Articles by alphabetic order
 Ā Ī Ñ Ś Ū Ö Ō
1 2 3 4 5 6 7 8 9 0

Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele

From Tibetan Buddhist Encyclopedia
Jump to navigation Jump to search

Jayarava Attwood

Independent Scholar,

Cambridge, United Kingdom


A transcription of the Fangshan Stele of the Heart Sūtra is presented in an English language Buddhism Studies context for the first time. While the text of this Heart Sūtra is relatively unremarkable, the colophon reveals that work on the stele commenced in 661 CE. This is not only the earliest dated reference

to the Heart Sūtra in any language, but the date falls during Xuanzang’s 玄奘 lifetime (ca. 602–664). The status of the Heart Sūtra as an authentic Indian sūtra rests almost entirely on the supposed historical relationship with Xuanzang since it fails to meet the standard criteria for being a sūtra. The

historical connection between Xuanzang and the Heart Sūtra is critically re-evaluated in the light of the Fangshan Stele and recent scholarship from the fields of history, philology, and bibliography.

Keywords: Heart Sūtra, Xinjing, Prajñāpāramitā, Xuanzang, Fangshan Stele

 I thank Ji Yun for drawing my attention to the existence of the Fangshan Stele in June 2018. In writing this article I benefitted greatly from my correspondence with Jeffery Kotyk, especially in the area of Chinese historiography. He also made many useful comments on the first draft generally. Another friend, who wishes to remain anonymous, helped me to decipher the colophon and spotted many typos in Chinese. 2 Journal of Chinese Buddhist Studies Volume 32 (2019)

從房山石碑看玄奘與《心經》的關係 Jayarava Attwood 英國劍橋獨立研究學者


房山石碑《心經》的抄本,首次呈現在英語佛教研究的語境之中。雖 然經文的內容沒有特別引人注目的地方,但是最後的碑記揭示造刻於公元 661 年。這不僅是所有語言中最早涉及《心經》的日期,而且是在玄奘的 有生之年。《心經》作為一部真正的印度佛經,其地位完全取決於與玄奘 的歷史關係,因為它並不符合佛經的標準規範。本文根據房山石碑與近年 來歷史學、文獻學與目錄學的研究成果,對玄奘與《心經》的歷史關聯進 行批判性的重新評估。 關鍵詞: 心經、般若波羅蜜多、玄奘、房山石碑


The Heart Sūtra is, for the most part, synonymous with the Chinese Xinjing 心 經, or to give the full canonical title, the Boreboluomiduo1 xinjing 般若波羅 蜜多心經, i.e. sūtra No. 251 in the Taishō shinshū daizōkyō 大正新修大藏經.2 The title has varied in different contexts but has been stable as the Xinjing

for over 1,000 years.3 This is the text that has been chanted by pious Buddhists, in their local pronunciation, throughout Asia for at least thirteen centuries. This is the text on which learned East Asian scholars composed commentaries. Although other versions are in use or are preserved in Tibet and

Nepal, these can ultimately be traced back to the Xinjing.4 The canonical Xinjing has an annotation attributing the translation to the Tang dynasty pilgrim

and scholar-monk, Xuanzang 玄奘 (ca. 602–664). This association is long-standing and taken at face value in Buddhist circles and amongst some Buddhist Studies scholars. However, the traditional attribution

1 般若 (Sanskrit prajñā; Pāli paññā) is variously transcribed as bore, banre, and banruo. There is some discussion of this in the Digital Dictionary of Buddhism entry for 般若, which acknowledges the widespread use of bore in Buddhist contexts, perhaps as a kind of shibboleth. The DDB editors recommend

banre. Meanwhile, some prominent Prajñāpāramitā scholars, such as Zacchetti and Huifeng, use banruo. 2 The Taishō edition lists a number of variant readings from earlier editions of the Tripiṭaka. Other minor variations can be found in the Chinese text of T 256, in the texts embedded in early

commentaries by Kuījī and Woncheuk, and in the text of various inscriptions including the Fangshan Stele. Other variations are apparent in Heart Sūtra texts preserved at Dunhuang. For some preliminary remarks on the Dunhuang texts see Nourse, “The Heart Sūtra at Dunhuang” (I’m grateful to the author for

supplying a copy of his presentation). To date, there is no English language study of the editions of the Heart Sūtra 3 Other common Chinese titles for the text include Duo xin jing 多心經, Bore xin jing 般若心經, Boreboluomiduo damingzhou jing 般若波羅蜜多大明呪經, (Fukui cited in Ji, “Is the Heart Sūtra an

Apocryphal Text,” 37–8). A similar situation holds for the Synoptic Suvarṇaprabhāsottama-sūtra (T 664) ascribed to Paramārtha: “there is therefore no single ‘correct’ title for it” (Radich, “On the Sources,” 209 n.8). Also compare Stefano Zacchetti’s notes on the title of Dharmarakṣa’s translation of the

Larger Prajñāpāramitā Sūtra (Zacchetti, In Praise of Light, 3 n.5). 4 Lopez, The Heart Sūtra Explained, 6–8, points out that Indian commentaries preserved in Tibetan and Tibetan commentaries are all on the extended version of the sūtra and only the extended version is found in the Kanjur. The extended text is represented in Chinese by T 252, 253, 254, 255, and 257.

is considered apocryphal by others. Jan Nattier’s landmark article on the Heart Sūtra summed up the reasons for doubting the tradition.5 The text of Xinjing appears to be an edited version of passages copied from Kumārajīva’s translation of the Large Sūtra. The Xinjing does not draw on Xuanzang’s

various Large Sūtra translations included in his Da boreboluomiduo jing 大般 若波羅蜜多經 (T 220). Had the Xinjing been a translation by Xuanzang, we would have expected it to be included in T 220, but it was not. The Neidian Catalogue, completed in 664 CE, has several entries for the Heart Sūtra: one

attributes the translation to Xuanzang6 whereas another classifies it under the heading “Mahāyāna sūtras with unknown translators.”7 Tradition suggests that the Damingzhoujing (T 250),8 a Heart Sūtra text attributed to Kumārajīva, was superseded by the Xinjing (T 251). However, Chinese Buddhists generally

found Xuanzang’s translations pedantic and verbose and preferred translations by Kumārajīva. Furthermore, Nattier refers to the attribution of Damingzhoujing to Kumārajīva as “highly suspect” and concludes: “What we can state with certainty at this point is that [the Damingzhou jing] is neither

Kumārajīva’s nor an independent translation from Sanskrit.9 The first dated evidence of the Damingzhou jing comes in the Kaiyuan Catalogue compiled in 730 CE.10 There is a crucial piece of evidence, long known about in China but absent from these discussions in English, i.e. that the Heart Sūtra is inscribed on the Fangshan Stele.11 The inscription is dated March 13, 661, almost exactly three

5 Nattier, “The Heart Sūtra,” 189–93. 6 T 2149, 55: 282c6. 7 T 2149, 55: 294a6. The other two entries are under the headings: Dacheng jing yiyi 大乘經一譯 [[[Mahāyāna Sūtras]] with one translation], T 2149, 55: 305a16; and Dacheng jing zhengben 大乘經正本 [[[Mahāyāna Sūtras]] that are original copies], T 2149, 55:

320a17. 8 Boreboluomiduo damingzhoujing 般若波羅蜜多大明呪經 (*Mahāprajñāpāramitā-mahāvidyā-sūtra). 9 Nattier, “The Heart Sūtra,” 189. 10 Zhisheng 智昇, Da Tang Kaiyuan shijiao lu 大唐開元釋教錄 (Catalogue of Śākyamuṇi’s Teachings of the Kaiyuan Era of the Great Tang), T 2154, 55. 11 One of the anonymous

reviewers pointed out that Fukui discusses the Fangshan Stele in Japanese, though this discussion has not made it into the English language literature. For example, Kazuaki Tanahashi, who relied on Japanese scholars (including Fukui) for his “comprehensive guide” to the Heart Sūtra, discusses the Beilin Stele, dated 672 CE, as the oldest dated Heart Sūtra (Tanahashi, The Heart Sutra, 81, 95–7).

years before the Biography records the death of Xuanzang on March 7, 664,12 and two years before he completed his Prajñāpāramitā Sūtra translations in late 663. The Fangshan Stele has been discussed in a number of Chinese language publications following the rediscovery of the cache of stone tablets at Yunju si

雲居寺 (Cloud Dwelling Temple), on Fangshan 房山 (Repository Mountain), in the early twentieth century. Two mentions of the Stele can be found in English language articles written for art history journals.13 In 2016, a wave of news reports about the Fangshan Stele swept through the Chinese media, including at least one outlet in English, but this report was not picked up by English language media. Therefore, I wish to report on this important artefact to an

English-speaking Buddhist Studies audience, apparently for the first time, and to consider the implications of it for the history of the Xinjing. Below I provide a complete transcription of the Stele based on my own examination of published images of the rubbing taken in the 1930s, particularly plate 9 in Fangshan Yunjusi shi jing 房山雲居寺石經 (Stone Sūtras of Yunju Temple, Mount Fang) and the first image in He and Xu’s “The Early Recessions of the Heart Sūtra.”14

I have also consulted two published transcriptions of the colophon.15 The text of the sūtra itself is relatively unremarkable with a few variant characters. After commenting on the colophon, I will reflect on how this affects our understanding of the history of the Heart Sūtra and its connection with Xuanzang.

The Fangshan Stele Text

The effort to preserve sūtras in stone at Yunju Temple was initiated during the turbulent Sui dynasty by a Buddhist monk named Jingwan 靜琬, who was convinced that he was living in the period of the decline of the Dharma, i.e. mofa 末法. We don't know the exact year Jingwan began his project to record

12 麟德元年二月五日 (Fifth day, second month of the first year of Linde). The date information is scattered through various pages of T 2053. Year from 50: 276c2, month from 277a25, and day from 277b4. This is sometimes reported as February 5, 664. 13 Ledderose, “Changing the Audience,” 395; Lee,

“Transmitting Buddhism,” 55. 14 Chinese Buddhist Association, Stone Sūtras of Yunju Temple, Mt Fang, Beijing: Cultural Relics Publishing House, 1978; He and Xu, “The Early Recessions of the Heart Sūtra,” 13. A copy was kindly supplied to me by Ji Yun. 15 Lin, “A General Survey”; Zhang, Collection of Tripiṭaka Research; Beijing Library, et al., Classified Compilation.

Buddhist texts in stone but his contemporary, Tang Lin 唐臨, says that it was in the Daye 大業 period, 605–616 CE.16 In the main part of the project, Buddhist texts of various types were engraved on about 5,000 stone tablets and stored in artificial caves at nearby Mount Fang. It was very much a case of

the medium is the message. The Fangshan Stele is one of about 10,000 stone tablets bearing Buddhist inscriptions that were buried in a courtyard of Yunju Temple between 1117 and ca. 1200.17 The monks who carried on the project after Jingwan’s death in 639 apparently began to accept commissions. The Fangshan Stele is not only the earliest dated Heart Sūtra but the earliest of the Yunju tablets engraved at the behest of donors.18 The dimensions of the stone tablet are 85 x 56 cm. The inscription presents the Xinjing and a colophon recording the donor and the date work commenced in a 15 x 26 grid of 15 columns

with a maximum of 26 evenly-spaced characters (reproduced below). For ease of reference, I have numbered the columns right to left and used letters of the alphabet to label the rows of characters. The Xinjing comprises columns 1–12 and the colophon columns 13–15. There is a half-column width space between the

text and colophon that I leave unlabelled. The bottom left corner of the tablet is missing, meaning that we have lost four characters from column 15, three from columns 13–14, and one from column 11. Additionally, the tablet has been broken in half, leaving a ragged line across the middle of the rubbing

(through rows m and n) that partially obscures some characters. The surface of the tablet has been damaged meaning that the rubbing is unclear in many places. I have filled in the missing characters of the sūtra from the canonical text. Apart from a few minor variations noted below, this text is not

significantly different from T 251. There are some spaces in the inscription, which take up a full measure. In the transcription, the symbol “□” indicates a character-sized space, “?” represents an unreadable or missing character. My full transcription of the stele follows. Missing or entirely unreadable parts of the inscription are greyed out.

16 His account is found in the Ming bao ji 冥報記, translated in Gjertson, Miraculous Retribution, 165. 17 Ledderose, “Changing the Audience,” 386. 18 Lee, “Transmitting Buddhism,” 55.


家 母 雍 提 羅 大 蜜 罣 a 眷 段 州 莎 蜜 多 礙 亦 b □ 櫟 婆 多 故 故 在 波 c 不 菩 羅 d 此 扈 諸 異 蜜 e 說 上 耨 恐 多 f 徳 懷 騎 g 齊 慶 將 波 曰 羅 遠 𦀰 h 軍 羅 □ 不 異 □ i 正 嗣 守 蜜 揭 等 藐 顛 菩 盡 波 三 j 左 多 諦 等 三 倒 提 乃 k ? 器 衛 菩 夢 l □ 涿 𦀰 諦 m 顯 貞 城 一 □ 除 故 究 依 n 慶 女 府 卷 一 知 竟 照 奘 o 六 大 左 羅 奉 p 娘 果 揭 槃 波 五 □ q 二 二 毅 諦 真 波 三 羅 詔 r 月 娘 都 □ 皆 譯 s 隸 尉 波 不 蜜 諸 多 盡 t 禓 羅 虛 多 u 造 巫 一 v ? 揭 說 大 集 乃 w ? ? □ 諦 罣 滅 至 x ? ? 父 □ 波 礙 如 厄 y ? ? ? 菩 波 舍 z

Column 1 contains the title of the text: 般若波羅蜜多心𦀰, Prajñāpāramitā Heart Sūtra, where jing 𦀰 is an archaic variant form of jing 經.19 This is followed by a space and the attribution: 三藏法師玄奘奉□詔譯 “Tripiṭaka Dharma-master Xuanzang translated with imperial authorisation.” (1.j–s). The phrase 奉□ 詔譯 is a mark of imperial authority. The space before the

character zhao 詔 is a sign of respect for the Emperor, although there was also a taboo against writing the name of the reigning emperor. The Digital Dictionary of Buddhism defines this character:

詔 “An imperial edict. To decree. Appearing in the colophons of translated scriptures, it indicates official authorization at the highest level, indicating the high level of the translator’s reputation.”20

The text of the Heart Sūtra follows (columns 2–12) and is more or less the standard canonical text, with some minor variations. The character wu is inscribed using the ancient variant form, . This substitution is common in ancient inscriptions “from at least the fourth century BCE.”21 In the dhāraṇī,

di 帝 is written as di 諦, which we also see in the Beilin Stele. It may be that the scribe saw the latter character as more significant since it is used in Chinese translations of the “two truths,” i.e. er di 二諦, though of course in the dhāraṇī it is used for its phonetic rather than semantic value.

Alternatively, the character 帝 means “Emperor” and it might have been politesse to choose a variant with the same pronunciation and a different connotation. The final characters of the dhāraṇī (12.b–d)—sa po he 婆訶—are evidently a transcription of Sanskrit svāhā, where sa and po are intended to

convey the conjunct svā. The Taishō edition and the CBETA version instead give seng sha he 僧莎訶, which seems to be a poor representation of svāhā. Taishō notes that Song, Yuan and Ming editions of the Tripiṭaka all had 薩婆 訶.22 The stele suggests that the modern edition is incorrect.23 Finally, following the sūtra (12.n), there is what appears to be a space followed by the single character juan 卷 (fascicle). I think we can safely

20詔. 21 Galambos, “Simplified Characters,” 191. 22 T 251, 8: 848c23 and note 7. 23 Kuiji's commentary, Boreboluomiduo xinjing you zan 般若波羅蜜多心經幽贊, has sha he 莎訶 for svāhā (T 1710, 33: 542c8) while Woncheuk’s commentary, Boreboluomiduo

xinjing zan 般若波羅蜜多心經贊, has sha po he 莎婆呵 (T 1711, 33: 551c10). Another early Tang dynasty commentary by Facang 法藏 (702 CE), Boreboluomiduo xinjing lüe shu 般若波羅蜜多心經略疏, has sa po he 薩婆 訶 (T 1712, 33: 555a6). This suggests that seng sha he 僧莎訶 may be the result of an eye-skip, copying from earlier in the line. I think these variants lend credence to John McRae’s suggestion that the Dhāraṇīsamuccaya translation by Atikūṭa (T 901) might be the source of the dhāraṇī in the Heart Sūtra (“Ch’an Commentaries,” 107 n.10) but this needs more research.

Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele 9 conjecture that 卷 is preceded by yi 一 (one), though it is obscured by the crack where the tablet broke in half. We can now turn to the colophon.

The Colophon

Most of the interest generated by the Fangshan Stele has focussed on the colophon since the date occurs there. It is the text of the colophon rather than the sūtra that is reproduced in the literature. I will cite two published transcriptions below. In some cases, the Beijing Library Group used modern simplified characters and in order to facilitate comparisons, I have converted them to the forms used on the stele itself. I will present the inscriptions as they occur and then add a commentary. Lin’s transcription skips over some details and has added punctuation:

雍州櫟陽縣遊騎將軍守左衛涿城府左果毅都尉楊社生、 母段、妻扈、息懷慶、玄嗣、玄黎、玄…… 眷屬、緣此功德、齊成正覺。顯慶六年二月八日造。24

In 1987, the Beijing Library Metal and Stone Group and The Chinese Buddhist Books and Cultural Relics Museum Stone Sūtra Group (henceforth abbreviated to Beijing Library Group) published a more complete transcription:

雍州櫟陽縣游騎將軍守左衛淥城(府)左果毅都尉楊社生 母段□妻扈息懷慶玄嗣玄器玄貞女大良二娘秣利巫山 家眷屬緣此功德齊成正覺□顯慶六年二月八日造[[[經]]]25

I used these as a starting point and compared them with the images of the stele. I found it necessary to make some minor corrections and have inferred another two positions (13.x–y) in the missing section. My corrected and extended version of the colophon reads:

24 Lin, “A General Survey.” 25 The character in parenthesis in the first line (府) was omitted and added by the present author. The rubbing clearly has 娘 for 良 in the second line. The character in square brackets in the 3rd line [[[經]]] was added by the Beijing Library Group and is not visible in the rubbing. Beijing Library, et al., Classified Compilation, 199.

10 Journal of Chinese Buddhist Studies Volume 32 (2019) 13. 雍州櫟陽縣遊騎將軍守左衛涿城府左果毅都尉楊社生□父? 14. 母段□妻扈息懷慶玄嗣玄器玄貞女大娘二娘隸利巫山??? 15. 家眷屬緣此功德齊成正覺?□顯慶六年二月八日造????

The first line of the colophon mainly describes the donor. The inscription was commissioned by Yang Shesheng 楊社生 (13.u–w). The name Yang 楊 is very significant in Chinese history of this period because it was the family name of the Sui dynasty emperors (581–618) and of the mother of Wu Zetian 武則 (624–705). It is not clear whether or how Yang Shesheng was related to these important figures. Yang’s ancestral home was Yongzhou 雍州 (Yong Province), where the Tang capital Chang’an (長安) was located. More specifically, his family was from Yueyang County 櫟陽縣 (13.c–e), about 50 km northeast of

Chang’an. The inscription tells us that Yang was an officer in the military.26 The early Tang military was based on the fubing 府兵 system of self-supporting garrisons,27 each of which had an overall commander or “general” who was assisted by two “courageous commanders,” left and right. Yang held the

“prestige title” or titular rank of Youji jiangjun 游騎將軍 (General of Mobile Cavalry) (13.f–i).28 Yang served (shou 守 13.j) in the garrison (wei 衛 13.l) of a place called Zhuochengfu 涿城府 (13.m–o),29 which seems to correspond to modern-day Zhuozhou 涿州 located about 17

26 “By Tang times, regimental offices were no longer necessarily identified with local power. However, there were still good reasons for local elites to seek command positions in the fubing system. They offered the prestige of government office, the possibility of upward mobility through promotion to, say,

a generalship in the Guards, and distinct advantages for wealthy, landholding families under the equalfield system of land distribution (since officers were entitled to up to 600 mu of “office land” [zhifen tian 分 田 ] and might be able to claim additional landholdings on the basis of honorific rank

[xun guan 勳官] won in battle).” Graff, “The Reach of the Military,” 262. 27 The fubing system conscripted men mainly from wealthy families. From ages twenty to sixty they served as required and the rest of the time tended land allotted to them. They were concentrated around Chang’an and in the northwest.

Expeditionary forces were swelled by irregular conscripts as required. Graff, “The Reach of the Military,” 245–6. 28 Hucker, Dictionary, 584 (s.v. yu-chi chiang-chün). 29 The Beijing Library Group have Lu cheng fu 淥城府 199, but I can find no reference to this place name. The name 涿城府 does occur at CBETA B 146, 26: 110a13.

Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele 11 km southeast of Mount Fang.30 Yang’s actual role in the military was reflected in the rank of Guoyi duwei 果毅都尉 (Courageous Commander) (13.q–t) of the

Left (zuo 左) (13.p).31 Line two of the colophon (column 14) begins with Duan 段, the name of Yang’s mother (mu 母), (14.1–2), followed by a space. Duan would probably be her family name. Next is his wife (qi 妻), Hu 扈 (also a family name); followed by his sons (xi ): Huaiqing 懷慶, Xuansi 玄嗣, Xuanqi 玄器32, and Xuanzhen 玄貞; and his daughters (nu 女) Da’niang 大娘33 and Erniang 二娘 (First daughter, Second daughter). Finally, someone named Li Wushan 利巫

who is probably a servant or dependent (li 隸) is included. The character can also indicate a slave, but given that he is treated as part of the family we could think of him as a “ward” or “retainer.” The person missing from this list is Yang’s father and it is unthinkable that he would be left out. Since there are potentially three characters missing from column 13 (xyz), it seems likely that they included the word “father” (fu 父) and his name. Since

Yang’s mother’s name was followed by a space (14.c), we can conjecture that the missing characters in column 13 were a space □, the character 父 “father” and the father’s name in one character. The father’s family name was obviously also Yang 楊, so perhaps one of his other names would have appeared here. The third line (column 15) asks that family (jia 家) members (juan shu 眷 ) be caused (yuan ) by this merit (ci gong de 此功德) to attain awakening

(cheng zheng jue 成正覺) together (qi 齊). The character at 15.l is unreadable and none of the other commentators has hazarded a guess. This is followed by a space and then the date: 顯慶六年二月八日. The nian hao 年號 (reign title) Xianqing 顯慶 refers to a period of the rule of Emperor 唐高宗 Tang Gaozong

(649–683 CE). The 6th year of Xianqing, 2nd month, 8th day corresponds to the date 13th March 661 CE.34 The date is interesting because Xuanzang is recorded to have died almost exactly three years later on 7th March 664.35

30 “又称涿州为“涿城府””. Zhuozhou Museum, 31 Hucker, Dictionary, 298 (s.v. kuo-i fu; cf 545 s.v. tu-wei). 32 Lin has 玄黎, but the rubbing is fairly clear at this point. 33 We might have expected 太 for 大 in this case. 34 Date conversion by 35 Da Tang Da ci'en si sanzang fashi zhuan 大唐大慈恩寺三藏法師傳, T 2053, 50: 275c.

The final visible character (15.v) is zao 造: “make, fashion; construct; begin, commence.”36 The Beijing Library Group added the character jing in their transcription. In the photographs of the rubbing, we can see what may be part of a stroke. However, the mark we see is not consistent with the character jing , especially given the variant character 𦀰 used elsewhere in the text. A comparison with some of the other plates in the Fangshan Yunjusi wi jing shows that some colophons simply end with 造.37


The Fangshan Stele unequivocally treats the Xinjing as a translation by Xuanzang, three years before his death in early 664. This is consistent with the traditional history of the text. On the other hand, we have a compelling body of evidence that the Xinjing was a not a translation at all, but rather a

digest text composed in Chinese, drawing on Kumārajīva’s Large Perfection of Wisdom Sūtra translation and superficially edited to make it look like one of Xuanzang’s translations by the addition of a few terms he introduced. How do we reconcile these two conflicting views? One approach would be to assume the

truth of each proposition and see which produces the best explanation of the known facts. However, part of assessing an explanation is to examine the foundations on which it is based and in this case, some of the foundations are quite shaky.

Traditional Historiography

Apart from the Fangshan Stele, the first dated attribution of the Heart Sūtra to Xuanzang occurs in the Neidian Catalogue,38 a bibliography of translated Buddhist texts compiled in the year of Xuanzang’s death (664 CE) by Daoxuan 道宣 (596–667 CE). The close association between Xuanzang and the Xinjing was bolstered by a traditional story, told in the Biography,39 a hagiography of

36造. 37 Chinese Buddhist Association, Stone Sutras, plates 48, 56, and 58. 38 Datang neidian lu 大唐內典錄 (Catalogue of the Inner Canon of the Great Tang), T 2149, 55. 39 Da Tang Da ci'en si sanzang fashi zhuan 大唐大慈恩寺三藏法師傳 (Biography of the Tripiṭaka Master of the Great Ci’en Monastery of the Great Tang Dynasty), T 2053, 50.

Xuanzang attributed to Huili 慧立 and Yancong 彥悰, with a colophon by the latter dated 688 CE.40

Formerly, when the Master was in the region of Shu, he once saw a sick man suffering from a foul skin ulcer and dressed in rags. With a feeling of pity, he took the man to his monastery and gave him money to purchase clothes and food. Being ashamed of himself, the sick man taught the Master this sutra, which he [the master] often recited.41

The ancient region of Shu 蜀, in the vicinity of modern-day Sichuan, was a remote province, far from the centers of power. The Biography recounts that Xuanzang and his brother fled to Shu to escape the chaos that ensued from the collapse of the Sui dynasty. The Biography is routinely treated as a reliable

historical source by historians of Buddhism, but this is problematic. Max Deeg has helpfully discussed the parallel problem of naïve use of Xuanzang’s Record 42 by historians.43 Deeg has shown that the Record is often acting as propaganda for Buddhism rather than as an accurate historical account of

Xuanzang’s travels. Part of Xuanzang’s aim in composing the Record seems to have been to admonish the Emperor Taizong 太宗 (r. 626–649) to be a good ruler

by Buddhist standards. According to its preface, the Biography was composed by Huili, a disciple of Xuanzang’s, and completed, some years later, by

Yancong, a Buddhist monk, in 688 CE. This was a time when Buddhists were taking Wu Zetian’s side in the ongoing internecine conflict within the Tang court that saw her take the imperial throne in 690.44 In the Biography, emperors Taizong and

40 The preface of the Biography suggests half of it was composed by Huili at an earlier (unspecified) date and the other half was added by Yancong at a later date, but it is not clear who wrote which parts. Li, Biography, 5–9. 41 Li, Biography, 26. Translating T 2053, 50: 224b8–10. 42 Da Tang Xiyuji 大唐

西域記 (Great Tang Record of the Western Regions), T 2087, 51. 43 Deeg has addressed this issue in at least three papers: “Has Xuanzang really been in Mathura?,” “Show Me the Land Where the Buddha Dwelled,” and “The Political Position of Xuanzang.” 44 The historiography of Wu Zetian is complex since the

imperial sources, such as Liu’s Jiu Tang shu (The Old History of the Tang) and Ouyang’s Xin Tang shu (New History of the Tang), paint a biased picture and her story is still in the process of being revised. Modern revisionist accounts of Wu include Guisso, Wu Tse-t’ien, and Rothschild, Wu Zhao. Accounts such

as Eisenberg, “Emperor Gaozong,” also to some extent rehabilitate Gaozong, giving him a greater role in promoting Wu and actively sharing power with her.

Gaozong 高宗 (r. 649–683) are portrayed as pious Buddhists with an improbably high degree of interest in the minutia of Xuanzang’s life and work. However, non-Buddhist sources suggest that both were renowned for their disinterest in religion.45 Much of the Biography describes Xuanzang in superlative or

miraculous ways consistent with what Joseph Bulbulia has called “charismatic signalling.” The primary purpose of charismatic signalling is to provide a way to “align prosocial motivations” in large religious movements: “Charismatic culture supports cooperative outcomes by aligning powerful emotions,

motivations, and intentions among potentially anonymous partners, toward collective goals.”46 The Biography appeared in 688 CE, just two years before Wu Zetian took the throne. It portrays the early Tang emperors as favourable towards Buddhism and thus could have provided an important reference point for

Chinese Buddhists as Wu Zetian rallied support for her move from regent to sovereign. The sick man story is inserted into a fairly standard Buddhist miracle tale. As outlined by Robert Campany, these involve “a compassionate, salvific, and clear intervention in human affairs by some powerful being,

typically the bodhisattva or buddha on whom the sūtra focuses.”47 Inspired by the Lotus Sūtra, or more specifically, by chapter 25 of Kumārajīva’s translation (T 262) which also circulated separately as the Guanshiyin jing 觀世音經, many miracle stories involve a protagonist in jeopardy who becomes

absorbed in the act of invoking Guanyin,48 who then intervenes to save them from misfortune. However, in the sick man story, reciting the name Guanyin and

addressing prayers to him does not work,49 which allows the storyteller to introduce the Heart Sūtra—but only once. Immediately after this, Xuanzang is once again in peril and again invokes Guanyin, who does save him this time. Clearly, the

45 Taizong’s attitude to Buddhism is detailed in Weinstein, “Imperial Patronage,” 265–306. He seems to become increasingly hostile to Buddhism after taking the throne from his father. However, Taizong and Gaozong both saw the political expediency of imperial support for Buddhism at a time when many of

the aristocracy had converted; compare in particular, Weinstein, “Imperial Patronage,” 265–7. 46 Bulbulia, “Charismatic Signalling,” 545. 47 Campany, “Notes,” 30–1. 48 Campany, “Notes,” 32. In the Biography the expression is “he concentrated ceaselessly” 心心無輟 (T 2053, 50: 224b27). This is not one of

the standard phrases listed by Campany but has the same illocutionary force. 49 Li, Biography, 27. Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele 15 episode of chanting the Heart Sūtra is not fully integrated into the Biography: was this because of having two authors, or was a third party involved? The story of the sick man in the Biography is the only narrative flashback in an otherwise relentlessly linear chronological narrative. The event is near the

beginning of his journey to India. In order to explain how he came to know the Xinjing, the narrative returns to his time in Shu. Whatever the probative value of the explanation, we can infer from this that the authors felt an explanation was required. Campany says, “Authors and collectors of such stories

about the efficacy of reciting the Guanshiyin Sūtra apparently fashioned them quite selfconsciously to authenticate the sūtra’s claims for itself”.50 It seems that the sick man story is also a self-conscious attempt to authenticate the Heart Sūtra, but one that is crudely superimposed on the Biography. The

Heart Sūtra does not feature in the many other perilous situations that Xuanzang faced on his sixteen-year odyssey. It is mentioned one other time in the Biography which I describe below. In the Record, Xuanzang does not mention the Heart Sūtra at all. The sick man story is taken to mean that the Heart Sūtra

existed before Xuanzang left on his pilgrimage to India, ca. 629 CE. This fits with the idea that the Damingzhoujing (T 250) is a translation of the Heart Sūtra by Kumārajīva, completed in the early fifth century, although as we’ve seen there are many reasons to doubt this account. Apart from the Damingzhoujing, some other early lost translations are sometimes postulated.

Lost Translations

Some modern scholars have attempted to connect the Xinjing and Damingzhoujing to records of short Prajñāpāramitā texts found in early medieval bibliographies, in particular, the Collection of Records compiled by Sengyou 僧祐 (445–518 CE) in 515 CE.51 The Collection of Records purports to reproduce the entries of an earlier bibliography by Dao’an 道安 (312–385), compiled in 374 CE but now lost. That Sengyou’s citation of Dao’an was reliable is, again, simply stipulated by Buddhist historians. According to Sengyou, Dao’an listed two short Prajñāpāramitā texts:

50 Campany, “Notes,” 33. 51 Chusanzang jiji 出三藏記集 (Collection of Records about the Production of the Tripiṭaka), T 2145, 55.

摩訶般若波羅蜜神呪一卷 = Moheborebeluomi shenzhou in one scroll.

般若波羅蜜神呪一卷(異本) = Borebeluomi shenzhou in one scroll (different version).52

However, these texts are not listed as sūtras but are instead shenzhou 神呪 which literally means “magical incantation.”53 The two texts are not attributed to any translator, which, for bibliographers like Sengyou, undermined their claim to authenticity. The idea that shenzhou is necessarily a translation of

some Sanskrit term is moot; even if it were not, the idea that shenzhou is a translation of hṛdaya (heart) is far-fetched. The translation of hṛdaya as xin (heart) is all too obvious since they have more or less the same denotation and connotations. Against this, we have Japanese scholar Fukui Fumimasa’s

1987 argument that can be interpreted as dhāraṇī in this context.54 The idea is supported by the fact that most of the late Nepalese manuscripts refer to the text as a dhāraṇī also. By contrast, in Chinese, the Xinjing always calls itself a sūtra (), although it was (and is) certainly used like a

dhāraṇī as described by Paul Copp.55 While shenzhou is a plausible translation of dhāraṇī, I know of no other dhāraṇī that also fits the Chinese bibliographic category of digest text so perfectly as the Xinjing does. If these shenzhou texts did indeed exist in 374 CE then they predate both

Kumārajīva and Xuanzang, which simplifies our problem. All the extant versions of the Heart Sūtra reuse passages from Kumārajīva’s 404 CE translation of the Large Perfection of Wisdom Sūtra (T 223). Thus the shenzhou texts are not versions of the Heart Sūtra. Another “lost translation,” Boreboluomiduona

jing 般若波羅蜜多那經, is mentioned for the first time in the Kaiyuan Catalogue of 730 CE, this time attributed to Bodhiruci (d. 727), an Indian monk who translated texts during the reign of Wu Zetian. He is said to have collaborated with her in interpolating

52 T 2145, 55: 31b9–10. 53 Some of the Sanskrit Prajñāpāramitā texts also lack the appellation sūtra, for example the Prajñāpāramitā-vajracchedikā. 54 Fukui, Hannnya shingyō no rekishiteki kenkyū, cited, with approval, by Nattier, “The Heart Sūtra,” 175. Ji Yun, “Is the Heart Sūtra an Apocryphal Text,” 37–8, gives more detail about Fukui’s reasoning, but Fukui’s work is only available in Japanese and I have been unable to consult it directly. 55 Copp,

The Body Incantatory,

prophecies of a female emperor into Buddhist texts. Ji Yun discusses the idea of a Bodhiruchi translation and concludes that it is doubtful at best.56 Translated in 649

The received tradition also asserts that Xuanzang translated the text into Chinese only after his return from India. Specifically, the Kaiyuan Catalogue57 compiled in 730 CE records that he made the translation in 649 CE.58 Since there is no record of his encountering the Heart Sūtra anywhere else, or that he

returned from India with a Sanskrit manuscript of the Heart Sūtra, this suggests that he received a Sanskrit text in Shu. The phrasing of the sick man story in the Biography suggests oral transmission of the Xinjing, i.e. that the sick man taught (shou 授) the sūtra to Xuanzang, who subsequently recited it (song 誦). While some Sanskrit texts did circulate amongst the Chinese Buddhist literati, very few people at any given time had the opportunity to study

Sanskrit and they would all have been Buddhism monks living in or around the translation centres in the imperial capitals. The idea that a Sanskrit text was in oral circulation in far-flung Shu is far-fetched at best and it raises a whole raft of questions about the provenance of such a text that cannot be

answered. If we accept the Kaiyuan date for the translation of the Xinjing then, according to the Biography, Xuanzang was staying at the Cuiwei Palace at

the request of Emperor Taizong at the time.59 The Biography depicts Taizong converting to Buddhism during the 4th month of that solar year, and regretting that he met Xuanzang so late in life.60 Taizong died on the 27th day of the 5th month, but it seems quite unlikely that he ever converted to Buddhism.61 The discrepancies between Buddhist and non-Buddhist historical accounts deserve

56 Ji, “Is the Heart Sūtra an Apocryphal Text,” 47–8. 57 Zhisheng, Da Tang Kaiyuan shijiao lu, T 2154, 55. 58 The 649 CE date is discussed in Nattier, “The Heart Sūtra,” 174, 206 n.42, and in McRae, “Ch’an Commentaries,” 105 n. 2. In the Kaiyuan Catalogue it is given in a note at T 2154, 55: 555c3–4 in

the form 貞觀二十三年五月二十四日. In the Gregorian calendar this corresponds to July 8, 649. Date conversion by 59 Li, Biography, 221. 60 Li, Biography, 221. 61 “Whether he made such a statement is doubtful; if he did it must have been a deathbed conversion, totally at

variance with his life-long hostility towards the Buddhist church and Buddhist doctrine” (Wechsler, “T’ai-tsung,” 219). Wright, “T’ang T’ai-tsung,” 239–263, paints a more nuanced picture of Taizong turning against Buddhism only after taking the throne and a series of misfortunes. much more detailed and careful scrutiny, especially as the Buddhist sources are often used uncritically.

A Gold-Lettered Text

The Biography does not record the translation of the Heart Sūtra, but at nearly the same time, not long before the death of Taizong, it does record that Xuanzang made a new translation of the Jingang bore jing 金剛般若經 (Vajracchedikā-prajñāpāramitā Sūtra) at the request of the Emperor.62 It may be that

later accounts kept this story but changed the name of the text that was translated. However, it is also quite unlikely that Taizong, of all people, asked Xuanzang to retranslate a Buddhist text because he was dissatisfied with the earlier efforts. Later in the Biography, Huili and Yancong reprinted a letter

from Xuanzang to Emperor Gaozong (dated 656) in which he gives the emperor a “Gold-lettered Prajñā Heart Sūtra” (Jin zi bore xinjing 金字般若心經) in one fascicle to congratulate him and the Empress on the birth of a son.63 The same letter occurs in a collection preserved in Japan and it appears that this

source might have been used by Yancong when editing the Biography.64 With reference to the name of the sūtra, I note some graphical similarity between 金剛般若經 and 金字般若心經, which might fit the idea that Xuanzang completed an early translation of the Vajracchedikā and the title of the text was changed to

fit the emerging narrative of the Heart Sūtra. The fact that the letter is preserved independently argues against this and suggests that Xuanzang himself might have had a copy of the Xinjing by 656 CE.

Assessing Traditional History

Even if, for the sake of argument, we set aside Jan Nattier’s observations about the Chinese origins of the Heart Sūtra (and the follow-up work by Huifeng and Attwood), this leaves traditional historiography heavily reliant on the Biography and it is suspect as a historical source. The Biography seems to be at odds with non-Buddhist sources and it is a problem for Buddhist historiography that texts like the Biography are still used naively as reliable historical sources and are not compared to non-Buddhist sources.

62 This episode occurs in Li, Biography, 215–6 (translating T 2053, 50: 259a13–a28). 63 T 2053, 50: 272b12. The Empress is Wu Zetian as Empress Consort. 64 T 2119, 52: 825a16–17. My thanks to Jeffrey Kotyk for drawing my attention to this text.

The Heart Sūtra episode in the Biography seems to have been crudely inserted into the text. Despite attempting to supersede the salvific power of Guanyin, whoever added it failed to notice that that Biography reverts to the standard trope immediately afterwards. Although they felt the need to explain the presence of the Heart Sūtra, they don’t explain why invoking Guanyin works sometimes and not at others. The Biography suggests that Xuanzang received the

Heart Sūtra orally before his trip to India. It is difficult to believe in the presence of an oral Sanskrit text in Shu at this time. It is not impossible, but medieval Chinese Buddhists showed a distinct preference for Chinese language texts. Xuanzang is portrayed as having a remarkable memory, but could he

have accurately memorised an oral Sanskrit text and twenty years later reproduced it accurately enough to make sense of it? The tradition is also dependent on two ideas that are first found in the Kaiyuan Catalogue: that the Damingzhou jing is an early translation by Kumārajīva and that the text attributed to

Xuanzang was translated in 649 CE. The Damingzhou jing is definitely not what it appears to be. The date of 649 CE is tied up with Buddhist attempts to align themselves with emperors via a putative friendship between Xuanzang and Taizong. Many elements of this story are implausible. It seems at least

possible that a story involving the Vajracchedikā was altered to fit the Heart Sūtra narrative. The Fangshan Stele adds little or nothing to this picture and has little explanatory power. The best we can say is that it appears to confirm the existence of the Xinjing during Xuanzang’s lifetime.

Modern Historiography

Modern scholarship of the Heart Sūtra really begins with Fukui Fumimasa in 1987 and his suggestion that the “sūtra” is, in fact, a dhāraṇī intended for ritual use65 although neither Kuiji nor Woncheuk saw it this way.66 Until this point scholarship occurred within a largely uncritical traditional framework.67 In 1992, Jan Nattier published her landmark article showing that the Heart Sūtra

65 Fukui, Hannnya shingyō no rekishiteki kenkyū, cited in Nattier, “The Heart Sūtra,” 175. 66 See Nattier, “The Heart Sūtra,” 206–7 n.33. 67 Probably because he believed that the Heart Sūtra defies “ordinary logic,” Edward Conze did not notice that he had committed a number of simple grammatical errors in his Sanskrit edition: see Attwood, “Heart Murmurs” and “A Note on Niṣṭhānirvāṇa in the Heart Sūtra.”

was composed in Chinese.68 Despite determined resistance from Japanese scholars,69 we cannot simply set aside Nattier’s argument or the supporting evidence published in the last few years.70 If the Heart Sūtra was composed in Sanskrit, then we would expect the core passage copied from the Large Sūtra to

resemble other extant Sanskrit Prajñāpāramitā texts and for the Chinese texts to be significantly different in the way that the core passage is different in the translations by Mokṣala, Kumārajīva, and Xuanzang.71 In fact, the Chinese Heart Sūtra is almost identical to Kumārajīva’s Large Sūtra, with only minor changes to make it more like a Xuanzang translation (but not Xuanzang’s actual Large Sūtra translations). The Sanskrit Heart Sūtra and Large Sūtra

texts could hardly be more different since the Heart Sūtra consistently chooses idioms that are not used in any Sanskrit Prajñāpāramitā text.72 The Xinjing is not an Indian sūtra but is a Chinese collection of copied passages, mainly from Kumārajīva’s Large Perfection of Wisdom Sūtra translation (T 223). This

is acknowledged in the early commentaries by Kuiji and Wonchuek.73 The core of the text copies the passage found at T 223, 8: 223a10–20, with a few word changes and a line excised (T 223, 8: 223a16– 17).74 The epithets passage comes from another chapter of the same source.75 McRae and Fukui noted that the dhāraṇī has counterparts in other texts76 and

68 Nattier was to some extent relying on Fukui, Hannnya shingyō no rekishiteki kenkyū. I thank the anonymous reviewer who drew my attention to subsequent works by Fukui, i.e., Yoroppa no Tōhōgaku to Hannya shingyō kenkyū no rekishi and Hannya shingyō no sōgōteki kenkyū. Unfortunately, I cannot read Japanese

and Fukui’s work is only dimly reflected in contemporary English language scholarship. 69 See for example Ishii, “Issues.” 70 Huifeng, “Apocryphal,” 72–105; Attwood, “Epithets” and “The Buddhas.” 71 T 221, 8: 6a6–13; T 223, 8: 223a13–24; and T 220, 7: 14a11–a26. 72 For example, the Heart Sūtra says rūpān

na pṛthak śunyatā; when in other Prajñāpāramitā texts this concept is always expressed in the form nānyā śūnyatā anyad rūpaṃ (Attwood, “Form,” 52–80). 73 Nattier, “The Heart Sūtra,” 206 note 33. 74 The Damingzhoujing starts earlier including the lines T 223, 8: 223a10–13, has identical wording, and includes

the missing line. 75 The passage is common, but the probable source is T 223, 8: 286b28–c7 (Attwood, “Epithets,” 42). An inscription from Mount Sili, Shandong Province, dated before 561, is the same passage sourced from Chapter 3 of the Xiaopin by Kumārajīva, i.e. T 227, 8: 543b25–c5 (Wang and Ledderose, Buddhist Stone Sutras in China, 421–5). 76 Cited in Nattier, “The Heart Sūtra,” 211, n.52 and n.53.

Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele 21 I think a case can be made that Atikūṭa’s translation of the Dhāraṇīsamuccaya77 is the actual source. The phrase du yiqie ku e 度一切苦厄 may well come from the Dafangguang shi lun jing 大方廣十輪經 (T 410).78 The translator of Dafangguang shi lun jing is not recorded, but the title is recorded in a bibliography

of Buddhist translations made during the Northern Liang Dynasty (北涼) ca. 397–439 CE.79 We can now also say that the Sanskrit text contains calques from Chinese such as tryadhvavyavasthitāḥ sarvabuddhāḥ and aprāptitvād.80 The inescapable conclusion is that the Heart Sūtra was composed in Chinese using passages copied from other Chinese texts, principally the Large Sūtra translation by Kumārajīva (or perhaps the commentary he translated concurrently). It

was then translated into grammatically correct but not idiomatically correct Sanskrit. And thus we have to take a new approach to the historiography of the Heart Sūtra. Digest Text It is now clear that Robert Buswell’s suggestion to Nattier that the Heart Sūtra was a 抄經 (chao jing) or “digest text” was correct.81 According to the early

77 Tuoluoni ji jing 陀羅尼集經 (Collection of Spells), T 901, 18. 78 T 410, 13: 708a26–7. There is no extant Sanskrit witness, but the title has been reconstructed as *Daśacakra-kṣitigarbha-sūtra. 79 The *Daśacakra-kṣitigarbha-sūtra was also translated by Xuanzang (T 411) and he translated this phrase

as tuo yiqie you ku 脫一切憂苦. 80 “When Buddhist Sanskrit texts refer to the buddhas of the three times, they always use the dvandva compound, i.e., atīta-anāgata-pratyutpannā buddhāḥ ‘past, future, and present buddhas’ or, rarely, atīta-anāgata-pratyutpannā sarvabuddhāḥ ‘all past, future, and present buddhas.’ In Chinese translations we find the equivalent of this in the form of 過去未來現在諸佛 (guoqu weilai xianzai zhu fu) ‘buddhas of past, future, and

present.’ but we also commonly find the expression used in the Heart Sūtra, i.e., 三世諸佛 (san shi zhu fo) ‘buddhas of the three times.’ The exact Sanskrit equivalents of 三世佛 and 三世諸佛 i.e. tryadhva-buddhāḥ, tryadhvā buddhāḥ and tryadhva-sarva-buddhāḥ or tryadhvāḥ sarva-buddhāḥ are never found either as a compound or as individual words in Prajñāpāramitā texts” (Attwood, “The Buddhas of the Three Times,” 14). See also the confusion of aprāptitvād

and anupalambhayogena caused by a translator misreading 以無所得 故 as the former when the latter was intended (Huifeng, “Apocryphal Treatment for Conze’s Heart Problems,” 72–105). 81 Buswell made this suggestion in a private communication to Jan Nattier in 1992 (“The Heart Sūtra,” 210 n.48). On chao jing generally see Tokuno, “Evaluation,” and Storch, The History of Chinese Buddhist Bibliography. Ji Yun also argues that

22 Journal of Chinese Buddhist Studies Volume 32 (2019) sixth century bibliographer, Sengyou 僧祐—who recorded some 450 such texts—“digests were produced by Chinese people who cut the existing translations into pieces and arranged them to their liking.”82 Kyoko Tokuno adds the important detail that the digest was “presumed to convey the text’s essential

meaning,”83 an attribute strongly associated with the Heart Sūtra. The practice of making digests was popular in China from the early encounter with Buddhism up to the early Tang. 84 Although we do see anthologies in India, such as the Śikṣasamuccaya and the Sūtrasamuccaya, they coexist with the texts they quote and are called samuccaya (“anthology”) or similar, so there is usually no confusion about what they are. Early medieval Chinese bibliographers,

by contrast, were concerned by potential confusion between chao jing, wei jing 偽經 (fake texts), and zheng jing 正經 (genuine texts). Despite their best attempts, a number of locally produced texts such as the Śuraṅgama Sūtra were accepted as being translations of Indian texts, down to modern times. The Heart Sūtra appears to be the only chao jing to slip through the net and this seems to be because of a deliberate effort to disguise its true origins by

attributing it to Xuanzang and by translating the chao jing into Sanskrit. For the early medieval bibliographers, to be considered a genuine Buddhist sūtra, a text had to meet four criteria: 1. Have a known connection with India, preferably to be a translation from a manuscript brought back by a named pilgrim; 2. Have a named translator, preferably someone with a good reputation; 3. Have been couched in Buddhist language (with no mixture of Daoism or

Confucianism); 4. Have the characteristic features of a sūtra: beginning evam maya śrutam; announcing the place it was preached; being spoken or endorsed by the Buddha; and being celebrated by the audience. The bibliographies themselves had quite sophisticated hierarchical categories of authenticity. Texts that met all the criteria were considered the most genuine; those that met only some criteria might still be considered genuine, but less so, on a sliding

scale. Those texts that met none of the criteria

the Heart Sūtra is a chao jing (“Is the Heart Sūtra an Apocryphal Text,” 41–5) although he later equivocates. 82 Translated and cited in Storch, The

History of Chinese Buddhist Bibliography, 64. 83 Tokuno, “Evaluation,” 39. 84 Storch, The History of Chinese Buddhist Bibliography. Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele 23 were considered fake. Digest texts tended to be categorised towards the fake end of the scale, especially as time went on.85 The Xinjing does not have any

of the characteristic features in the fourth criteria which ought to have disqualified it from being categorised as a sūtra, though other genres of genuine texts were available. These features were added in the extended version, but this happened later and the extended version was never important in China (e.g., there are no commentaries on it). The qualification of the Xinjing as a genuine sūtra is entirely on the grounds that the text was associated with

and translated by Xuanzang, the most famous pilgrim, scholar, and translator of his day. Thus a lot rides on the nature of the connection between Xuanzang and the Xinjing.

Assessing Modern History

The modern approach to the Heart Sūtra explains why the sūtra lacks the defining features of a sūtra. It explains the predominance of unidiomatic expressions in the Sanskrit text and the presence of calques from Chinese. It explains the lack of any evidence of the Heart Sūtra’s existence before the

mid- to late seventh century and also the emergence of such evidence in China a century before evidence from India. Per Nattier, the similarities and differences in the four versions of the core passage (that is, Sanskrit and Chinese versions of the Heart Sūtra and Large Sūtra) are difficult to explain otherwise. The presence of Guanyin in a Prajñāpāramitā text makes sense in a seventh century Chinese milieu whereas it does not in a fourth century Indian

context. Seen in the context of Chinese Buddhist history, it is neither unusual that a digest text was produced, nor that a Chinese text was mistaken for an Indian text. The Sanskrit translation does seem to be unusual, but it can be explained as part of a concerted effort to pass the Chinese text off as a translation to the Tang Buddhist establishment. A consequence of this approach is the need to reassess the Biography as a historical source. A lot more

work needs to be done on the relationship between medieval Buddhist historiography and Chinese historiography more generally. Similarly, we have to reconsider the bibliographic works that supplied (or confirmed) crucial details such as attribution and date of the translation. The Neidian and Kaiyuan Catalogues contributed to an emerging myth of the Heart Sūtra. They have tended to be treated as works of science, but in this view, the

85 Tokuno, “Evaluation,” 39.

respective authors were both involved in myth-making and we may need to reconsider their role in reconstructing the history of Chinese Buddhist texts. It is curious that in the historiography of the Heart Sūtra, the traditional sources are eager to associate both Xuanzang and the Heart Sūtra with

contemporary emperors, but they make almost no mention of Wu Zetian. Given her dominant role in Chinese politics from at least 655 (when she became Empress Consort), it seems remiss to ignore her. We know that Wu Zetian had Buddhist monks insert prophecies of a female emperor into a commentary and later into a

sūtra translation.86 In fact, the period of composition of the Heart Sūtra coincides with Wu’s return to the Court as Gaozong’s high ranked concubine, becoming the Empress consort in 655, and then in 660 taking de facto control of the Tang during Gaozong’s first period of illness. It was a period of

widespread palace conspiracies and political manoeuvring that the Buddhist establishment was very much involved in, along with their Daoist and Confucian rivals (who are never mentioned in the Biography). Is it not strange that Xuanzang only ever deals directly with the emperor and never with functionaries?

The fact that in 688 CE the Biography retrospectively painted Taizong and Gaozong as Buddhists, just two years before Wu Zetian takes the throne in her own right, can hardly have been a coincidence. Because the Fangshan Stele is dated within Xuanzang’s lifetime, it raises some interesting questions about his

involvement. It seems highly unlikely that a pilgrim who returned from India with literally hundreds of Indian texts and spent the rest of his life translating and commenting on them would feel the need to pass off a short Chinese digest text as Indian. In my view, this rules him out as a suspect.

Furthermore, if he had known about it, he would surely have objected to a translation being falsely attributed to him. Therefore it seems unlikely that he even knew about it. It is entirely possible that Xuanzang never knew of the text on the Fangshan Stele. In November of 659, he moved his translation group

to the Yuhua Gong 玉華宮 (Palace of Jade Flowers) in the mountains about 100 km north of Chang’an, where he stayed in seclusion translating the Prajñāpāramitā literature he had brought back from India. This work was completed in late 663, but by this time Xuanzang was seriously ill and he died

there in March 664 without ever rejoining society. Fangshan is some 850 km to the northeast and Xuanzang could easily have known nothing about events in that region. However, this still leaves open the questions of who created the digest and who translated it into Sanskrit.


At first glance, the Fangshan Stele is solid evidence for the tradition of Xuanzang’s involvement in the Heart Sūtra. The received tradition may seem to be vindicated by words set in stone. However, this historical rock is based on some rather unsuitable foundations. On reflection, the existence of the stele

in 661 CE has little explanatory power. For example, there is no point insisting on the date of the translation or the name of the translator when we know for a fact that the text on the stele is not a translation. The association of the Heart Sūtra with Xuanzang was vital for its acceptance as an Indian sūtra in translation. We can see other pieces of information that emerged over time as contributing to this acceptance in the longer term. However, because

the Fangshan Stele date is within Xuanzang’s lifetime, it leaves unanswered questions. In the traditional view, such questions never arise and thus no one sought to answer them. Perhaps by asking new questions scholars of Chinese texts may look again at their sources and discover answers. The Fangshan Stele gives us a terminus ante quem for the Chinese Heart Sūtra of 661 CE. The terminus post quem is less certain. It had to be after Kumārajīva finished his

Large Sūtra translation in 404 CE. I think we can now say that it had to have been composed after Xuanzang returned from India in 645 CE since it uses some idiosyncratic translations that he introduced (and evidence for earlier texts does not stand scrutiny). If the Biography is accurate, which is doubtful, then a version of the Heart Sūtra existed in some form by 656 CE. If the dhāraṇī was sourced from Tuoluoni ji jing (T 901) then this gives us a terminus

post quem of 654 CE. One caveat is that I have taken the date on the Fangshan Stele at face value throughout this article. Given that the evidence points away from Xuanzang being involved at all, one might wish for some confirmation that Yang Shesheng was a real person who lived at that time. As yet I have found none. The question of who could have made the Sanskrit translation remains. The translation was made before the end of the seventh century since

Woncheuk, who died in 696 CE, mentions a Sanskrit text in his commentary. 87 Competency in Sanskrit was extremely limited in China, meaning that it had to have been a Buddhist monk, probably living in one of the translation centres of

87 Lusthaus, “The Heart Sūtra,” 83. And note that Kuiji does not mention a Sanskrit text.

Chang’an or Luoyang. The translator was competent in Sanskrit but unfamiliar with Sanskrit Prajñāpāramitā idioms. The timing of the appearance of the text and the possible Yang family connection make a speculative connection to Wu Zetian intriguing, but there is too little evidence for anything more than

speculation. The attribution of a lost translation to Bodhiruchi, Wu Zetian’s accomplice in faking Buddhist prophecies, is also intriguing. Although I am involved in making (and am persuaded by) arguments for a revisionist history for the Heart Sūtra, I also think the genuine/apocryphal dichotomy as usually

framed is artificial and unhelpful. Buddhist texts are composed by human beings who, at their best, have insights into the nature of experience that they wish to communicate. The Heart Sūtra still seems to me to epitomise such insights and, arguably, its curious history makes it more interesting rather than

less. There will be those who find fault with the argument presented here—for example, my account of the historiography of the Heart Sūtra is far from complete and the gaps may seem fatal to some. Some may find my arguments lack salience in the light of authoritative Japanese scholars having inveighed

against Nattier’s thesis. Appeal to authority and special pleading may still win the day in religious arguments. Opponents may retort that there is no point in trying to cast doubt on the translation when the fact of it being a translation is carved in stone. To opponents I reply that the evidence

presented since 1992 decisively shifts the burden of proof onto those who argue that the Heart Sūtra is an Indian text. For this view to be credible we require some evidence, any evidence at all, of its existence in an Indic language (with all its peculiarities of expression) prior to the fifth century, that is prior to Kumārajīva’s translation of the Large Sūtra.


Attwood, Jayarava. “The Buddhas of the Three Times and the Chinese Origins of the Heart Sūtra.” Journal of the Oxford Centre for Buddhist Studies 15 (2018): 9–27. ———. “Epithets of the Mantra in the Heart Sūtra.” Journal of the Oxford Centre for Buddhist Studies 12 (2017): 26–57. ———. “Form is (Not) Emptiness: The Enigma at the Heart of the Heart Sūtra.” Journal of the Oxford Centre for Buddhist Studies 13 (2017): 52–80. ———. “Heart Murmurs: Some

Problems with Conze’s Prajñāpāramitāhṛdaya.” Journal of the Oxford Centre for Buddhist Studies 8 (2015): 28–48. ———. “A Note on Niṣṭhānirvāṇa in the Heart Sūtra.” Journal of the Oxford Centre for Buddhist Studies 14 (2018): 10–17. Beijing Library Metal & Stone Group and The Chinese Buddhist Books & Cultural Relics Museum Stone Sūtra Group, eds. Classified Compilation of Headings and Records of the Stone Scriptures on Mt. Fang. Beijing: Bibliographic Literature

Publishing House, 1987. [北京圖書館金石組、 中國佛教圖書文物館石經組編,《房山石經題記彙編》,北京:書 目文獻出版社,1987]. Bulbulia, J. “Charismatic Signalling.” Journal for the Study of Religion, Nature, and Culture 3, no.4 (2009): 518–551. Buswell, Robert E., ed. Chinese Buddhist Apocrypha. University of Hawaii Press, 1990. Campany, Robert F. “Notes in the Devotional Uses and Symbolic Functions of Sūtra Texts as Depicted in Early Chinese Buddhist Miracle

Tales and Hagiographies.” Journal of the International Association of Buddhist Studies 14, no.1 (1991): 28–72. Chinese Buddhist Association, ed. Stone Sūtras of Yunju Temple, Mt Fang. Beijing: Cultural Relics Publishing House, 1978. [中國佛教協會編, 《房山雲居寺石經》,北京:文物出版社,1978]. Also CBETA B 146, 26. Copp, Paul. The Body Incantatory: Spells and the Ritual Imagination in Medieval Chinese Buddhism. New York: Columbia University Press, 2014. Deeg,

M. “Has Xuanzang Really Been in Mathura? Interpretation Sinica or Interpretation Occidentalia: How to Critically Read the Records of the Chinese Pilgrims.” In Essays on East Asian Religion and Culture: Festschrift in Honour of Nishiwaki Tsuneki on the Occasion of His 65th Birthday, edited by Christian Wittern and Shi Lishan, 35–73. Kyōto: Editorial Committee, 2007.

28 Journal of Chinese Buddhist Studies Volume 32 (2019) ———. “The Political Position of Xuanzang: The Didactic Creation of an Indian Dynasty in the Xiyu ji.” In The Middle Kingdom and the Dharma Wheel: Aspects of the Relationship between the Buddhist Saṃgha and the State in Chinese History, vol. 1, edited by Thomas Juelch, 94–139. Vol. 133 of Sinica Leidensia.

2016. ———. “‘Show Me the Land Where the Buddha Dwelled...’ Xuanzang’s 玄奘Record of the Western Regions’ (Xiyu ji 西域記): A Misunderstood Text?” China Report 48, no. 1–2 (2012): 89–113. Eisenberg, A. “Emperor Gaozong, the Rise of Wu Zetian, and Factional Politics in the Early Tang.” Tang Studies 30 (2012): 45–69. Fukui Fumimasa 福井文雅. Hannnya shingyō no rekishiteki kenkyū 般若心経 の歴史 的 研 究 [Study of the history of the Heart Sūtra]. Tokyo:

Shunjūsha, 1987. ———. Hannya shingyō no sōgōteki kenkyū: rekishi, shakai, shiryō 般若心経 の総合的研究:歴史.社会.資料 [Comprehensive research about the Heart Sūtra: Historical, social, and research materials]. Tokyo: Shunjūsha, 2000. ———. Yoroppa no Tōhōgaku to Hannya shingyō kenkyū no rekishi ヨロッパ の東方学と般若心経研究の歴史 [The history of European East Asian (Oriental) studies of the Heart Sūtra]. Tokyo: Goyō shobō, Hatsubaisho Meitoku shuppansha 五曜

書房、明徳出版社, 2008. Galambos, Imre. “Simplified Characters.” In Demystifying China: New Understandings of Chinese History, edited by Naomi Standen, 187–195. Lanham: Rowman & Littlefield, 2012. Gjertson, Donald E. Miraculous Retribution: A Study and Translation of T’ang Lin’s Ming-pao chi. Berkeley: Centers for South and Southeast Asia Studies, University of California at Berkeley, 1989. Graff, David A. “The Reach of the Military: Tang.” Special issue, Chinese

Military Institutions: Journal of Chinese History 1, no. 2 (July 2017): 243– 268. Guisso, R.W.L. Wu Tse-T’ien and the Politics of Legitimation in T’ang China. Washington, D.C.: Western Washington University, 1978. He Ming and Xu Xiaoyu. “The Early Recessions of the Heart Sūtra.” In Stone Sutras Studies, vol. 1, edited by Fangshan Stone Sūtras Museum & Research Center of Fangshan Stone Sūtras and Yunju Temple, 12–28. Beijing: Beijing Yanshan Chubanshe,

2017. [賀銘、續小玉,〈早期 《心經》的版本〉,《石經研究》第一輯,房山石經博物館、房山 石經與雲居寺文化研究中心編,頁 12–28,北京:北京燕山出版社, 2017]. Hucker, Charles O. A Dictionary of Official Titles in Imperial China. Stanford, California: Stanford University Press, 1985. Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele 29

Huifeng. “Apocryphal Treatment for Conze’s Heart Problems: ‘Nonattainment’, ‘Apprehension’ and ‘Mental Hanging’ in the Prajñāpāramitā Hṛdaya.” Journal of the Oxford Centre for Buddhist Studies 6 (2014): 72– 105. Ishii, Kōsei. “Issues Surrounding the Prajñāpāramitā-hṛdaya: Doubts Concerning Jan Nattier’s Theory of a Composition by Xuanzang.” Translated by Jeffrey Kotyk. Originally published as『般若心経』をめぐる諸問題:ジャ

ン.ナテ ィエ氏の玄奘創作説を疑う. Journal of Indian and Buddhist Studies (Indogaku Bukkyogaku Kenkyu) 64, no. 1 (2015): 499–492. Ji Yun 紀贇. “Is the Heart Sūtra an Apocryphal Text?—A Re-examination.” Translated by Chin Shih-Foong. Singapore Journal of Buddhist Studies, 4 (2017): 9–113. Citing a 2018 version circulated by the translator with different pagination: 1–68. Originally published as 《心經》疑偽問題再 研究. Fuyan Buddhist Studies 福嚴佛學研究 7 (2012):

115–182. Ledderose, Lothar. “Changing the Audience: A Pivotal Period in the Great Stra Carving Project at the Cloud Dwelling Monastery near Beijing.” In Religion and Chinese Society, vol. 1, edited by John Lagerwey, 385–409. Hong Kong: The Chinese University Press and École française d’ExtrêmeOrient, 2004. Lee, Sonya S. “Transmitting Buddhism to a Future Age: The Leiyin Cave at Fangshan and Cave Temples with Stone Scriptures in Sixth Century China.” Archives

of Asian Art 60 (2010): 43–78. Li Rongxi. A Biography of the Tripiṭaka Master of the Great Ci’en Monastery of the Great Tang Dynasty. Berkeley, Calif.: Numata Center for Buddhist Translation and Research, 1995. Lin Yuanbai. “A General Survey of Fangshan Stone Sutras from the Tang Dynasty.” Modern Buddhist Studies 3 (1958): not paginated. Accessed June 1, 2018,

from-the-tang-dynasty-cachedcopy/. [林元白,〈唐代房山石经刻造概况〉,《現代佛學》3,1958] Liu Xu 劉昫. Jiu Tang shu 舊唐書. Beijing: Zhonghua Shuju, 1992. Lopez, Donald S. The Heart Sutra Explained: Indian and Tibetan Commentaries. Delhi: Sri Satguru Publications, 1988. Lusthaus, Dan. “The Heart Sūtra in Chinese Yogācāra: Some Comparative Comments on the Heart Sūtra Commentaries of Wŏnch’ŭk and K’uei-chi.” International Journal of Buddhist Thought & Culture

3 (2003): 59–103. McRae, John R. “Ch’an Commentaries on the Heart Sūtra: Preliminary Inferences on the Permutation of Chinese Buddhism.” Journal of the International Association of Buddhist Studies 11, no.2 (1988): 87–115 Nattier, Jan. “The Heart Sūtra: A Chinese Apocryphal Text?.” Journal of the International Association of Buddhist Studies 15, no. 2 (1992): 153–223.

Nourse, Benjamin. “The Heart Sūtra at Dunhuang.” Paper presented at the North American Graduate Students Conference on Buddhist Studies, Toronto, Canada, April 10, 2010. Ouyang Xiu 歐陽修. Xin Tang shu 新唐書. Beijing: Zhonghua Shuju, 1992. Radich, Michael. “On the Sources, Style and Authorship of Chapters of

the Synoptic Suvarṇaprabhāsa-Sūtra T 664 Ascribed to Paramārtha (Part1).” Annual Report of the International Research Institute for Advanced Buddhology at Soka University 17 (2014): 207–244. Rothschild, N. Harry. Wu Zhao: China’s Only Woman Emperor. New York: Pearson Longman, 2008. Sen, T. Buddhism, Diplomacy, and Trade: The Realignment of Sino-Indian Relations, 600–1400. Honolulu: Association for Asian Studies and University of Hawai’i Press, 2016.

Storch, T. The History of Chinese Buddhist Bibliography. Amherst, NY: Cambria Press, 2014. Tanahashi, Kazuki. The Heart Sutra: A Comprehensive Guide to the Classic of Mahayana Buddhism. Boulder: Shambala, 2014. Tokuno, Kyoko. “The Evaluation of Indigenous Scriptures in Chinese Buddhist Bibliographical

Catalogues.” In Chinese Buddhist Apocrypha, edited by Robert E. Buswell, Jr., 31–74. Hawai’i: University of Hawaii Press, 1990. Wang, Yongbo, and Lothar Ledderose, eds. Buddhist Stone Sutras in China: Shandong Province, Volume 1. Wiesbaden: Harrassowitz; Hangzhou: China Academy of Art Press, 2014. Wechsler, H. J. “T’ai-tsung (reign 626–49) the Consolidator.” In Sui and T’ang China, 589–906, Part I, edited by Denis Twitchett. Vol. 3 of The Cambridge History of

China. Cambridge: Cambridge University Press, 1979. Weinstein, S. “Imperial Patronage in the Formation of T’ang Buddhism.” In Perspectives on the T’ang, edited by Arthur F. Wright and Denis Twitchett, 265–306. New Haven and London: Yale University Press, 1973. Wright, A. F. “T’ang T’ai-tsung and Buddhism.” In Perspectives on the T’ang, edited by Arthur F. Wright and Denis Twitchett, 239–263. New Haven and London: Yale University Press, 1973. Zacchetti,

Stefano. In Praise of the Light: A Critical Synoptic Edition with an Annotated Translation of Chapters 1–3 of Dharmarakṣa’s Guang zan jing , Being the Earliest Chinese Translation of the Larger Prajñāpāramitā. Tokyo: The International Research Institute for Advanced Buddhology, Sōka University, 2005. Zhang Mantao, ed. Collection of Tripiṭaka Research. 2 vols. Taipei: Mahayana Culture Press, 1977. [張曼濤主編,《大藏經硏究彙編》(2 冊),台 北:大乘文化出版社,1977].