19 January 2018

The Chinese Origins of the Heart Sutra Revisited

In 1992, Jan Nattier published the watershed article in which she made a very strong argument that the Heart Sutra was compiled/composed in China. As I have discussed, the reaction in Japan was one of horror, denial, and rejection. Not much of this has filtered through to the English-speaking world, except through the Zen-based commentaries of Red Pine and Kazuaki Tanahashi. I'm working on quantifying the proportions, but most English-speaking scholars accept or at least do not reject the thesis, while some remain sceptical and on the fence (largely because there has been little follow up).

This essay will outline the case as it stands now, i.e. as stated by Jan Nattier in 1992 and extended by Huifeng in 2014, and by me in 2015 and 2017 (though I will also draw on an article that is out for peer review and two more that I'm working on that I hope to submit in 2018). There are two main areas of interest: 1. where the Heart Sutra is a quotation from the Pañcaviṃśatisāhasrikā-prajñāpāramitā-sūtra (Pañc) or "Perfection of Wisdom in 25,000 Lines" and 2. where it is an original composition. Nattier compared the words and sentence structure mainly from the former, but Huifeng and I have each extended this analysis into the conclusion.

Nattier compared four texts and showed that the most plausible way understanding their history was like this

Pañc (Sanskrit)
Pañc (Chinese)
Heart Sutra (Chinese)
Heart Sutra (Sanskrit)

The result is that the Sanskrit Heart Sutra often paraphrases Pañc. You can get a sense of what this process is like by getting Google translator to translate "form is not different from emptiness" into Mandarin, and then have Bing translator translate it back into English (note Babelfish does much less well).

There are some complications such as the potential confusion between Pañc as it appears in the Sūtra translation (T223) and as it appears embedded in the Upadeśa or commentary (T1509). But these are minor and do not affect the accuracy of the thesis.

Below are ten clues to the Chinese origins of the Heart Sutra. Since the text itself is only about 250 words, this is a very dense cluster of evidence. No.8

Core Section

In this section I will show the text as it appears in the Gilgit manuscript of Pañc, followed by Kumārajīva's translation of a similar Pañc text, followed by the parallel passage in the Sanskrit Heart Sutra. And then summarise how this contributes to the Chinese origins thesis. 

1. Form is no different from emptiness

nānyad rūpaṃ anyā śūnyatā 
rūpān na pṛthak śūnyatā

If we were being pedantic, then na anya X anya Y means "X is not one thing and Y another"; whereas X na pṛthak Y means "X is not different from Y". Two ways of saying that X and Y are the same. However, although it is grammatically correct the X na pṛthak Y  idiom is not found in the Sanskrit Prajñāpāramitā literature, whereas the na anya X anya Y idiom is. (Nattier).

Some confusion arises because there are two Chinese ways of writing this idea: 1. 非色異空 and 2. 色不異空. Version 1 negates 非 the phrase 色異空 "form is different from emptiness". Version 2 only negates the verb/adjective 異 "is different from". To distinguish them we might translate 1 as, "it is not the case that form is different from emptiness" and 2 as, "Form is not different from emptiness". T250 uses 1 and T251 uses 2.

Some older editions of the Chinese Tripiṭaka use 2 in Kumārajīva's translation of the Large Sutra, but Taishō uses 1. Taishō has 2 in the Upadeśa. It's not entirely clear what this means, but it is possible that the whole quoted text in the Heart Sutra comes from the Upadeśa.

2. All dharmas are marked with emptiness

yā śūnyatā na sā utpadyate... 
是諸 法空相不生不滅 
sarva-dharmāḥ śūnyatā-lakṣaṇā anutpannā

In Pañc it is emptiness itself that doesn't arise etc, and "all dharmas" are not mentioned (the same is true of the later Nepalese manuscripts). However, Kumārajīva's Chinese translation introduces "all dharmas" 諸 法 and syntactically makes them the subject of the sentence, changing the meaning substantially. The Heart Sutra follows Kumārajīva's Chinese translation rather than the Sanskrit text it supposedly quotes from.

The grammatical form also changes. Verbs are replaced by adjectives. See 3.

3.  Emptiness does not arise or pass away

na sā utpadyate, no nirudyate, na saṃkliśyate, na vyavadāyate, na hīyate, no vardhate 
不生 不滅不垢不淨不增不減 
anutpannā aniruddhā, amalā avimalā, anūnā aparipūrṇāḥ.

In the Sanskrit Heart Sutra a series of finite verbs in the 3rd person singular utpadyate are replaced by a series of adjectives in the masculine plural (to go with the noun dharmāḥ). 

And this is precisely the kind of confusion that medieval Chinese introduces. A character like 生 can be used as a verb utpadyate or as an adjective utpanna, or for any other nominal or verbal derivative of ut√pad and probably a number of other verbal roots. How we read it is up to us. Without a very detailed knowledge of the Prajñāpāramita idiom in Sanskrit, we are likely to make the wrong choice in this circumstance. And the translator does. 

Note also that some of the adjectives in the Heart Sutra have similar meanings, but have changed roots. For example, na hīyate "does not fall short" (< √) is translated as 不增, but then back-translated as an-ūna "not deficient". 

The list in Pañc is used frequently (in part and in full), with the same verbal roots used in this order but with different derivatives (past participles and action nouns). The list in the Heart Sutra is not found elsewhere, meaning that it was created ad hoc rather than following the usual Buddhist practice of giving standard lists. (Nattier)

This is very strong evidence for the Chinese origins thesis but is often overlooked in discussions of Nattier's article.

4. Negated lists
na cakṣur na śrotraṃ na ghrāṇaṃ na jihvā kāyo na manaḥ 
na caksuḥ-śrotra-ghrāṇa-jihvā-kāya-manāṃsi

Such lists are frequent and often combined into one long compound. However, in Pañc the compounded form is only used for positive forms. Where the terms are negated, as here, Pañc always negates each individually. On the other hand, in Chinese we see the convention of supplying one negative particle for the list as a compound.

The Sanskrit Heart Sutra follows the Chinese convention rather than the Sanskrit convention.  The fact that we find a Chinese convention in a Sanskrit text is again a strong argument for the Heart Sutra being composed in Chinese (Nattier).

Conclusion Section

Leaving behind the quoted section, we move onto the original composition. Since this section was composed in Chinese, arrows go away from the Chinese. Below the Chinese is the received translation. Above the Chinese, the Sanskrit word/phrase marked by an * is an attempt at conveying the meaning of the Chinese more accurately in the light of modern research. If you like, it is how the translator ought to have translated the text if they were more familiar with Sanskrit Prajñāpāramitā texts. On occasion, we can trace parts of the conclusion to Pañc as well. 

5. Practising non-apprehension

Kumārajīva uses 得 to represent a verbal noun from pra√āp, i.e. prāptiḥ. The author immediately uses the same character in the next phrase and it was, naturally, assumed to denote another word derived from pra√āp, i.e. aprāpritvād. However, Huifeng showed that Kumārajīva invariably translated the Sanskrit phrase anupalambhayogena using this Chinese phrase 以無所得故. The translation aprāptitvād could not have been composed in India because it relies on the ambiguity of the Chinese characters.

What's more, Huifeng argued that this word really goes with the quoted section. This qualifier moves us away from metaphysics and towards and epistemic reading of the text. (Huifeng) That is it tells us that being in the state of emptiness and practising non-apprehension of dharmas, is the only time that "no form" applies. (Attwood)

6.  His mind does not become attached

*asya cittaṃ na kvacit sajjati 
viharaty acittāvaraṇaḥ

Huifeng showed that cittāvaraṇa is simply the wrong translation here. 心無罣礙 translates "mind without attachment", where 罣礙 is a verb rather than a noun. He proposed to read 罣礙 as "hang", but I argued that it was more straightforward to read it as "attached". Similarly, the Sanskrit verb sajjati means "attach" or "become attached".  So, āvaraṇa "impediment" is also clearly the wrong translation.

The phrase na kvacit sajjati occurs in both Aṣṭa and Pañcaviṃśāti. So even though this is not a quote, we have a clear view of how Kumārajīva used this combination of characters (though Kumārajīva could be inconsistent as we have seen). Āvaraṇa is not a bad guess, but it's not consistent with the Prajñāpāramitā idiom. Which argues against composition in India. (Huifeng).

There is nothing in the Chinese that could be read as viharati "he dwells". My supposition is that the translator was struggling for a word here, especially having read 罣礙 as a noun instead of a verb, and did not know the verb sajjati. They had to improvise and this was the best they could do.

7. Not being attached


If 心無罣礙 translates "mind without attachments" then 無罣礙故 means "because [it] is without attachments". The Sanskrit Heart Sutra renders this, "because of the non-existence of mental impediments". The construction nāstitvād "because of the non-existence of " is very strange and appears to be a one-off in Sanskrit. A Sanskrit-reader can see what it means, but there are simpler and more elegant ways to negate a noun (i.e. by adding the negative prefix a-). This construction implies someone familiar with the rules of Sanskrit, who did not feel bound by the conventions of idiom. It also continues the misreading that begins above.

8. Buddhas of the Three Times

atītānāgatapratyutpannā buddhā
tryadhvavyavasthitāḥ sarvabuddhāḥ

This is the smoking gun. All being well, I'll be publishing something on this in 2018, but this phrase in the Sanskrit Heart Sutra could only have come from translating a Chinese text because it involves an idiom that developed in Chinese and is never seen elsewhere in Buddhist Sanskrit texts.

9.  Prajñāpāramita is a vidyā

mahavidyeyaṃ bhagavan yaduta prajñāpāramitā | anuttareyaṃ bhagavan vidyā yaduta prajñāpāramitā asamasameyaṃ bhagavan vidyā yaduta prajñāpāramitā 
T250: 故知般若波羅蜜是大明呪 無上明呪 無等等明呪 (8.847.c24)
prajñāpāramitā mahāmantro mahāvidyāmantro anuttaramantro ‘samasamamantraḥ

In the passage from PañcPrajñāpāramitā was described as a great vidyā (mahāvidyā 大明呪 ) and unsurpassed vidyā (anuttarā vidyā 無上明呪) and an unequalled vidyā (無等等明呪). Kumārajīva uses 明呪 to translate vidyā. But it is mistaken for two words, 明呪 "bright dhāraṇī". 

After the advent of Tantric Buddhism in China (7th C) 呪 is used to translate mantra. Tantra subsumed the previously existing spell practices under the category of mantra. This makes no sense from the perspective of a few centuries early where dhāraṇī existed entirely outside the Tantric milieu. Therefore the Sanskrit Heart Sutra came into being after the advent of Tantra when mantra chanting was finally accepted as a Buddhist practice. That said Woncheuk (613–696) makes brief mention of having a Sanskrit text, though he does not treat it as authoritative.

The form found in T250 can only have come from T223, while T251 has been modified to reflect the wording of Xuanzang's translation in T220 while keeping Kumārajīva's phrasing.

Note: Mantra recitation is still seen as non-Buddhist and frowned on in Aṣṭa. It doesn't become a feature of Buddhism until the mid-7th Century in India and about a century later in China.

10. True and not hollow

*satyā na tucchakā
satyam amithyatvāt

These are adjectives of prajñāpāramitā and should be in the feminine gender. The translator seems to have misread them as related to a mantra (grammatically neuter). He also misread 虛 which means "hollow, empty, vain" for which tucchaka is a more obvious translation than mithyā "contrarily, incorrectly, improperly".

The translator has a penchant for abstract nouns in the ablative case, which adds the sense of "because of being in the state of [the noun]". So satyam amithyatvāt literally means "truth because there is no contrariness". If these are not adjectives then this is not a well-formed sentence.


We can now conclusively say that the Heart Sutra was composed in China without any equivocation or hedging. Not only is there a weight of evidence, but No.8 is the clincher. The "three times" idiom in the Heart Sutra can only be Chinese. It is not simply that there are some suspicious looking paraphrases, but that there are passages that look like Sanskrit translations of Chinese phrases. In the case of the three times, there is no other way to construe it. The Sanskrit is definitely a translation from Chinese. 

Again, we can unequivocally say that the Heart Sutra is an apocryphon. But then so are all Mahāyāna texts. Arguably all Buddhist texts are apocryphal. There is no Buddhist equivalent of divine revelation or the preserved word of god. The best a believer could argue is that the sutras were based on a true story. There is a great deal more internal contradiction and incoherence in the literature than is usually admitted and this militates against a single source. For example, the Pāḷi suttas clearly came from multiple sources.

We can also say that the person who translated the text from Chinese into Sanskrit was unfamiliar with the idioms of the Sanskrit Prajñāpāramitā literature. They had a bias for taking active sentences in Chinese and rendering them as a series of compound adjectives, and a preference for using abstract nouns in the ablative case, even when this was inelegant. They seem to have been forced to improvise on several occasions, by a limited Sanskrit vocabulary. Lastly, they produced a unique form of Chinese influenced Sanskrit—preserving Chinese literary conventions in Sanskrit translation—which to my knowledge has no parallel. In this sense, the Heart Sutra is unique.

To my eye, this does not look like the work of someone who translated millions of words of Sanskrit Prajñāpāramitā texts into Chinese and who is still acknowledged as a master of the art of translation. In other words, the idea that Xuanzang produced this shoddy work is not really credible. It is clearly the work of an inferior and parochial mind whose point of reference for the Prajñāpāramitā tradition was Kumārajīva's Chinese translations. Which ought not to surprise us because Kumārajīva's translations have always been more popular that Xuanzang's. 

As for Tanahashi's idea that Avalokiteśvara transmitted the text of T251 to Xuanzang in India as a divine revelation (allowing him to claim that it is an "Indian text"), we would want to know why either the bodhisatva or the expert translator would only change a few key terms in Kumārajīva's text, while leaving the worst features—the mistakes—of it intact. This is not credible. 

Of course, I will need to properly frame these ideas and present the evidence to my "peers" in academia. I expect this to happen in due course. I'm hoping to get the last of the necessary corrections published this year along with one or two other papers about the text. Though getting published is less than half the battle. 80% of all articles in the humanities are never cited by another article. To date, I don't think any of my work on the Heart Sutra has been cited. There is little or no interest in the Heart Sutra in academia and little or no interest in critical scholarship amongst Buddhists. 


Related Posts with Thumbnails