The evolution of the Heart Sutra has been largely obscured by the historically dominant narratives and by the reluctance of Buddhist Studies to go beyond description and seek explanations. Watanabe Shōgo (1990) and Jan Nattier (1992) showed that the historical narratives about the Heart Sutra are pious fictions and pointed to another, rather unexpected history: the Heart Sutra was composed in China in the mid-seventh century. Their insights were subsequently confirmed by Huifeng (2014), and then I started publishing on this topic in 2015, both confirming the existing observations and adding a few of my own. While the field of Buddhist Studies (and the Buddhist world) has yet to catch up, it is now certain that the Chinese origins theory is correct.
Part of my contribution has been to step outside the usual descriptive mode of Buddhist Studies and propose explanations for the origins and evolution of the Heart Sutra. To date, my main focus has been on origins since this seemed to be the most urgent problem. More recently, I have begun to look at how the text evolved once it appeared ca 656 CE. In particular, I published an article on the varieties and relationships between the extended versions (Attwood 2021a).
In this essay, I will present a first attempt at an overview of the origins and the evolution of the Heart Sutra. I will explain why the variant texts on the Heart Sutra were produced and why they took the form that they did. In particular, I will argue that all of the major variants were created to bolster the perceived authenticity of the Heart Sutra. That the Heart Sutra appeared to lack authenticity in some eyes is hardly surprising given what we now know about its origins.
The centre of this argument is a simplified version of the stemma (or genealogy) of the Heart Sutra that I published in 2021. This diagram shows the relationships between the main inputs to the Heart Sutra and the five main versions that subsequently appeared.
Here a solid arrow represents the lines of descent, and the dotted arrow reflects the fact that Chinese extended versions repeat the text of Xīn jīng (T 251) where they overlap. Vertical spacing reflects relative chronology.
There are two processes to consider: an initial convergence in the Xīn jīng and a subsequent divergence into numerous versions of the text.
As much as the Xīn jīng reflects a convergence of texts, it also reflects a convergence of Indian and Chinese Buddhism. This may not be obvious since writing about Indian Buddhism and Chinese Buddhism tends to happen in different academic contexts that don’t communicate very well. This is sometimes referred to as “the silo mentality”. Even when there is some crossover, such as when scholars of Pāli literature study Chinese translations of Āgama texts, they see the Chinese translations as reflecting Indian culture rather than Chinese culture. Little or no attempt is made to read translated Āgama texts as Chinese texts.
This may be understandable in the case of Āgama texts but it doesn't work in the case of the Heart Sutra. The text was created in a Chinese Buddhist milieu and this is important for understanding it. However, the principal ideas in the text—Avalokiteśvara bodhisatva, Prajñāpāramitā, and dhāraṇī—all come from, and must understood in terms of, Indian Buddhism as well. Understanding the Heart Sutra requires us to have a foot in both camps, which may explain why the text has been so neglected and many of the articles that appear in print are low quality.
Convergence
The late Stefano Zacchetti (2005: 32) says that Kumārajīva's translation of Pañcaviṃśatisāhasrikā Prajñāpāramitā (Pañc) occurred during the period 29 May 403–13 Jan 404 CE. The Móhē bānrě bōluómì jīng «摩訶般若波羅蜜經» (T 223; Móhē) was completed with the help of several expert assistants and was a significant improvement on previous translations. In parallel Kumārajīva and his team translated the *Mahāprajñāpāramitopadeśa an extensive commentary attributed to Nāgārjuna. The Dàzhìdù lùn «大智度論» was completed on 1 Feb 406 CE.
During the process of translating the commentary, it became clear that Móhē required some revisions. Zacchetti says that these were complete by 18 May 404, but also says in a footnote (128) that one of Kumārajīva's principal assistants, Sengrui (僧睿; 371–438 CE), in his preface to the sutra, mentions revisions continuing to be made throughout the process of producing the Dàzhìdù lùn. The commentary and its text have guided the Chinese understanding of Prajñāpāramitā from that time onwards.
While we still don't know for sure who composed the Xīn jīng, it seems increasingly likely to have been Xuanzang. His name is associated with the earliest mention of the text, he is named in the oldest artefacts, and the earliest commentaries were by some of his close associates. My thorough exploration of this evidence has been submitted for peer review and with luck will be published in 2024. In this essay, after long and detailed consideration of the evidence, I assume that Xuanzang was the author. This is a provisional conclusion that may be subject to revision if some plausible refutation appears (implausible refutations already exist, but can be safely ignored).
Sometime between 654 CE and 26 December 656 CE, Xuanzang composed the Bānrě bōluómìduō xīn jīng «般若波羅蜜多心經» (T 251; Xīn). The earlier date reflects when the dhāraṇī was translated in the Tuóluóní jí jīng «陀羅尼集經» (T 901) by Atikūṭa. Since dhāraṇī transcription in China was never standardised, where we see an identical transcription in two different texts it is highly likely that one copied from the other. Given the nature of Xīn, which is mainly copied passages, I provisionally assume that Xuanzang copied the dhāraṇī from the Tuóluóní jí jīng. This means that Xīn could not have existed before this date. Note that Watanabe Shōgo (1990) definitively refuted the idea of Heart Sutra texts existing prior to the composition of the Xīn.
The later date is when the text is first mentioned in Buddhist literature, i.e. in the Biography of Xuanzang, Dà Táng dà Cí’ēnsì sānzàng fǎshī chuán xù «大唐大慈恩寺三藏法師傳序» (T 2053), composed by Yàncóng 彥悰 in 688 CE. The best translation of the Biography is Li (1995), but see also remarks on its historicity in Kotyk (2019).
There is no evidence of the Heart Sutra, of any kind, from any place, before 654 CE. From that date onwards, evidence in the form of inscriptions, manuscripts, catalogue entries, and commentaries proliferated and began to spread to neighbouring polities in Tibet, Korea, Japan, and Vietnam. There is no evidence of any kind from India and it now seems extremely unlikely the Heart Sutra was ever known there. The so-called Indo-Tibetan commentaries are better thought of as Tibetan commentaries attributed to Indian authors (a legitimising strategy).
Xin consists of some copied passages from Móhē, to which Xuanzang added some touches of his own (notably some novel "spellings" and the figure of Guanyin) and the dhāraṇī. Xīn became the standard Heart Sutra from that time onwards in China, Japan, Korea, and Vietnam. It is Xīn that people refer to when they say "The Heart Sutra is the most popular Mahāyāna text".
This part of the stemma also emphasises that passages from Pañc found in Hṛd did not arrive there directly. Contra the historically dominant narrative, the copied passages arrived via a Chinese intermediary, i.e. Móhē. That is, the passages copied from Pañc were not copied directly in Sanskrit from a Sanskrit source. Rather they were selected from Móhē, and only later were they (inexpertly) translated back into Sanskrit.
As Jan Nattier (1992: 170) pointed out, Hṛd bears all the hallmarks of a "back-translation". These include “unmatched but synonymous equivalents” for some Sanskrit terms and “incorrect word order, grammatical errors that can be traced to the structure of the intermediary language, and incorrect readings (due to visual confusion of certain letters or characters in the intermediary language)”.
Thus the Heart Sutra can be explained as the the result of a series of convergent processes and reflects also a convergence of Indian and Chinese Buddhist cultures. However, the text soon began to diverge into numerous versions and it is to these that that we now turn.
Divergence
From the Xīn we see three main lines of development that, as yet, cannot be precisely dated.
- The creation of the Dàmíngzhòu jīng (T 250) and its attribution to Kumārajīva
- The creation of the Sanskrit text titled Prajñāpāramitāhṛdaya.
- The creation of two extended texts.
Note that the dotted line from Xīn to the Chinese extended sutras (T 253, 254, 257) reflects the retention of the Chinese text of Xīn where they overlap. The phrase "two extended texts" refers to (1) the Chinese text of T 252, and (2) the Sanskrit Pañcaviṃśatikā-prajñāpāramitā-hṛdaya nāma dhāraṇī “Dhāraṇī named The Heart of the Perfection of Insight in Twenty-five [Lines]”. The Pañcaviṃśatikā was subsequently translated into Chinese (T 253, 254, 257).
I will take each of these versions in turn and try to show that each adds something that was perceived to be missing from the Heart Sutra. I don’t argue that there was any coordination between the three processes and, indeed, they seem to have occurred independently and over quite a long timeframe. However, together with the hagiographic stories about Xuanzang, they were embraced into the established myth of the Heart Sutra as an Indian Buddhist text.
1. Dàmíngzhòu jīng
The Heart Sutra was associated with Xuanzang from the outset and this might have been enough to ensure its place as an authentic Indian Buddhist sutra. The four early Chinese commentaries, however, still exhibit some anxiety on this score. The commentaries are:
- Bōrě bōluómìduō xīn jīng yōuzàn «般若波羅蜜多心經幽贊» “Profound Explanation of the Sutra of the Essence of Prajñāpāramitā”, by Kuījī 窺基 (T 1710)
- Fú shuō bōrě bōluómìduō xīn jīng zàn «佛說般若波羅蜜多心經贊» “Explanation of the Sutra of the Essence of Prajñāpāramitā” by Woncheuk 圓測 (T 1711)
- Bōrě bōluómìduō xīn jīng shū «般若波羅蜜多心經疏» “Commentary on the Sutra of the Essence of Prajñāpāramitā” by Jìngmài 靖邁 (X 522)
- Bōrě bōluómìduō xīn jīng lüèshū «般若波羅蜜多心經略疏» “Brief Commentary on the Sutra of the Essence of Prajñāpāramitā” by Fǎzàng 法藏 (T 1712).
Fǎzàng's commentary is traditionally dated to ca. 702 CE and he died in 712 CE. The other four are not dated, but Kuījī and Woncheuk died in 682 and 696 CE respectively. Jìngmài’s precise dates are unknown but he was roughly contemporary with Xuanzang. Thus they all date from the late seventh or early eighth centuries, and span perhaps twenty years (682–702).
Each commentator notes that the Heart Sutra lacks the expected introduction and conclusion of an authentic sutra. They also note that it consists of extractions from Prajñāpāramitā, which at that point in history seems to have been a reference to Móhē. All four men went ahead and composed their commentaries, but they left some ambiguity. Each of the subsequent developments in the Heart Sutra seems to address this ambiguity.
One approach to securing the authenticity of the text was to create the impression that previous translations existed, notably a translation attributed to the greatest of all the Chinese translators, Kumārajīva, i.e. Móhē bānrě bōluómì dàmíngzhòu jīng «摩訶般若波羅蜜大明呪經» (T 250; hereafter Dàmíngzhòu jīng). Note the title does not include the word xīn 心 "heart". Many of Kumārajiva's translations from the early fifth century are still in use today.
The idea of a Kumārajīva translation and the title it was given (Dàmíngzhòu jīng) were used to make links to another story about an even earlier translation, titled Móhē bānrě bōluómì shénzhòu 摩訶般若波羅蜜神呪. This created the myth of the "lost translation" by Zhī Qiān 支謙 (fl. 3rd century).
Watanabe (1990) thoroughly debunked this story, pointing out that it relies on a two-step process: (1) the false attribution of the shénzhòu text to Zhi Qian—in the catalogue Lìdài sānbǎo jì «歷代三寶紀» (T 2034), compiled by Fèi Chángfáng 費長房 in 598 CE—and (2) the conflation of this shénzhòu text with Dàmíngzhòu jīng. The debunking of this story (some 34 years ago) has not stopped commentators from continuing to use the idea of the "lost translations" to push back the date of translation and assert the validity of the claim that the text is Indian in origin. To be clear, neither Kumārajīva nor Zhi Qian translated the Heart Sutra. This is a false trail, deliberately laid.
In fact, the first evidence of the Dàmíngzhòu jīng, of any kind, is a mention in the Dàtáng nèidiǎn lù «大唐內典錄» "Catalogue of the Inner Canon of the Great Tang" (T 2149), published in 730 CE. As far as I can tell there are no physical texts of Dàmíngzhòu jīng before the eleventh century. The idea that a translation by Kumārajīva could be lost and then rediscovered some three hundred years after his death is extremely far-fetched and scholars have long doubted this attribution, starting with Matsumoto (1932).
That a text produced after Xīn might be retrospectively attributed to Kumārajīva to bolster its perceived authenticity is entirely plausible. It is not merely theoretical to say that Dàmíngzhòu jīng might have been used this way since this is exactly the use that has been made of it in practice. Indeed, we may say that legitimising Xīn is more or less the only use that has been made of the Dàmíngzhòu jīng. In stark contrast to Xīn, there are no commentaries on Dàmíngzhòu jīng, for example, and no prominent inscriptions or famous manuscripts. To my knowledge, Dàmíngzhòu jīng was never transmitted outside of China or translated into another language.
Dàmíngzhòu jīng, then, seems to have been created with the intention of making Xīn appear to be more authentic by pushing back the date of its composition.
2. Prajñāpāramitāhṛdaya.
In the historically dominant view, Xīn, the main text used in China, is a translation of this authentic Sanskrit version of the text. What some scholars still call "the Sanskrit original" proves that the Heart Sutra is an authentic Indian Buddhist sutra.
This view is spoiled by a detailed analysis of the text which shows that Hṛd definitely could not have borrowed its copied passages directly from Pañc. Rather the passages were clearly copied into Xīn from Móhē and then translated back into Sanskrit, leaving numerous telltales of the "back-translation" process. This was the gist of Nattier (1992) but has been confirmed numerous times by Huifeng (2014) and yours truly (see especially Attwood 2021b). The Sanskrit text is a translation of the Chinese. As such, it is not a stretch to refer to it as a "Chinese forgery".
It seems that the Sanskrit text was produced at around the same time as Dàmíngzhòu jīng was being created, i.e. in the late seventh or early eighth century. To date, we have no information on who did the translation or when. There is an ambiguous reference to "a Sanskrit text" (fàn běn 梵本) in Woncheuk's commentary (T 1711), though he does not name Hṛd and might have been referring to Pañc, since he says:
The reason there is no introduction or conclusion is that [this text] selects the essential outlines from all the Prajñā texts. It has only the main chapter, without an introduction and conclusion, just as the Guānyīn jīng is not composed of three sections (Adapted from Hyun Choo 2006: 138: emphasis added).
The Guānyīn jīng «觀音經» being originally the twenty-fifth chapter of the Miàofǎ liánhuá jīng «妙法蓮華經» (T 262; Skt. Saddharmapuṇḍarīka Sūtra), where it is titled Guānshìyīn púsà pǔmén pǐn 觀世音菩薩普門品 “Chapter of the Universal Gate of Avalokiteśvara Bodhisattva”.
I assume that a manuscript of Hṛd was forged for this purpose, and passed off as an Indian “original” since later copies of a Sanskrit text do exist (such as the famous Hōryūji manuscript). A Sanskrit manuscript is the sine qua non of authenticity for a Buddhist text in China.
Now that it has been revealed to be a back-translation, Hṛd has little philological value. Those existing studies that treat this text as "the Sanskrit original" have to be deprecated. In a forthcoming article in Asian Literature and Translation, I revise Conze's unparsable Sanskrit edition of Hṛd to make it parsable. But even this was perceived as a dead end by one of the reviewers. Study of the Sanskrit text is now quite pointless except as a unique historical artefact from early Tang China.
It is not that rare for a Chinese Buddhist text to turn out to have been composed in Chinese. Examples of this have been well documented, even in antiquity. It is exceedingly rare for a Chinese text to be translated into Sanskrit. A few examples of this have been noted. To my knowledge, the Hṛd is unique for having been successfully passed off as an authentic Indian text.
The single most important sign of the authenticity of a Buddhist text in Tang China was precisely the existence of a Sanskrit manuscript. Once again, Hṛd appears to have been created to fill a perceived gap in the authentication of Xīn.
3. Extended Texts
All of the early commentators on Xīn comment on—and attempt to explain—the absence of the usual introduction and conclusion that we expect in Buddhist sutra (I cited Woncheuk on this above; the others make similar comments). The extended text is an attempt to supply exactly these missing sections and this appears to have happened at least twice.
The first extended text appears to be Pǔbiàn zhì cáng bānrě bōluómìduō xīn jīng «普遍智藏般若波羅蜜多心經» (T 252). This text has an introduction and conclusion that are substantially different from all other extended Heart Sutra texts. The introduction is much longer and has specific details —like the presence of 100,000 bhikṣus and 70,000 bodhisatvas—that are absent in all the other texts. At the same time, the conclusion of Pǔbiàn is much shorter and more perfunctory. It is quite striking that the significant differences here have been almost entirely overlooked by other scholars. For more, though still incomplete, detail see Attwood (2021a).
All the other extended texts are clearly from one source, probably in Sanskrit, though with Tibetan and Chinese translations. The Chinese versions of the extended text (i.e. T 253, 254, 257) appear to be genuine translations from Sanskrit. That said, all the Chinese versions, including Pǔbiàn, retain the text of Xīn and only retranslate the extensions.
Ben Nourse (2010) has noted several variant extended texts in the Dunhuang cache. He suggests that these may be hybrids of standard and extended texts, but an alternative explanation might be that they are additional attempts at creating an extended text. More work needs to be done on the Dunhuang texts.
So for a third time, we see the Heart Sutra being modified to better fit a Chinese preconception about authenticity, in this case, that a real sutra has an introduction and conclusion. Only here, however, do we see two (or possibly more) attempts at the same modification.
Conclusion
I have been engaged in explaining the origins of the Heart Sutra for around twelve years. It already seems like old news to me and I find it frustrating that no one in Buddhist Studies seems willing or able to keep up with my oeuvre. At some point, the textbooks will have to change and I only hope I live long enough to see this. How this affects Buddhists is anyone's guess, but I suspect that they will continue to resist all attempts at a deflationary explanation.
The evolution of the Heart Sutra beyond its origins has been of even less interest to the field (and of no interest to Buddhists). The existence of multiple versions is, of course, well known. However, the dates of these versions have been obscured by presuppositions and this has hampered any attempts to understand how the text evolved. Watanabe (1990) debunked the attributions to Kumārajīva and Zhi Qian, making it clear that Xīn is the earliest version of the text. But his work has largely been ignored (including by me until 2023). The dhāraṇī tells us that Xīn cannot have existed before 654 CE, when Atikūṭa transcribed it in Tuóluóní jí jīng «陀羅尼集經» (T 901). This is our starting point.
As we have seen, Xīn diverged into four other versions—ie. Dàmíngzhòu jīng, Hṛd, Pǔbiàn, and Pañcaviṃśatikā. I have argued that we can see these versions as the result of three processes:
- One attempt to push back the date of composition
- One attempt to create a "Sanskrit original"
- Two attempts to provide the missing introduction and conclusion.
In other words, the evolution of the Heart Sutra was driven by conscious attempts to make the origins of the Heart Sutra fit preconceived notions of authenticity in China. These attempts largely succeeded and the associated ideas were incorporated into the historically dominant narratives of the Heart Sutra as an ancient Indian sutra text.
What my work shows is that the Heart Sutra was never ancient, Indian, or a sutra. It was created in the mid-seventh century, in China, and is modelled on a chāo jīng 抄經 "digest text" (a Chinese genre of Buddhist text). And this created anxieties related to authenticity that were addressed in a variety of ways.
I hope that it is becoming clear to my readers that the historically dominant narrative, the myth of the Heart Sutra, is largely a fiction, created quite consciously (thought without much coordination) by Chinese Buddhists. If the Heart Sutra had been merely another ancient Indian sutra, it would have been quite prosaic, notable only for its popularity in East Asia. The idea that it was composed in China and deliberately (and successfully) passed off as an ancient Indian sutra is far more interesting (even a little exciting for a textual scholar).
While I am still not an expert in Chinese Buddhist texts, if I am right about this, it makes the Heart Sutra unique amongst Buddhist texts. Moreover, I think I am right because my approach has a great deal more explanatory power than the historically dominant narratives: expanding on existing work, I have explained the origins of the text in detail. I hope this essay shows that my approach can also explain subsequent developments in the Heart Sutra as well.
~~oOo~~
Bibliography
Attwood, Jayarava. (2021a). "Preliminary Notes on the Extended Heart Sutra in Chinese." Asian Literature and Translation 8(1): 63–85. DOI: http://doi.org/10.18573/alt.53
——. (2021b). “The Chinese Origins of the Heart Sutra Revisited: A Comparative Analysis of the Chinese and Sanskrit Texts.” Journal of the International Association of Buddhist Studies 44: 13–52.
Huifeng. (2014). "Apocryphal Treatment for Conze’s Heart Problems: Non-attainment, Apprehension, and Mental Hanging in the Prajñāpāramitā Hṛdaya." Journal of the Oxford Centre for Buddhist Studies 6: 72-105.
Hyun Choo, B. (2006). "An English Translation of the Banya paramilda simgyeong chan: Wonch’uk’s Commentary on the Heart Sūtra (Prajñāpāramitā-hṛdaya-sūtra)." International Journal of Buddhist Thought & Culture 6: 121-205.
Kotyk, Jeffrey. (2019). ‘Chinese State and Buddhist Historical Sources on Xuanzang: Historicity and the Daci’en si sanzang fashi zhuan 大慈恩寺三藏法師傳’. T’oung Pao 105(5-6): 513–544. DOI: https://doi.org/10.1163/15685322-10556P01
Li, Rongxi (1995). A Biography of the Tripiṭaka Master of the Great Ci’en Monastery of the Great Tang Dynasty. Berkeley, CA.: Numata Centre of Buddhist Translation and Research.
Matsumoto, Tokumyo. (1932). Die Prajñāpāramitā Literatur. Rheinischen Friedrich-Wilhelms-Universitat zu Bonn.
Nattier, Jan (1992). "The Heart Sūtra: a Chinese apocryphal text?" Journal of the International Association of Buddhist Studies 15 (2): 153-223.
Nourse, Benjamin. (2010) "The Heart Sutra at Dunhuang." Paper presented at the North American Graduate Students Conference on Buddhist Studies. Toronto, Canada. April 10, 2010.
Watanabe, Shōgo. (1990). “Móhē bānrě bōluómì shénzhòu jīng and Móhē bānrě bōluómì dàmíngzhòu jīng, As Seen in the Sutra Catalogues.” Indogaku Bukkyōgaku Kenkyū 39-1: 54–58. [= 渡辺章悟. 1990. 「経録からみた『摩訶般若波羅蜜神呪経』と『摩訶般若波羅蜜大明呪経』」印度学仏教学研究 39-1: 54–58.]
Zacchetti, Stefano. (2005) 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. (Bibliotheca Philologica et Philosophica Buddhica, 8). Tokyo: The International Research Institute of Advanced Buddhology, Soka University.
