Notes on T253 and T254

In this post, I will show that the extended Heart Sutra text T254 is a lightly redacted or edited version of T253, rather than an independent translation from Sanskrit. Moreover, T253 incorporates all of T251, including parts with no Sanskrit counterpart, which could only result from redaction in Chinese. Moreover, since the main text is essentially just a repeat of T251, only the extended opening section and additional closing sections of T253 were translated from Sanskrit. Given the history of this text, we do have to wonder if T253 was entirely composed in Chinese. I make no judgement on the value of Chinese-produced Buddhist texts, I merely wish to clarify the history of the text.

Both texts are titled: Bānrě bōluómìduō xīn jīng «般若波羅蜜多心經» "The Sutra of the Heart of Perfection of Insight". T253 is attributed to a Khotanese monk, Bānrě 般若 (Prajñā), a Chinese monk Lìyán 利言, and "others" (děng 等), and traditionally dated to ca. 788 CE. (A note in the Zhēnyuán lù «貞元錄» (T 2157), suggests that Bānrě 般若 had Sanskrit Heart Sutra text - I'll likely cover this in a  separate post). The text of T253 is 545 characters (not counting spaces or punctuation). It seems that Prajñā did not speak Chinese, and Lìyán 利言 acted as his translator in China (note: I will soon post an exploration of the dual identity of Lìyán 利言).

T254 is attributed to Zhìhuì Lún 智慧輪 (*Prajñācakra), a noted tantrika, and traditionally dated to 861 CE. I still cannot find where this attribution comes from. The text is 569 characters.

T253 is recorded in the Tripitaka Koreana, but T254 is not. In yet another forthcoming essay will look at the TK, which only records T250, T251, 252, T253, and 257.

After 12 years of literature searching, I know of no secondary literature on this issue except for my own preliminary investigations here and in Attwood (2021).

My method is simple. I break the text into 18 sentences. For sentences from the introduction and conclusion, I will paste the sentences from the two texts alongside each other, and I will compare them both with the Sanskrit, and occasionally with T251. There appear to be some errors in both T253 and T254, which I will indicate using strikethrough and explain in notes. At the very end, I've appended a list of all the differences, sentence by sentence.

My aim was to have the same punctuation in both T253 and T254, which Taishō does not provide and thus it required some changes. I have not noted such changes. Sometimes the punctuation is incorrect, which I have noted. 

I will also provide Middle Chinese transcriptions where it seems helpful. My approach to MC is indicative rather than systematic, meaning that anyone interested in MC should consult the experts. Two main problems make attempting to provide MC transcriptions difficult: no one system of reconstruction covers all the characters used in the Heart Sutra; and all of the many systems use different notation. The notation used is often idiosyncratic, but even when the International Phonetic Alphabet is used, most of the "letters" are unfamiliar most readers. That said, I have tried to use the IPA because at least it is easy to look up. 

Since the titles are identical, we begin with the text itself.

1. Introduction

Sentence 1.

253	如是我聞：一時佛　　在王舍城耆闍崛山中，與大比丘眾及　菩薩眾俱。
254	如是我聞：一時薄誐梵住王舍城　鷲峯山中，與大苾蒭眾及大菩薩眾俱。
Skt.	evaṃ mayā śrutam: ekasmin samaye bhagavān rājagṛhe viharati sma gṛdhrakūṭe parvate mahatā bhikṣusaṃghena sārdhaṃ mahatā ca bodhisattvasaṃghena |

One time 一時, buddha 佛 dwelled 在 Rājagṛha 王舍城, Gṛdhrakūṭa Mountain 耆闍崛山 on 中; together [with] 與 a great 大 bhikṣu 比丘 congregation 眾 and 及 a [great 大] bodhisatva 菩薩 congregation 眾 accompanied 俱.

The word order of the Chinese is similar to English, except for certain words that we would put at the beginning; go at the end. Here the word jù 俱 "accompanied by" would come earlier in English, as would zhōng 中 when used as a locative preposition "in, on".

Qí dū jué 耆闍崛 (MC: /gi dʑia gʷiᴇt/) is a partial transcription of Gṛdhrakūṭa (Pāli: Gijjhakūṭa), plus shān 山 "mountain"; while Jiù fēng shān 鷲峯山 is a translation: vulture 鷲 peak 峯 mountain 山.

Note that final sounds (in particular) were affected by changes in pronunciation of both Indic and Chinese languages. Notably for many Indic speakers, a final vowel, especially -a, would be de-emphasised to the point of ambiguity or absence (noted by Edgerton in his Grammar of Buddhist Hydrid Sanskrit). A modern Hindi speaker, for example, would pronounce Gṛdhrakūṭa as /ɡɾɪ d̪ʱɾə kuːʈ/ (Pāli /ɡɪd͡ʒ.d͡ʒʰə.kuːʈə/). So the transcription 耆闍崛 /gi dʑia gʷiᴇt/ could accurately reflect what a Chinese interpreter heard an Indian informant saying, even if the spelling seems a little off. And in the case of these two texts, neither of the named translators, Bānrě 般若 nor Zhìhuì Lún 智慧輪, could speak Chinese. 

Báo'éfàn 薄誐梵 (MC: /bɑk ngɑ bɨuam/), in T254, is an unusual transcription of Bhagavan that is used only this one time. Elsewhere T254 uses the expected Chinese phrase Shìzūn 世尊, literally "world honoured".

This is very much a standard opening for a Buddhist sutra. Compare these other examples from Kumārajīva's oeuvre:

• T 227: 如是我聞：一時佛在王舍城耆闍崛山中，與大比丘僧千二百五十人俱，...
• T 235: 如是我聞：一時佛在舍衛國祇樹給孤獨園，與大比丘眾千二百五十人俱。
• T 245: 如是我聞：一時佛住王舍城耆闍崛山中，與大比丘眾八百萬億，...

---

Sentence 2.

253	時，佛世尊即入三昧，　名廣大甚深。
254	爾時，世尊　入三摩地，名廣大甚深照見。
Skt	tena khalu punaḥ samayena bhagavān gambhīrāvabhāsaṃ nāma dharmaparyāyaṃ bhāsitvā samādhiṃ samāpannaḥ |

[Then 爾] at that time 時, Buddha 佛 Bhagavan 世尊 then 即 entered 入 [a] samādhi 三昧， named 名 vast 廣大 and profound 甚深.

Contrast shí 時 "then" with ěr shí 爾時 "at that time", both of which translate Sanskrit: tena samayena and ignore khalu "indeed" and punaḥ "again". Note that in Sentence 3: T253 爾時 vs T254 時.

While T254 began with a transcription Bhagavan 薄誐梵, here it reverts to the more usual translation Shìzūn 世尊 "World honoured". 

Sānmèi 三昧 and Sānmóde 三摩地 both transcribe samādhi. In S16 both texts have 三摩地.

T254 appends zhàojiàn 照見 (a binomial verb meaning "to observe") to the end of the sentence, but this is an obvious mistake. Possibly an eye-slip given that the same binomial occurs in the next sentence. Whatever the explanation, this phrase makes no sense here (and has no counterpart in Sanskrit). Were I editing this text, I would have omitted these two characters.

After the completely standard opening, we immediately run into problems here because the Sanskrit is substantially different from the "translations".

Then tena indeed khalu again punaḥ at that time samayena the Well-Endowed bhagavān profound illumination gambhīrāvabhāsaṃ named nāma Dharma teaching dharmaparyāyaṃ having spoken bhāsitvā meditation samādhiṃ entered samāpannaḥ |

The main clause is "At that time the Bhagavan entered samādhi." (tena samayena bhagavān samādhiṃ samāpannaḥ). The subordinate clause is "having spoken a Dharma teaching named profound illumination" (gambhīrāvabhāsaṃ nāma dharmaparyāyaṃ bhāsitvā). The Sanskrit text unambiguously tells us that having spoken (bhāsitvā) the dharmaparyāya the Buddha entered meditation (samādhiṃ samāpannaḥ). But no Chinese text has a term corresponding to dharmaparyāya.

Words like khalu and punaḥ tend to be ignored in the Chinese translations (as they often are in English translations).

Surprisingly, neither T253 nor T254 gets this sentence right. In T253 and T254 (and all of the other extended texts, including the Tibetan) the name gambhīrāvabhāsa is attached to the samādhi, not to the dharmaparyāya. Indeed, the dharmaparyāya is entirely omitted from T253 and T254. This kind of gross mismatch is problematic for the idea that either T253 or T254 are translations from Sanskrit. 

Reading this passage in isolation, it's not entirely clear what has happened here. But it is quite an oddity when every single one of the extant "translations", in both Chinese and Tibetan, has misapprehended, and thus mistranslated, this sentence in more or less the same way. Given the history of the text, perhaps we should be questioning which came first? Or perhaps the Sanskrit text was changed after the fact and we no longer have the original that was available in the early eighth century?

---

Sentence 3.

For this section, I include the relevant text from T251. I've colour-coded based on T251 to make it easier to see how all of this first sentence was assimilated into the expanded introduction. You can see that the whole of the text from T251 is preserved in both T253 and T254, with only minor variations.

251	觀自在菩薩行深般若波羅蜜多時，照見五蘊，皆空，度一切苦厄。
253	爾時，眾中有菩薩摩訶薩，名觀自在，行深般若波羅蜜多時，照見五蘊，皆空，離諸苦厄。
254	時，眾中有一菩薩摩訶薩，名觀世音自在。行甚深般若波羅蜜多行時，照見五蘊，自性皆空。
Skt.	tena ca samayena āryāvalokiteśvaro bodhisattvo mahāsattvo gambhīrāyāṃ prajñāpāramitāyāṃ caryāṃ caramāṇaḥ evaṃ vyavalokayati sma pañca skandhāṃs tāṃś ca svabhāvaśūnyaṃ paśyati sma |

Then 爾 at that time 時 within the congregation 眾中 there was 有 a bodhisatva mahāsatva 菩薩摩訶薩，named 名 Guanzizai 觀自在，practiced 行 deep 深 prajñāpāramitā 般若波羅蜜多 {while 時}, examined 照見 the five skandhas 五蘊, all were absent 皆空，[he] got away from 離 all 諸 suffering 苦厄.

Here the Sanskrit largely follows the text of the standard Prajñāpāramitāhṛdaya, including doubling up words from √car: cāryāṃ caramānaḥ. This is not idiomatic Buddhist Sanskrit. We never see "practising the practice of prajñāpāramitrā" in other Sanskrit Prajñāpāramitā texts. The word cārya seems never to be applied to Prajñāpāramitā in this way. On the other hand, we regularly see terms such as bodhisatvacārya and brahmacārya.

For example in volume one of Kimura's (2009) edition of Pañc alone there are 193 occurrences of passages ending ...bodhisattvena mahāsattvena prajñāpāramitāyāṃ śikṣitavyam "...bodhisatvas should be trained in prajñāpāramitā". According to convention, one may "train in prajñāpāramitā" (indicated by verbs from √śikṣ), and one may "practice prajñāpāramitā" (verbs from √car), but no one ever "trains in the practice of prajñāpāramitā".

Moreover, there are 73 occurrences of ...bodhisattvena mahāsattvena prajñāpāramitāyāṃ caratā with the present active participle caratā, and none at all with the present middle participle caramānaḥ. In the standard text, a mismatched synonym (like this) is taken to be a result of back-translating Chinese to Sanskrit. That such expressions were retained unchanged in the extended Sanskrit text is significant because it tells us that the "translators" were aware of the standard text and gave it priority. 

Here, it is the Chinese texts that conform to the expected idiom, which is one line of argument for the composition of the Heart Sutra in Chinese.

Notably, no known Sanskrit text has an equivalent to dù yīqiè kǔ è 度一切苦厄. Although T253 changes the verb from dù 度 "overcome" to lí 離 "separate [from]", and yīqiè 一切 "everything, all" to zhū 諸 "all", the overall meaning is the same. While the term is absent from this sentence of T254, the phrase lí zhū kǔ è 離諸苦厄 is appended to the end of S7 (see below).

There is also some variation in the name. The basic form of the name in Sanskrit is Avalokiteśvara bodhisatva. On this name see: Revisiting Avalokiteśvara in the Heart Sutra (10 April 2020). The Sanskrit has been elaborated by the addition of āryya "noble" and mahāsatva.

Over time, it becomes increasingly common to see mahāsatva added to bodhisatva. I still don't really understand what mahāsatva means. The traditional explanations ignore Classical Sanskrit grammar. As a suffix, it forms bahuvrīhi compounds of the type X-sattva, meaning "whose nature is X", "having X as their essence", and so on. A bodhisatva, therefore, is a person "whose nature is awakening".

T251 uses Xuanzang's preferred transcription of the name: Guānzìzài púsà 觀自在菩薩. T253 also uses the name Guānzìzài 觀自在, but does so in the context of being within "the congregation 眾中有 of bodhisatva 菩薩 mahāsatvas 摩訶薩"; all later occurrences in T253 add púsà móhēsà 菩薩摩訶薩 after the name as convention dictates.

However, both before and after Xuanzang, Chinese Buddhists strongly preferred the name Guānshìyīn 觀世音, often abbreviated to Guānyīn 觀音. Xuanzang's spelling never really took off. The fact that T253 uses Guānzìzài 觀自在 suggests that Bānrě 般若 or Lìyán 利言 had some a connection to Xuanzang, though such a connection is not obvious or explicit, since Xuanzang died in 664, some decades before Bānrě 般若 arrived in China.  

T254 adopts an unusual hybrid of the two forms Guānzìzàishìyīn 觀世音自在, throughout. Both of the later translations, i.e. T255 and T257, also spell the name Guānzìzài 觀自在, long after everyone else reverted to Guānyīn. Perhaps this reflects the lasting influence of T251, but it certainly speaks to the power of the existing Chinese texts in creating these new translations.

爾時/時 already discussed above.

T253 and T254 consciously place Guanyin amidst the congregation:  in the saṃgha 眾中 there was 有 [one 一] bodhisatva mahāsatva 菩薩摩訶薩.  T254 adds yī 一 "one", seemingly to emphasise that only one of the bodhisatvas assembled was called 觀世音自在. Which seems completely unnecessary, but is not wrong. This change cannot have been inspired by the Sanskrit text, because S3 of that version contains no mention of the bodhisatvasaṃgha.

T254 adds a superfluous superlative shèn 甚 "very, extremely" to the already superlative adjective shēn 深 "deep, profound" (Skt. gambhīra).

A problem I have commented on before is that the Sanskrit text has a structure based on complimentary verbs for "looking" (vyava√lok) and "seeing" (√paś). This look/see structure appears to be absent in the Chinese expression that T253 has copied verbatim from T251: zhàojiàn wǔyùn, jiē kōng 照見五蘊, 皆空 (universally printed without the comma I have added). In my view, "looking" is explicit in 照見, but "seeing" is implicit and what was seen was jiē kōng 皆空 "all absent". Thus I see the canonical 照見五蘊皆空 as two clauses: 照見五蘊，皆空 "[He] observed the five branches, [and what he saw was that] all [were] absent."

T251 has no term corresponding to svabhāva. T253 simply reproduces T251, but T254 has added a term meaning svabhāva, i.e. zìxìng 自性. This may reflect influence from Sanskrit, though T254 still doesn't translate paśyati sma "he saw", it's still a moot question whether is passage was translated from Sanskrit or composed/redacted in Chinese. 

---

Sentence 4.

T253	即時，　　舍利弗，承佛威力，合掌恭敬，白觀自在　　菩薩摩訶薩言：
T254	即時，具壽舍利子，承佛威神，合掌恭敬，白觀世音自在菩薩摩訶薩言：
Skt.	athāyuṣmān śāriputro buddhānubhāvena āryāvalokiteśvaraṃ bodhisattvam etad avocat-

At that 即 time 時，Śāriputra 舍利弗，receieved 承 buddha 佛 power 威力，joining hands 合掌 respectfully 恭敬，addressing 白 Guānzìzài 觀自在 bodhisatva 菩薩 mahāsatva 摩訶薩 he said 言：

T254 adds the honorific jùshòu 具壽, i.e. jù 具 "possessing, having" shòu 壽 "longevity, life" which equates to Skt. āyuṣmant "Elder" (āyuḥ life; -mant a possessive suffix). In Prajñāpāramitā literature generally, we expect someone like Śāriputra will always be referred to as Elder Śāriputra (āyuṣmān śāriputraḥ), although it is sometimes dropped when he is being addressed as an inferior.

Note also that honorifics such as āyuṣmant always precede the name; whereas descriptors indicating rank or status, e.g. bodhisatva, or sthavīra, always follow the name. 

There is no counterpart of hézhǎng gōngjìng 合掌恭敬 "joining hands respectfully" in any Sanskrit manuscript. Though it is a common Chinese expression and a very common Buddhist way of showing respect. This again suggests redaction in Chinese rather than translation from Sanskrit. 

Some notes on anubhāva. A strange word whose etymology is no help. The word is used to indicate that someone is acting from an impetus emanating from the Buddha, i.e. his anubhāva. This could be something as banal as the Buddha's charisma and authority, but it tends to suggest a more supernatural power that could influence his followers to state truths that they might not be able to attest from their own insight.

However, in Aṣṭa the Buddha asks Subhūti to instruct the bodhisatvas and Śāriputra wonders:

kim ayam āyuṣmān subhūtiḥ sthavira ātmīyena svakena prajñāpratibhānabalādhānena svakena prajñāpratibhānabalādhiṣṭhānena bodhisattvānāṃ mahāsattvānāṃ prajñāpāramitām upadekṣyati utāho buddhānubhāveneti?

“Will this Elder Subhūti senior, explain the perfection of insight of the bodhisatva-mahāsatvas by employing the power (bala) his own (ātmīyena) personal (svakena) insight and eloquence, based on the power of his own insight and eloquence, or will he speak by the authority (anubhāva) of the Buddha?”

Subhūti's answer is that a disciple who has practised the Dharma and realised the truth for themselves and carries that realisation with them always speaks with the anubhāva of the Buddha. See also my discussion of this passage: Aṣṭasāhasrikā: Insight and Ongoing Transformation. (01 December 2017).

In Chinese, we see wēilì 威力 (T253) and wēishén 威神 (T254) for anubhāva. Wēi 威 means "power". Since lì 力 also means "strength, power, etc.", T253 could be said to have a double translation. By contrast, shén 神 refers to supernatural beings, powers, and events (giving anubhāva a magical quality). Elsewhere, wēishénlì 威神力 has been used to translate both anubhāva and adhiṣṭhāna.

---

Sentence 5.

T253	「善男子！若有欲學甚深般若波羅蜜多行者，云何修行？」。
T254	聖者！若有欲學甚深般若波羅蜜多行，　云何修行？」
Skt.	yaḥ kaścit kulaputro vā kuladuhitā vā asyāṃ gambhīrāyāṃ prajñāpāramitāyāṃ caryāṃ car[i]tukāmaḥ, kathaṃ śikṣitavyaḥ?

"Kulaputra 善男子 if 若 there is 有 desire [for] 欲 training [in] 學 the extremely 甚 profound 深 prajñāpāramitā 般若波羅蜜多 practice 行，how 云何 cultivate 修 [that] practice 行?"

I've commented in print that it is odd for Śāriputra to address Guanyin as "kulaputra" since it's generally used for social juniors (Attwood 2021: 74-75). It would be unremarkable for Gunyin to refer to Śāriputra as kulaputra. Perhaps Zhìhuì Lún 智慧輪, the redactor of T254, felt this as he changed the vocative to shèngzhě 聖者 "Venerable One", which carries the implication that the one being addressed is awakened. This is a more appropriate form of address in the context.

In fact, it is rare in Prajñāpāramitā for any person to be directly addressed as kulaputra, except for one circumstance, the sadhukāra, which we encounter in Sentence 17. This involves saying "sadhu sadhu" to the person (always twice); and is the Prajñāpāramitā it is very often "sadhu sadhu kulaputra" Sadhu likely derives from √sādh "succeed" and means "good". The sadhukāra reflects approval by, and acknowledgement from, a superior. 

More often in Prajñāpāramitā literature, kulaputra (in the nominative case) is the agent in a hypothetical: "How does a kulaputra cultivate prajñāpāramitā?" This is how the Sanskrit Heart Sutra text uses the term, but it is not how the Chinese texts use it. 

Note that neither Chinese text includes the kuladuhitṛ here in the question, but both include her in the answer (see S7). The extant Sanskrit texts mentions the kuladuhitṛ in both cases. 

If T253 and T254 are translations, and if the extant Sanskrit texts reflect the text translated, then Bānrě 般若, Lìyán 利言, et al have made an error here. It seems that Zhìhuì Lún 智慧輪 attempts to correct this error in T254 so that at least Guānyīn is not being addressed inappropriately as kulaputra, but he has not corrected it so that T254 conforms to the Sanskrit. In this case, even if we postulate a non-extant Sanskrit source, it would be very odd indeed if Avalokiteśvara (or anyone) was addressed in it as kulaputra. This is consistent with T254 being redacted in Chinese with little or no reference to the Sanskrit.

In Attwood (2021: 74-75), I outlined my view that, in this context, kulaputra is a way of referring to a bhikṣu in the abstract; and kuladuhutṛ to a bhikṣunī. Kula is a general term for any group of humans or animals which carries no implication of status (high or low). So kulaputra is literally "group-son" and kuladuhitṛ "group-daughter" (duhitṛ is cognate with daughter). Both terms seem to be tatpuruṣa compounds: a son or daughter of the group. Note that it was not uncommon for bhikṣus to refer to themselves as śakyaputra "a son of the Śakya(s)"; where "the Śakya" is the Buddha and "the Śakyas" refers to the Buddha's tribe. So putra is literally "son", but is used figuratively to mean something like "disciple".

The association with "nobility", "goodness" or "high status" is also implied by the Chinese translation shàn nán zi 善男子 literally "good male child". Shàn 善 "good, kind, virtuous, friendly; apt, adept, expert" is also a character used to translate kuśala "good, adept, etc". However, this appears to be a mistake based on misreading certain early Buddhist texts which appear to conflate kulaputra with sujāta "well born" (notably the Ambaṭṭha Sutta DN 3). In fact, these are two different adjectives in a longer list of distinctive qualities. They are not synonyms. So I think all such translations as "noble son", "son of good family", etc are simply wrong.

This passage is important because it sets things up so that the unattributed statements in the main body of the text appear to be the answer to this question. As in the TV game show Jeopardy, the "translator" had to come up with a suitable question to fit an existing answer. It's artful enough in its own way, but we know that the answer—i.e. the text of T251—came first. 

Common forms of indicating who is speaking—such as idam avocat, evaṃ ukte, or āha (and their Chinese counterparts)—do not occur in the standard Heart Sutra. The inescapable conclusion is that, despite centuries of tradition, the words in the text are not uttered by Guānyīn. These passages come from Pañc where they are spoken by the Buddha (something that Woncheuk is cognizant of in his commentary). See: Guanyin Does Not Speak in the Heart Sutra (9 February 2024).

In The Extended Heart Sutra: Avalokiteśvara Preaches (14 August 2020) I noted that cartukāma (yù 欲) is a variant of caritukāma "one whose desire is to practice"; a bahuvrīhi compound combining the infinitive caritum/cartum "to practice" and kāma "desire".

T254 drops the zhě 者 from xíngzhě 行者, which makes more sense, since if xíng 行 is "practice" then xíngzhě 行者 should be "a practitioner". And this doesn't fit the context.

In this passage, the changes to T254 appear to be corrections of mistakes in T253, though without reference to a Sanskrit text.

---

Sentence 6.

253	如是問已，爾時，觀自在菩薩摩訶薩　　告具壽舍利弗言：
254	如是問已，爾時，觀世音自在菩薩摩訶薩 告具壽舍利子言：
Skt.	Evam ukte āryāvalokiteśvaro bodhisattvo mahāsattvaḥ āyuṣmantaṃ śāriputram etad avocat:

Thus 如是 asked 問 already 已, at that time 爾時, Guānzìzài 觀自在 bodhisatva 菩薩 mahāsatva 摩訶薩 spoke [to] 告 Elder 具壽 Śāriputra 舍利弗 saying 言：

Of the three texts, the easiest to understand is the Sanskrit: "That being said, Noble Avalokiteśvara bodhisatva mahāsatva said this to Elder Śāriputra". This is very much a standard Buddhist locution in Sanskrit and Pāli. We see such sentences on almost every page of every sutra/sutta.

The Chinese text is unnecessarily complicated. The phrase rúshì wèn yǐ 如是問已 "having been asked" appears to correspond to Sanskrit evaṃ ukte "when this was said". As per the Sanskrit, this is one grammatically simple sentence. If 如是問已 is to be included then the following full stop in CBETA is clearly incorrect because this is one sentence. I've changed the full stop to a comma. However, note that the Sanskrit merely says "thus spoken" (evam ukte), while the Chinese explicitly mentions the "question" (wèn 問).

The real problem here is ěr shí 爾時 "at that time" in both Chinese texts. It's not really needed or wanted here and could be omitted with no change to the meaning of the sentence. Nor is it implied by the extant Sanskrit text. So once again the connection between the Sanskrit "source" and the Chinese "target" texts is not a straightforward "translation". 

Note that T253 includes Śāriputra's honorific this time, having previously omitted it. The inclusion is not compulsory but it is included more often than not in Prajñāpāramitā literature.

---

Sentence 7.

253	「舍利子！若　善男子、善女人，行甚深般若波羅蜜多行時，應觀五蘊性空。
254	「舍利子！若有善男子、善女人，行甚深般若波羅蜜多行時，應照見五蘊自性皆空，離諸苦厄。
Skt.	yaḥ kaścicChāriputra kulaputro va kuladuhitā vā [asyāṃ] gambhīrāyāṃ prajñāpāramitāyāṃ caryāṃ cartukāmaḥ, tenaivaṃ vyavalokitavyam-pañca skandhāṃs tāṃś ca svabhāvaśūnyān samanupaśyati sma |

"Śāriputra 「舍利子！if 若 [a] kulaputra 善男子、[or] kuladuhitṛ 善女人 [desires] to practice 行 very profound 甚深 prajñāpāramitā 般若波羅蜜多 practice 行 when 時，[they] should 應 observe 觀 five 五 skandhas 蘊 nature 性 absence 空。

T254 inserts yǒu 有 into ruò shàn nánzǐ 若善男子 "If a kulaputra..." giving ruò yǒu shàn nánzǐ 若有善男子, which seems to say "if that kulaputra...", or "if there is a kulaputra..."; whereas the Sanskrit still seems to discuss kulaputra in the abstract.

Despite asking the question solely in terms of kulaputra, T253 and T254 both answer in terms of both kulaputra and kuladuhitṛ. This inconsistency seems to be the result of poor editing.

While T253 and T254 now omit yù 欲 (cartukāma), it is repeated in the Sanskrit. Perhaps in Chinese, the context makes it clear, but without yù 欲, the sentence doesn't entirely make sense on its own. 

Note the addition of 離諸苦厄 in T254 at the end (it occurs in Sentence 3 of T253). Placed here it doesn't fit the context. In sentence 3, it follows on from the realisation that the skandhas are absent. Here, the sentence is recommending this same practice to the unenlightened kulaputra or kuladuhitṛ. So it makes no sense to say "free from all suffering" at this point. Note the absence of any such statement in Sanskrit. The introduction of T254 appears to have been edited in Chinese and the editor seems not to have been cognizant of the Sanskrit text.

---

2. Middle Section (i.e. T 251)

The middle of T253, sentences 8-14 is identical to T251 except for (14) the dhāraṇī. T254 has a few changes.

In sentence 9:

253	舍利子！是諸法空相，不生不滅、不垢不淨、不增不減。
254	舍利子！是諸法性相空，不生不滅、不垢不淨、不減不增。

Where T251 and T253 have "Here 是 all 諸 dharmas 法 [are] absence 空 marked 相", T254 has an extra character xìng 性, which also means "quality, nature" so really just reinforces xiāng 相 which generally translates lakṣana "characteristic".

Note also that T254 reverses the order of the last pair of phrases: 不減不增. This actually brings it into conformity with the Sanskrit text, i.e. T251/253 have their pairs reversed compared to the Sanskrit.

The original passage from Pañc is: ya Śāradvatīputra śūnyatā na sā utpadyate, no nirudhyate, na saṃkliśyate, na vyavadāyate, na hīyate, no vardhate. This clearly places "not diminishing" (na hīyate) before "not growing" (no vardhate). So this is the order we expect. Still, if we were translating into English, it feels more idiomatic to say "not growing, not diminishing", especially in the light of "not arising, not ceasing". Similarly, in English we'd prefer to say "not pure, not stained", giving the positive quality first and the negative second.

Where T253, sentence 11, has jiùjìng nièpán 究竟涅槃 equating to niṣṭhanirvāṇa, T254 has jiùjìng jìrán 究竟寂然。I cannot explain this change. It seems entirely arbitrary, since nièpán 涅槃 is a very common translation of nirvāṇa, and jìrán 寂然 is more often used to translate such Sanskrit terms as upaśānta, praśānta, śama, śamatha, and śānti, all representing the idea of "peace" or "peacefulness". Presuming that the creator of T254 was translating from a Sanskrit text that said nirvāṇa (which all of the extant documents do) this is a very odd change to make. Again, this redaction appears to have occurred in Chinese without reference to the Sanskrit text.

A similar change occurs in the passage 三世諸佛依般若波羅蜜多故，得阿耨多羅三藐三菩提。"All the buddhas of the three times relying on this prajñāpāramitā completely attained anuttarāṃ samyaksaṃbodhim." To this, T254 appends xiàn chéng zhèngjué 現成正覺, "manifests perfect awakening" which again corresponds to no known Sanskrit text. This phrase appears to be a translation of anuttarāṃ samyaksaṃbodhim whereas ānòuduōluó sānmiǎosānpútí 阿耨多羅三藐三菩提 is a transcription. Perhaps the addition of the translation was a commentarial gloss?

In the epithets section, sentence 13, T254 changes the ambiguous zhòu 呪 "incantation" to zhēnyán 真言 "true words", a term only used by Tantrikas and which can only mean "mantra". As we will see, T254 also changes the dhāraṇī into a mantra. This is consistent with the attributed author, Zhìhuì Lún 智慧輪 (*Prajñācakra), being a tantrika. And this change is consistent with the Sanskrit text.

---

Sentence 14. The Dhāraṇī

251	即說呪曰　揭帝　揭帝　般羅揭帝　般羅僧揭帝　菩提　薩婆訶。
253	即說呪曰「櫱諦　櫱諦　波羅櫱諦　波羅僧櫱諦　菩提　娑(蘇紇反)婆訶
254	即說真言「唵(引)　誐帝　誐帝　播(引)囉誐帝　播(引)囉散誐帝　冒(引)地　娑縛(二合)賀(引)
Skt.	Tadyathā: gate gate pāragate pārasaṃgate bodhi svāhā |

Neither T253 nor T254 completely follow the dhāraṇī transcription in T251. T254 takes an explicitly tantric approach.

Where we find tadyathā "this way" in Sanskrit, T251 has "now 即 speak 說 the incantation 呪 that says 曰". The word is frequently translated dátā 怛他 or some other variant. T253 preserves the reading in T251. T254 again substitutes 真言 for 呪, emphasising the idea that "mantra" was intended, even though we know that the original word (in Pañc) was in fact vidyā.

T254 inserts 唵(引) /ʔəm/, i.e oṃ. Here, (引) indicates a long vowel. In Sanskrit, e and o are always long; the short e and o were assimilated to short a prior to Vedic (aka Old Indic) emerging as a distinct language. Hence we don't usually mark them as long with the macron: ē and ō as we do with other long vowels, i.e. ā, ī, ū, and ṝ.

In T245, T251 揭 /kæt/ representing Skt ga is changed to 櫱 /ŋi̯ät/ in T253 and 誐 /ngɑ/ in T254. 

The anomalous 般 /pan/ in T251 becomes the expected 波 /pa/ and 播(引) /pā/

In T253 the expression sū hé fǎn 蘇紇反 represents a fǎnqiè 反切 formula, where sū 蘇 provides the initial sound and hé 紇 provides the final sound and tone. However, I don't think this makes sense in context and a Middle Chinese transcription is no help. T253 has a three-character transcription of svāhā, i.e. suō pó hē 娑婆訶, where 娑 and 婆 together represent svā. T254 reflects a more common approach: 娑縛(二合) which tells us to combine the two sounds (MC /sa bɨak/) into one conjunct consonant sound supposed to represent svā in a language that lacks a /v/ sound.

---

3. The Concluding Section

Sentence 15.

253	「如是，舍利弗！諸菩薩摩訶薩，於甚深般若波羅蜜多行，應如是行。」如是說已。
254	「如是，舍利子！諸菩薩摩訶薩，於甚深般若波羅蜜多行，應如是學。」
Skt.	evaṃ śāriputra gambhīrāyāṃ prajñāpāramitāyāṃ caryāyāṃ śikṣitavyaṃ bodhisattvena |

In this way 如是，Śāriputra 舍利弗！all 諸 bodhisatva 菩薩 mahāsatva 摩訶薩，regarding 於 very profound 甚深 prajñāpāramitā 般若波羅蜜多 practice 行，should 應 in this way 如是 practice 行。」

The addition of rúshì shuō yǐ 如是說已 "Thus, having spoken" at the end of the line in T253 should correspond to evaṃ ukte "when this was said" or to etad avocat "that being said". In fact, such expressions are absent from the Sanskrit text and it seems to have been added in error or despite the Sanskrit text. The same phrase occurs in sentence 17 of T254 where it does fit the context and T253 has Shuō shì yǔ yǐ 說是語已 "Having spoken these words.".

At the end, T253 has cíng 行 for both caryā and śikṣitavya. The last character of T254 substitutes xué 學 corresponding to √śikṣ "train", which better reflects the Sanskrit text. 

---

Sentence 16.

253	即時，世尊從 廣大甚深三摩地起，讚觀自在菩薩摩訶薩　言：
254	爾時，世尊從　三摩地安祥而起，讚觀世音自在菩薩摩訶薩言：
Skt.	atha khalu bhagavān tasmāt samādher vyutthāya āryāvalokiteśvarāya bodhisattvāya mahāsattvāya sādhukāram adāt |

At that time 即時，Bhagavan 世尊 from 從 the vast 廣大 and [very 甚] profound 深 samādhi 三摩地 he arose 起，praising 讚 Guānzìzài 觀自在 bodhisatva 菩薩 mahāsatva 摩訶薩 saying 言：

The main difference here is how they refer to the samādhi. T253: Guǎngdà shènshēn sānmódì 大甚深三摩地 (T253 S2 spelled samādhi as sānmèi 三昧).

T254 spells samādhi consistently but here does not refer to the samādhi by name, merely adding that the Buddha... "arose from samādhi, peaceful and composed" (從三摩地安祥而起).

On this occasion, T254 appears to follow the Sanskrit rather than T253. 

---

Sentence 17.

253	「善哉，善哉！善男子！如是，如是！如汝所說。甚深般若波羅蜜多行，應如是行。如是行時，一切如來皆悉隨喜。」
254	「善哉，善哉！善男子！如是，如是！如汝所說。甚深般若波羅蜜多行，應如是行。如是行時，一切如來悉皆隨喜。」
Skt.	sādhu sādhu kulaputra | evam etat kulaputra, evam etad gambhīrāyāṃ prajñāpāramitāyāṃ caryāṃ cartavyaṃ yathā tvayā nirdiṣṭam anumodyate tathāgatair arhadbhiḥ |

Good! 善哉，Good! 善哉！Kulaputra 善男子！It is so, 如是，it is so 如是 it is as you just said 如汝所說。Very 甚 profound 深 prajñāpāramitā 般若波羅蜜多 practice 行，[you] should 應 this way 如是 practice 行。in this way 如是 practising 行 when 時，all 一切 tathāgatas 如來 all 皆 completely 悉 rejoice. 隨喜。

This phrase sādhu sādhu kulaputra (善哉，善哉！善男子!) occurs several times in the later chapters of Aṣṭa (Vaidya 1960: 195, 238, 240, 247, 252, 255, 256, and 258). In the first occurrence (195), Māra is addressing a hypothetical bodhisatva (this event is absent from Conze's translation). The rest are in Chapters 30 and 31, on Sadāprarudita and Dharmodgata respectively. Thus phrase is absent from what is thought to be the core of the text and is mainly found in chapters widely considered to be later additions.

In the chapter on Sadāprarudita (Vaidya 238 ff.), the Bhagavan tells an avadāna story to Subhūti. In the story, the words sādhu sādhu kulaputra are spoken to Sadāprarudita by "a voice from the sky" (antarīkṣān nirghoṣaḥ). The second time (Vaidya 238) it is "the figure of a tathāgata" (tathāgata-vigrahaḥ) who appears and speaks.* The third time (247) it is Śakra (aka Indra) who addresses Sadāprarudita. A little later (still on 247), a merchant's daughter addresses Sadāprarudita, saying "You should come, Kulaputra" (ehi tvaṃ kulaputra...); at the end of Chp 30 (Vaidya 252), it is Sadāprarudita himself, recounting to Dharmodgata how the tathāgatas appeared to congratulate him.

* Note that Conze (1973: 279) translates tathāgata-vigrahaḥ as "a tathāgata-frame" which is both incomprehensible and wrong.

The pattern is similar in Pañc.

Here T253 and 254 are identical except for the transposition of two characters in the last phrase: 皆悉 "all completely" versus 悉皆 "completely all". The latter seems to be more idiomatic, and better expresses the sense of "completeness".

Again, I think it's unusual to see Guanyin addressed as kulaputra, even by the Buddha.

The latter part of the Sanskrit is a little different from the Chinese: "[it]... should be practised cartavyaṃ just as yathā by you tvayā has been indicated nirdiṣṭam, rejoiced in anumodyate by tathāgatas tathāgatair and arhats arhadbhiḥ."

---

Sentence 18.

253	爾時，世尊說是語已，具壽舍利弗，大喜充遍，觀自在菩薩摩訶薩亦大歡喜。時彼眾會天、人、阿修羅、乾闥婆，等，聞佛所說，皆大歡喜，信受奉行。
254	爾時，世尊如是說已，具壽舍利子，　　　觀世音自在菩薩，　　　　及彼眾會一切世間天、人、阿蘇囉、巘䭾嚩，等，聞佛所說，皆大歡喜，信受奉行。
Skt.	Idam avocad bhagavān, āttamanā āyuṣmān śāriputraḥ āryāvalokiteśvaraś bodhisattvo māhasattvas te bhikṣavas te ca bodhisattvā sā ca sarvāvatī pariṣat sadeva-mānuṣāsura-garuḍa-gandharvaś ca loko bhagavato bhāṣitam abhyanandann iti ||

Then indeed 爾時，Bhagavan 世尊 spoke 說 these 是 words 語 after 已，Elder 具壽 Śāriputra 舍利弗，[with] great 大 joy 喜 was entirely filled 充遍，Guānzìzài 觀自在 bodhisatva 菩薩 mahāsatva 摩訶薩 also 亦 great 大 delighted 歡喜。Then 時 that 彼 congregation 眾會 of devas 天、humans 人、asuras 阿修羅、gandarvas 乾闥婆 etc 等，heard 聞 [by] buddha 佛 what was said 所說，all 皆 greatly delighted 大歡喜，believed 信 accepted 受 respectfully 奉 put into practice 行。

The Chinese ending is characteristically Chinese and the structure of the Sanskrit is quite different here:

The Bhagavan said this: [And] overjoyed, Elder Śāriputra, Noble Avalokiteśvara bodhisatva mahāsatva, those bhikṣus and bodhisatvas, the whole gathering, and the world with its devas, humans, demons, garudas and gandharvas rejoiced in the words of the Bhagavan.

T254 omits dàxǐ chōng biàn 大喜充遍 "was entirely filled great joy". However, in T253 this appears to translate āttamanā "enraptured", so the omission from T254 seems to be a mistake.

Conclusion

Any conclusions about T253 and T254 need to be contextualised in terms of what we already know. T251 was assembled from some copied passages, some original composition, and a copied dhāraṇī. This composite text was later translated into Sanskrit. The Hṛd is a poor translation with numerous poor word and syntax choices, and a least one obvious calque of Chinese (Attwood 2021b). There are two main recensions of the extended text: (1) T252 and (2) the rest, i.e. T253, T254, 255, 257, the Sanskrit, and canonical Tibetan.

T252 (translated by Fǎyuè 法月 ca 741 CE) clearly had the same goal of completing the sutra by adding a suitable introduction and conclusion. This not only made the text seem like an Indian sutra, but also helped to clarify the otherwise ambiguous role of Guanyin. Apart from this, and the fact that T252 also incorporates the text of T251, it is in every other respect different from both T253 and T254. T252 is certainly not the same text as T253 but has to be seen as an independent attempt to supply the missing parts of a standard sutra.

From the present study, it seems that T254 is not an independent translation of the Heart Sutra. Rather it is a lightly edited version of T253, with only minor alterations. It's difficult to quantify, but I would guess that T254 is ~90% T253, and most of the variation is merely alternative spellings or synonyms.

The redacting of T253 to produce T254 appears to have occurred in Chinese. It is possible that a Sanskrit text was consulted, more so for the concluding section than the introduction. So for example, when T254 adds jùshòu 具壽 āyuṣman "Elder" to the name Śāriputra, this appears to reflect a Sanskrit practice that is not observed in Chinese translations. That said, T254 also seems to adopt readings that do not reflect the Sanskrit. Such as: adding hézhǎng gōngjìng 合掌恭敬 "joining hands respectfully" in Sentence 4; and omitting dàxǐ chōng biàn 大喜充遍 in Sentence 17.

When we look at T253, we see that it contains the entire text of T251, with a handful of minor alternations, even including a variant of a phrase from T251—lí zhū kǔ è 離諸苦厄—that is not found in any extant Sanskrit text.

T253 has several obvious mistakes that appear to be scribal errors: especially adding words and phrases where they do not belong. At present, I have no access to the witnesses that the editors of the Taishō used to create their edition. As such, I'm not sure whether such errors are found in the manuscripts/inscriptions or are the result of poor editing.

Neither T253 nor T254 really captures the Sanskrit of sentence 2. But then nor does any other version of the Heart Sutra (including the Tibetan). It's not clear how this happened.

The whole of T251 is integrated into T253 and this integration had to have happened in Chinese. This raises the question of whether T253 is a translation from a Sanskrit text at all. Perhaps it was composed in Chinese and the extended Sanskrit Heart Sutra came later? As with T254, T253 does not always follow the Sanskrit text.

At the very least Prajñā and Lìyán, like Fǎyuè 法月, salvaged the entirety of T251 in T253, seemingly without regard for any of the differences between T251 and their Sanskrit source. They all show the same blindness as Woncheuk who also supposedly had a Chinese source text.

It seems that the contribution of the "translators" only included the introduction and conclusion and a few minor changes to the rest of the text. This raises the question of whether T253 is an independent translation. For example, it is odd that T253 contains an expression hézhǎng gōngjìng 合掌恭敬 "joining hands respectfully" that has no counterpart in any known Sanskrit Heart Sutra, though it is a common enough expression.

Takeaways

1. T254 is a lightly redacted version of T253. Although most of the sentences have changes, they are mostly matters of alternate spelling or synonyms. Zhìhuì Lún 智慧輪 may have consulted a Sanskrit text in the process but all of T251 is preserved intact. This suggests that the redaction occurred in Chinese. Ergo, T254 is not a separate translation of the Heart Sutra. And is not really a "translation" at all.

2. T253 retains the entirety of T251 intact. Thus, even if it was initially a translation from Sanskrit, only the introduction and conclusion were translated afresh (if that). Given the integration of T251 within it, even if there was a Sanskrit text, T253 was likely substantially redacted in Chinese.

3. The extended Hṛd mirrors the redaction process of T253 in the sense that Hṛd_ext preserves the Hṛd_std text in its entirety; including all of the odd expressions, Chinese idioms, and calques. Thus it seems likely that Hṛd_ext was produced in much the same (Chinese) milieu as Hṛd_std and in much the same way.

~~oOo~~

Bibliography

Attwood (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

———. (2021) "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.

A list of all the differences

253 vs 254

Sentence 1

• 佛 vs 薄誐梵 (bhagavan)
• 耆闍崛山 vs 鷲峯山 (Gṛdhrakūṭa)
• 菩薩眾 vs 大菩薩眾 (cf 大比丘眾 in both)

S2

• 佛世尊 vs 世尊 (previously 薄誐梵)
• 時 vs 爾時
• 三昧 vs 三摩地 (cf S16 where both have 三摩地).
• T254 照見 added in error

S3

• 爾時 vs 時
• 觀自在 vs 觀世音自在
• 深 vs 甚深
• 般若波羅蜜多時 vs 般若波羅蜜多行時
• 皆空 vs 自性皆空
• 離諸苦厄 vs 254 omit (added S7)

S4

• 舍利弗 vs 具壽舍利子 (253 adds 具壽 in S18)
• 白觀自在 vs 白觀世音自在

S5

• 善男子 vs 聖者
• 行者 vs 行 (者 added in error)

S6

• 觀自在 vs 觀世音自在

S7

• 若 vs 若有
• 性空 vs 自性皆空
• 254 added 離諸苦厄。(from S3)

S8

• N/A

S9

• 是諸法空相 vs 是諸法性相空
• 不增不減 vs 不減不增

S10

• N/A

S11

• 究竟涅槃 vs 究竟寂然 (nirvāṇa)

S12

• 245 adds 現成正覺 (autocommentary?)

S13

• N/A

S14

• 櫱諦　櫱諦　波羅櫱諦　波羅僧櫱諦　菩提　娑(蘇紇反)婆訶
• 唵(引)　誐帝　誐帝　播(引)囉誐帝　播(引)囉散誐帝　冒(引)地　娑縛(二合)賀(引)

S15

• 舍利弗 vs 舍利子
• 應如是行 vs 應如是學 (improvement based on Sk)
• 253 adds 如是說已。 in error

S16

• 廣大甚深三摩地起 vs 三摩地安祥而起

253: vast 廣大 and profound 甚深 samādhi 三摩地 he arose 起
254: arose from samādhi 三摩地, peaceful 安祥 and composed 而起

• 觀自在 vs 觀世音自在

S17

• N/A

S18

• 語已 vs 說已
• 具壽舍利弗 vs 具壽舍利子 (253 adds 具壽)
• 大喜充遍 vs 254 omitted.
• 亦大歡喜。 vs 254 omitted
• 時彼眾會天 vs 及彼眾會一切世間天
• 阿修羅 vs 阿蘇囉 asura
• 乾闥婆 vs 巘䭾嚩 gandharva
• both have 信受奉行 with no Sk counterpart.

Revised Heart Sutra Editions and Translations.

After twelve years of research and a year-long editorial process, my revised editions of the Heart Sutra in Chinese and Sanskrit have now been published in Asian Literature and Translation along with translations (and copious notes) that reflect how I think it should be understood. I thank people in the article, but I also want to thank the editor, Ian Rapley, for his heroic patience with me.

Attwood, Jayarava. (2024). "Revised Editions of the Prajñāpāramitāhṛdaya and Bānrěbōluómìduō xīn jīng «般若波羅蜜多心經»." Asian Literature and Translation 11(1): 52-92. https://alt.cardiffuniversitypress.org/articles/10.18573/alt.63.

Since 2012, when I first noticed errors in Conze's Sanskrit, I have wanted to take this step of publishing revised editions. On the other hand, I have dreaded having to commit to a translation that would then be the subject of criticism from all sides. But the moment to take this step has arrived. So here are the "official" Jayarava texts and translations.

With help from many other scholars, I have at least identified the major issues with the existing editions and shown how I think they should be corrected. To the best of my ability, the Sanskrit text is now fully parseable as Sanskrit and, to the best of my knowledge, this is the first time in history that a Sanskrit Heart Sutra text can stake this claim. All the current Sanskrit editions have numerous grammatical errors and thus all existing translations based on those editions are also full of errors.

The revised Chinese text is less drastically changed and my revisions are mainly concerned with integrating insights from Huifeng (2014) into the structure of the text. Explanations for the changes can be found in the main article.

Please note that despite widespread beliefs to the contrary, the Heart Sutra is entirely in prose. So I avoid the faux-poetic arrangement of the text into disconnected short lines (that often ignore sentence breaks). I have added paragraphs to make reading/studying easier.

The Chinese Heart Sutra

Revised Chinese Text

般若波羅蜜多心經。

觀自在菩薩行深般若波羅蜜多時，照見五蘊皆空，度一切苦厄。

舍利子，色不異空，空不異色，色即是空，空即是色。受想行識亦復如是。

舍利子，是諸法空相不生不滅不垢不淨不增不減。

是故空中無色無受想行識，無眼耳鼻舌身意，無色聲香味觸法，無眼界乃至無意識界，無無明亦無無明盡乃至無老死亦無老死盡，無苦集滅道，無得無現觀，以無所得故。

菩提薩埵依般若波羅蜜多故心無罣礙，無罣礙故無有恐怖遠離顛倒夢想究竟涅槃。三世諸佛依般若波羅蜜多故得阿耨多羅三藐三菩提。

故知般若波羅蜜是大明呪無上明呪無等等明呪。能除一切苦真實不虛。

故說般若波羅蜜多咒即說呪曰　揭帝　揭帝　般羅揭帝　般羅僧揭帝　菩提　薩婆訶。

Translation

On the Essence of Perfect Insight.

Practising the deep perfection of insight, Guanyin bodhisattva observed the branches of experience, all were absent and he transcended all suffering.

Śāriputra, appearance is not different from absence and absence is not different from appearance; appearance is only absence and absence is only appearance. Valence, recognition, intention, and objectification are the same.

Śāriputra, all phenomena are characterised by absence that does not arise or cease, is not defiled or pure, and is not declining or growing.

In that state of absence—there is no appearance, valence, recognition, intention, or objectification; no eye, ear, nose, tongue, body, or mental sense; no appearance, sound, smell, taste, touch, or percepts; no eye-sphere etc up to no mental cognition sphere; no ignorance or cessation of ignorance etc up to no ageing and death or the cessation of ageing and death; no dissatisfaction, no origin, no cessation, no path; no attainment and no realisation—due to practising nonapprehension.

Since the bodhisatva relies on perfect insight their mind is not attached anywhere; being detached they are not afraid, transcend illusions and delusions, and attain final extinction. Relying on perfect insight, the buddhas of the three times attained full and perfect awakening.

Understand that perfect insight is great know-how, unexcelled know-how, and unequalled know-how. It can remove all suffering; it is true and not false.

Therefore recite the incantation of perfect insight, the incantation that goes: jiēdì jiēdì bānluójiēdì bānluósēngjiēdì pútí sàpóhē.

~o~

The Sanskrit Heart Sutra

Revised Sanskrit Text

āryāvalokiteśvaro bodhisatvo gambhīrāṃ prajñāpāramitācaryāṃ caramāṇo vyavalokayati sma pañcaskandhāṃs tāṃś ca svabhāvaśūnyān paśyati sma ||

iha śāriputra rūpam eva śūnyatā śūnyataiva rūpaṃ rūpān na pṛthak śūnyatā śūnyatāyā na pṛthag rūpaṃ | evam eva vedanā-saṃjñā-saṃskāra-vijñānaṃ ||

iha śāriputra sarvadharmāḥ śūnyatālakṣaṇā anutpannā aniruddhā amalā avimalā anūnā aparipūrṇāḥ ||

tasmāc chāriputra śūnyatāyāṃ na rūpaṃ na vedanā na saṃjñā na saṃskārāḥ na vijñānaṃ, na caksuḥ-śrotra-ghrāṇa-jihvā-kāya-manāṃsi na rūpa-śabda-gandha-rasa-spraṣṭavya-dharmāḥ na cakṣurdhātur yāvan na manovijñānadhātur nāvidyā nāvidyākṣayo yāvan na jarāmaraṇaṃ na jarāmaraṇakṣayo na duḥkha-samudaya-nirodha-mārgā na jñānaṃ na prāptiḥ ||

tasmāc chāriputra aprāptitvād bodhisatvaḥ prajñāpāramitām āśritya viharaty acittāvaraṇaś cittāvaraṇanāstitvād atrasto viparyāsātikrānto niṣṭhanirvāṇaḥ | tryadhvavyavasthitāḥ sarvabuddhāḥ prajñāpāramitām āśritya anuttarāṃ samyaksambodhim abhisambuddhāḥ ||

tasmāj jñātavyaḥ prajñāpāramitā mahāmantro mahāvidyāmantro ‘nuttaramantro ‘samasamamantraḥ sarvaduḥkhapraśamanaḥ satyam amithyātvāt ||

prajñāpāramitāyām ukto mantraḥ tadyathā | gate gate pāragate pārasaṃgate bodhi svāhā ||

iti prajñāpāramitāhṛdayaṃ samāptam ||

Translation

Practising the deep practice of perfect insight, noble Avalokiteshvara bodhisattva, examined the five branches of experience and saw that they lacked self-existence.

Obviously, Śāriputra, appearance only is absence; absence is only appearance. Absence is not different from appearance; appearance is not different from absence. The same applies to valence, recognition, intention, and objectification.

Obviously, Śāriputra all dharmas are characterised by absence that does not arise or cease, is not defiled or pure, and is not deficient or complete.

Therefore, Śāriputra, in that state of absence there is no appearance, valence, recognition, intention, or objectification; no eye, ear, nose, tongue, body, or mental sense; no appearance, sound, smell, taste, touch, or percepts; no eye-sphere, and so on up to, no mental-cognition-sphere; no ignorance or destruction of ignorance, and so on up to, no ageing and death or the destruction of ageing and death; no dissatisfaction, no origin, no cessation, no path; no knowledge and no attainment.

Therefore, Śāriputra, in the absence of attainment, the bodhisattva who is without mental obstructions dwells having relied on perfect insight, [and] being free of mental obstructions he is unafraid, has overcome delusions, and his extinction is complete. Having relied on perfect insight, all the buddhas appearing in the three times fully awakened to the unexcelled perfectly complete awakening.

Therefore, know that perfect insight is a great mantra, a great mantra of mastery, an unexcelled mantra, an unequalled mantra that allays all suffering and it is true and without wrongness.

Concerning perfect insight, a mantra goes like this: gate gate pāragate pārasaṃgate bodhi svāhā.

This Epitome of Insight is complete.

~~oOo~~

Guanyin Does Not Speak in the Heart Sutra

In this short essay, I will challenge a universal presupposition about the Heart Sutra, i.e. that the lines that appear to be spoken to Śāriputra in the core section are spoken by Guanyin. I will show that, by all the conventions of Buddhist literature, this is not true. Guanyin does not speak. This observation further undermines the already weakened historically dominant narrative about the Heart Sutra.

For some years, I have made a practice of reading every scholarly publication on this text (in English), as well as selected popular works. To the best of my knowledge, no modern scholars have previously noticed the absences I mark below. I think Chinese Buddhist commentators in the late seventh–early eighth centuries were aware of this. And there were subsequently efforts made to obscure this fact.

A Buddhist sutra is, above all else, a record of speech. In Buddhist texts, speech is almost always indirect speech and the forms of indicating who is speaking to whom are essential to understanding the text. Forms of present speech in Buddhist texts are highly formalized and standardized; to the point of being universal across genres and over time. And they are not complex. We can easily describe the main forms and note the Chinese reflexes of these forms. I will focus on how they appear in the early Prajñāpāramitā literature, if only because this is the appropriate context for thinking about the Heart Sutra.

To begin with, we expect to see the speaker “addressing” (āmantrayate) the audience. At the beginning of the Aṣṭasāhasrikā Prajñāpāramitā (Aṣṭa), for example:

Then the Blessed One addressed the senior Elder Subhūti...”
(tatra khalu bhagavān āyuṣmantaṃ subhūtiṃ sthaviram āmantrayate sma… Vaidya 1960 2).

Here āmantrayate sma is the "pleonastic past". Here adding sma to a present tense verb makes it a past tense, but is also used for the "present in the past" tense so commonly used in storytelling. In Kumārajīva’s translation—the Xiǎopǐn bānrě jīng «小品般若經» (T 227)—this becomes…

Then he Buddha addressed Subhūti
Ěr shí Fó gào Xūpútí 爾時佛告須菩提 (T 227: 8.537a29)

Here the verb is gào 告 "to address". Note that Kumārajīva omits the honorific ayuṣman "Elder" and Subhūti's monastic title sthavira "Senior [monk]". In Chinese we also sometimes see:

“Subhūti addressed the Buddha, saying…” 

Xūpútí bái Fó yán 爾時須菩提白佛言 (T 227, 8.537b06)

The character bái 白 is polysemic but here means “to make plain, to state clearly”, while yán 言 means “speak, talk”.

The extracts found in the Heart Sutra are from a version of the Pañcaviṃśātisāhasrikā Prajñāpāramitā (Pañc). An example of the same form from the Gilgit manuscript of Pañc:

The Bhagavan addressed Elder Śāradvatīputra...”
(bhagavān āyuṣmantaṃ śāradvatīputram āmantrayata... Zacchetti 2005: 375).

The use of āmantrayate (in various conjugations) usually marks the beginning of a passage of discourse, but within a given conversation, the speaker of individual passages is also marked. It is usual to spell this out laboriously, including the name and title of each participant. In the following passage from Chapter One of Aṣṭa, we find all of the most common forms of ongoing verbal address:

Then Elder Śāriputra said this to Elder Subhūti, “Elder Subhūti, does this mind that is a mind without mind actually exist?”

When that was said, Elder Subhūti said this to Elder Śāriputra, “Elder Śāriputra, concerning that which is without mind, is the existence of mindlessness known or apprehended?”

Śāriputra said, “Indeed not, Elder Subhūti”.

Atha khalv āyuṣmān śāriputra āyuṣmantaṃ subhūtim etad avocat - kiṃ punar āyuṣman subhūte asti tac cittaṃ yaccittamacittam?
Evam ukte āyuṣmān subhūtir āyuṣmantaṃ śāriputram etad avocat kiṃ punar āyuṣman śāriputra yā acittatā, tatra acittatāyām astitā vā nāstitā vā vidyate vā upalabhyate vā?
Śāriputra āha - na hy etad āyuṣman subhūte / (Vaidya 1960: 3).

The forms I wish to highlight are “said this” (etad avocat), “when this was said” (evam ukte), and “said” (āha). Both avocat and ukte derive from √vac “speak”, while the āha is from the defective verb √ah “say”. The same forms are used in the same way throughout the Pāli Suttapiṭaka also. An electronic search of the Chaṭṭa Saṅgāyana Tipiṭaka (4.1) suggests that in the four main Nikāyas: āmantesi occurs about 590 times, etadavoca occurs over 2200 times; evaṃ vutte occurs some 530 times; while āha occurs about 100 times. A few other such forms are used, but these are by far the most common. 

When a person is addressed in Sanskrit or Pāli, their name occurs in the vocative case, e.g. “O Subhūti” (subhūte). In the CBETA version of the Taishō edition of the Chinese Tripiṭaka, the “vocative case” is represented by "!" following the speaker’s name. 

Note the repetitive use of the title “Elder” (āyuṣman) most of the time. Given that both characters use the title, they appear to be social equals. I speculate that āyuṣman is omitted precisely when Śāriputra acknowledges Subhūti’s superior insight.

These forms are largely preserved in Chinese translations. Turning to Huifeng’s (2017: 205) translation of this same passage in the Xiǎopǐn:

Thereupon, Śāriputra said to Subhūti: “Does this mind which is mindless exist?”

Subhti said to Śāriputra: “That mind which is mindless, is it apprehendable as either existing or not existing?”

Śāriputra said: “Indeed not!”

Here, two different words are translated as “said”: yǔ 語 (here translating avocat) and yán 言 (translating āha).

None of these conventions for indicating that someone is speaking or for who is addressing whom occurs in the Heart Sutra. The only indication we get that some parts of the text are indirect speech is the use of the vocative śāriputre “O Śāriputra” (Shèlìzi 舍利子！). This is how we know that Śāriputra is being addressed. The Heart Sutra does not say who is speaking and everyone assumes that it is Guanyin. 

The passage in which Śāriputra is addressed has been traced to the Large Sutra (see: T 223, 8.223.a13-a20 and Zacchetti 2005: 393). When we read the passage in this context, the lines are spoken to Śāriputra not by Guanyin, who has no speaking part in any Prajñāpramitā text, but by the Buddha. Interestingly, Woncheuk’s (613-696 CE) commentary (T 1711) appears to take the Buddha to be speaking as well.

Question: [Since] this [teaching of] prajñāpāramitā is the dharma for the bodhisattva, why does the World Honored One preach not to the bodhisattva but to Śāriputra? (Hyun Choo 2006: 149)

and

Therefore, in the [Heart Sūtra], the Buddha preached to Śāriputra and intended to lead the Hīnayāna to the Mahāyāna as well. (Hyun Choo 2006: 149)

It seems to me that Woncheuk could only have deduced this by looking at the source of the passage, i.e. Pañc.

Conclusions

In this brief essay, I have tried to show that the universal view that Guanyin is speaking in the Heart Sutra is based on presupposition and unexamined assumptions. I did this by outlining Buddhist conventions for expressing who is speaking to whom. I argued that such conventional expressions are universal in Buddhist texts (in Pāli and Sanskrit) and that such conventions are absent from the Heart Sutra. All indications of who is speaking have been omitted. Taking the Heart Sutra at face value, no one is speaking.

We can explain this by pointing out that the Heart Sutra is not, in fact, a sutra. It is not a record of speech, rather it's a compilation of ideas and extracts from existing speeches. This much was obvious to the earliest commentators, though subsequently forgotten. The lines in their original context were spoken by the Buddha, as accurately reflected in the commentary by Woncheuk. 

The (now disproved) "fact" that Guanyin was speaking has always been a problem for scholars since Guanyin plays no active role in any non-Tantric Prajñāpāramitā text (and the Tantric Prajñāpāramitā texts are more Tantra than Prajñāpāramitā). Various unsatisfactory explanations have been advanced (I've made several previous attempts to explain), but they have always been ad hoc or post hoc rationalisations, rather than real explanations (they all amount to hand-waving).

The most striking attempts to make sense of this situation are the two recensions of the extended Heart Sutra text (1. T 252; 2. T 253 and all rest), probably composed in the early eighth century, possibly in the oasis town Dūnhuáng 敦煌, on the edge of the Gobi Desert (from where it was transmitted to Tibet). 

In the extended texts, the two redactors have attempted to better integrate Guanyin into the narrative, by expanding the first paragraph. However, both recensions retained the rest of the standard Heart Sutra unaltered and in this part of the text, the verbs of speech are still absent. This means that despite more clearly articulating the reason for the presence of Guanyin, neither of the extended texts addresses the problem. Guanyin's presence seems more natural, but he/she still does not speak the lines directed to Śāriputra in the "core section".

The fact is that, in the Heart Sutra, Guanyin does not speak. Guanyin is invoked and then plays no further role in the text. In their original context, the lines are spoken by the Buddha. At this point, we can say that more or less all of the historical dominant narrative about the Heart Sutra is a post hoc invention. 

~~oOo~~

Bibliography

Huifeng. (2017). “An Annotated English Translation of Kumārajīva’s Xiǎopǐn Prajñāpāramitā Sūtra.” Asian Literature and Translation 4(1): 187-236.

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 and Culture 6: 121-205.

Vaidya, P.L.(1960) Aṣṭasāhasrikā Prajñāpāramitā Sūtra. Darbhanga: The Mithila Institute. (Via the Gretil Archive, 2014. Including Karashima, S. (2013) On the "Missing" Portion in the Aṣṭasāhasrikā Prajñāpāramitā. ARIRIAB, 16: 189-192. Accessed 6 Feb 2024)

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 for Advanced Buddhology, Soka University.

The Extended Heart Sutra: Overview

This post is a first attempt to sum up my close reading of the extended Heart Sutra that spanned eight posts, over which I worked my way through the distinctive features of the different versions of the extended Heart Sutra, noting down and trying to explain differences, omissions, additions, and errors. One of the principle unanswered questions about the extended Heart Sutra is: what language was it composed in? Which raises methodological questions about how would we assess this. We also want to know when the extended text came into existence and who was responsible for it. 

I should stress that I have only consulted versions of the text in Sanskrit, Middle Chinese, and canonical Tibetan. At this point I have not consulted the secondary literature. As such, my conclusions are limited to philological points, except where my more broadly based historical research on the standard text clearly applies to the extended text as well. On the other hand, I'm not aware of any existing philological studies of the extended text, except for Silk's critical edition of the Tibetan variants. And Silk's work is a critical edition of the Tibetan, not a critical history. There are, of course, religious studies of the text in Tibetan by the Dalai Lama and others, but their methodology is very different from mine. Religious commentaries seek to justify beliefs (i.e. emotions about ideas). Donald Lopez's two books on the commentaries preserved in Tibetan by "Indian" authors are also on the long version, but Lopez takes the tradition on its own terms. The books offer a reading of the tradition, but little or no insight into the kinds of problems that interest me here (i.e. how such a tradition became established).

Of particular interest is the Manuscript kept at the Hasedera Temple, which was the foundation for both Müller's edition, the edition by Vaidya which is now widely available on the internet (e.g. DBSC), and Conze's critical edition. As far as I can tell, the Hasedera manuscript has not been digitised or published in any form since Müller (1884). Müller himself was working from hand copies made for him by Japanese Buddhists and everyone since then is working from Müller's edition (although this is not always acknowledged - by Conze, for example). I could not find any information about where the mss is now.

The extended Heart Sutra has passages added at the beginning and the end of the standard text. Pragmatically, what the extensions do is tell the story, in formulaic terms, of how Avalokiteśvara came to preach the Heart Sutra to Śāriputra and, almost as importantly, how it was received. Such extensions were presumably the result of internal pressure to make the Heart Sutra more authentic, since it lacked all of the features that are traditionally associated with authenticity. It is sometimes suggested that this was an Indian concern, but the Chinese had exactly the same concerns and were more concerned, because so many indigenous Buddhists texts were in circulation. There is little or no evidence that the Heart Sutra was even known in India.

The extensions infer that the narrator is Ānanda (the "me" in "thus by me heard" evaṃ mayā śrutam); tell us the occasion and place of the discourse, that the Buddha is present, and that the speaker speaks with the anubhāva or empowerment of the Buddha. At the end, the interlocutor praises the teaching, and the Buddha praises the speaker. And in Chinese the audience promise to faithfully practice the teaching. Without these features the Heart Sutra was not a sūtra. The fact that the Heart Sutra was considered to be a sūtra was down to the forging of a Sanskrit text and document, and the attribution of the Chinese to Xuanzang as a translation (See Attwood 2019, 2020).

I refer to the unknown composer of the original digest text (chāo jīng 抄經) we call Xīnjīng 心經 as the Author. Although we think the Author was Xuanzang, this is, strictly speaking, unproven and it is possible that it was one or more other people. I refer to the person who translated the text into Sanskrit as the Translator. And now I have to introduce the idea of the person who redacted the text to produce the extended recension of the text, whom I will call the Redactor. 

It is also useful to make a threefold distinction between work, text, and document. As I said in Heart Sutra: Work, Text, Document (2019):

In his recent article, Silk (2015:205-6) draws out a threefold distinction first made by Chaim Milikowsky. First we have the Work, which is the author's or editor's product. This may only exist conceptually and never have been committed to words. Or the author may have attempted to put it in words and be more or less satisfied that the result, but still consider this as inferior to their conception of the Work. A presentation of the Work in words is a Text. A single Work may generate multiple Texts; for example, one story that is told many times, but with minor differences each time. No single Text is the "original" in this case, because the Text is not the Work. Lastly a Document is some physical instantiation of a Text. Typically, in studying Buddhist manuscript cultures, we are faced with multiple Documents representing multiple Texts. This is certainly in the case of the Heart Sutra.

I think we have clear evidence of their being at least two Redactors: one responsible for Recension Two (T 252) and one for Recension One (all other texts) (I started this convention but I wish I'd used the numbers the other way around since T 252 looks like the first recension to me). Where necessary, I will refer to Redactor One and Redactor Two. Redactor One might have been Fǎyuè 法月, who is credited with the translation of T 252. Redactor Two might have been possibly Prajñā (Bānruò 般若) and his Chinese collaborator Lìyán 利言. At least one scholar has argued that Prajñā, in fact, composed the works for which he was credited with being the translator.

The headings I used were:

• Sources
• Nidāna (paragraph D)
• The Buddha's Samādhi (Para D)
• Enter Avalokiteśvara (Para E)
• Śāriputra's Question (Para F & G)
• Avalokiteśvara Preaches (Para H & I)
• The Buddha's Endorsement (Para T & U)
• Rejoicing (Para V & W).

And I will continue with this structure.

Sources

I worked with eight sources: T 252, 253, 254, 255, 257; Conze's Sanskrit edition; and the two Tibetan recensions described by Silk in his 1994 critical edition. There is still some work to do on some of these texts.

As Silk (1994) pointed out, Conze's Sanskrit edition is chaotic and unreliable. None of the Sanskrit sources is in pristine condition and some of them are very badly corrupted. However, even though we need to have a better edition, my experience of studying British Library Manuscript EAP676/2/5 suggests the chances are that examining more Nepalese manuscripts would be fruitless. Describing EAP676/2/5 required 142 footnotes on errors, omissions, and additions for a 280 word text. Even the best of the Nepalese mss is significantly degraded by repeated copying without error checking.

I would like to collect copies of all the canonical Chinese versions but don't have access to the necessary resources. Also, we lack a clear understanding of the attributions and dates traditionally assigned to the Chinese translations. This is important, given the chequered history of the Heart Sutra, in which none of the attributions or dates turns out to be reliable. I would place this above a new Sanskrit edition in importance.

The Heart Sutra may be an interesting candidate for the application of computerised phylogenetic techniques for analysing relatedness, though I'm not sure how such techniques cope with multiple languages. And I don't have the knowledge required to do it. I do have a preliminary stemma diagram prepared manually based on my own work and preliminary comments about the Dunhuang manuscripts by Ben Nourse.

NB: R1 & R2 are reversed here because I made this diagram for an article I'm preparing that covers this territory for publication.

The node labelled "Hṛdaya with added negations" is a notional intermediary. It may not have existed as a separate text and is not extant. But it does seem that most of the extended texts don't have the extra negations and neither does the Xīnjīng. Similarly, I don't know that there was a prototype of the Sanskrit extended text distinct from some actual document that formed the template for the rest. But in terms of diagramming the history of the text it's useful to posit these notional versions of the texts.

Nidāna

The nidāna is one of the most important features that mark the authentic sūtra and that is missing from the standard Heart Sutra. The nidāna begins with the immortal words "Thus have I heard" (Skt evaṃ mayā śrutam; Ch. Rúshì wǒwén 如是我聞). The phrase is traditionally understood to indicate that the narrator is Elder Ānanda, even for Mahāyāna sūtras. The nidāna then continues in the usual fashion to name the place at which the discourse was delivered and who was present: "At one time the Bhagavan was staying... (Skt ekasmin samaye bhagavān... viharati sma ) together with a great congregation of bhikṣus and a great congregation of bodhisatvas (mahatā bhikṣusaṃghena sārdhaṃ mahatā ca bodhisattvasaṃghena). The term bodhisattvasaṃgha (púsà zhòng 菩薩眾) is rare but not diagnostic for my purposes. 

Although the nidāna is considered evidence of authenticity by early medieval Buddhists, we also know that many texts lacked a nidāna. For example the Pāli version of the Burden Sūtra (SN 22:22;) lacks a nidāna while the Chinese versions both have one: Samyuktāgama 73 (T 2.19a15-19b1); Ekottarāgama 25.4 (T 2.631c11-632a5). Greg Schopen has detailed how the Mūlasarvāstivāda Vinaya preserved in Chinese contains a set of guidelines for adding a nidāna to a sūtra where one is missing (Schopen 2004). This suggests that the practice was routine. The nidāna of the extended Heart Sutra shows variations that can only be deliberate redactions (as opposed to accidents).

There is considerable variation in the nidāna of the extended Heart Sutra documents:

T 253		T 257		T 252
如是我聞：一時佛在王舍城耆闍崛山中，與大比丘眾及菩薩眾俱。		如是我聞：一時，世尊在王舍城鷲峯山中，與大苾芻眾千二百五十人俱，并諸菩薩摩訶薩眾而共圍繞。		如是我聞：一時佛在王舍大城靈鷲山中，與大比丘眾滿百千人，菩薩摩訶薩七萬七千人俱，其名曰觀世音菩薩文殊師利菩薩彌勒菩薩等，以為上首。皆得三昧總持，住不思議解脫。
Thus have I heard. At one time, the Buddha (佛) was in Rājagṛha on Vulture Peak Mountain, along with a great congregation of bhikṣus together with a congregation of bodhisatvas.		Thus have I heard. At one time, the Bhagavan (世尊) was in Rājagṛha on Vulture Peak Mountain, along with a great congregation of 1250 bhikṣus together with many bodhisatva mahāsatvas and together they circumambulated.		Thus I have heard: At one time, the Buddha was in Rājagṛha on the mountain of Gṛdhrakūṭa, together with a great bhikṣusaṃgha of 100,000 and 77,000 bodhisatva mahāsatvas in all, those named Avalokiteśvara Bodhisatva, Mañjuśrī Bodhisatva, Maitreya Bodhisatva, were the leaders.* All had attained the samādhi of always remembering, and abided in inconceivable liberation. *Possibly Maitreya was "at the head" (yǐ wéi shàng shǒu 以為上首), cf. Conze's translation of Pañcaviṃśati: "and Maitreya the Bodhisattva, the great being, at the head of many hundred thousands of niyutas of kotis of Bodhisatvas." (1975: 38).

What stands out is that, despite being also set in Rājagṛha and, despite being constrained by the conventions of Buddhist sūtra composition, T 252 is considerably different from the other versions of the extended sūtra (whether in Sanskrit, Chinese, or Tibetan) and continues to diverge. Any introduction that was added was bound to include certain phrases and references to the Buddha (by convention), while Avalokiteśvara and Śāriputra appear in the standard text and must be the main protagonists. Apart from these constraints, T 252 stands alone in giving exaggerated numbers of participants and naming bodhisatvas. By comparison the latest translation has added some detail to what is essential the same passage. As we will see in the next section, the story of T 252 takes a distinct path to get to the standard Heart Sutra.

The Buddha's Samādhi

Conze's Sanskrit text describes how the Buddha taught a dharma teaching (dharmaparyāyaṃ bhāṣitvā) named “profound illumination” (gambhīrāvabhāsaṃ nāma) and then entered a meditative state (samādhiṃ samāpannaḥ). The Hasedera Manuscript (which has strongly influenced other Sanskrit editions) truncates this to "At that time the Bhagavan entered a samādhi named deep understanding"* (tena khalu samayena bhagavān gambhīrāvasaṃbodhaṃ nāma samādhiṃ samāpannaḥ).

* reading avabodha for avasambodha which is not in any of my dictionaries.

It appears to be this text that is translated in T 253, 254.

T 253: "At that time, the Buddha, the Bhagavān, entered the samādhi named Vast and Extremely Profound" (時佛世尊即入三昧，名廣大甚深。).

The Tibetan recensions appear to have a hybrid, e.g. TibA.

de'i tshe bcom ldan 'das zab mo snang ba zhes bya ba chos kyi rnam grangs kyi ting nge 'dzin la snyoms par bzugs so //

and at that time the blessed one was entered into the concentration of the preaching of the Dharma called “profound illumination”

Note that TibB has what appears to be an eyeskip error at this point. The phrase chos kyi rnam grangs (dharmaparyāya) has been shuffled forward into the middle of the phrase zab mo snang ba. The result is comprehensible but less consistent with the other texts.

The word dharmaparyāya is present in Conze sources Na, Nb, and Ne (with a variant reading); Ca, Cd, and Ce which makes it a minority reading. However the Tibetan equivalent, chos kyi rnam grangs, is present in both TibA and TibB. A translation of the word is also present in T255 (fǎ zhī yì mén 法之異門) and 257 (xuān shuō zhèng fǎ 正法宣說). Thus it appears to belong.

T 253 and 254 call the samādhi "extensive" guǎng dà 廣大 (mahat, vaipulya) and "profound" shèn shēn 甚深 (gambhīra). T 255 and T 257 do a better job of conveying the Sanskrit, i.e. "profound (shèn shēn 甚深) illumination (míng liǎo 明了)" and "profound (shèn shēn 甚深) illumination (guāng míng 光明)".

There is some confusion here. It is possible that an earlier version represented in the translations T 253 and 254 left off dharmaparyāya and it was added later. But it is equally possible that it was simply left out of the early Chinese translations. 

Note that the narrative in T252 is again very different. Rather than the Buddha' entering a samādhi we find:

皆得三昧總持，住不思議解脫。
All had attained the samādhi of always remembering, and abided in inconceivable liberation.

It is de rigueur to mention samādhi at this point in a Mahāyāna sūtra. In Recension One the Buddha entering samādhi is a required detail to explain why he does not speak in the core part of the text (the standard Heart Sutra). But the story arc in T 252 is different. The mention of samādhi is really a final detail in the description of the audience. However, both recensions use this moment to introduce the main protagonist of the story.

Enter Avalokiteśvara

One of the primary purposes of the introductory extension is to introduce the only actor with a speaking part, i.e. Avalokitśvara bodhisatva. There is a distinction to be made here. When Śāripūtra is mentioned in the introduction he is called āyuṣman śāriputra or "Elder Śāriputra". The word āyuṣman literally means "one who possesses life" or something like "one who has lived". The word is invariably placed before his name as a title. When the word ārya "noble" is used in conjunction with the name Avalokiteśvara its occupies the same position. By contrast, bodhisatva is invariably placed after the name of the figure and thus is not a title or epithet but an adjective or predicate. Bodhisatva means "one whose nature or essence" (satva) is awakening or enlightenment (bodhi). The spelling is consistently satva in Prajñāpāramitā manuscripts and is a distinctive Buddhist variant that is usually obscured by editors who tacitly revert to the classical spelling sattva as though it were an error. Elder Śāriputra is a bhikṣu. Noble Avalokiteśvara is a bodhisatva.

James Apple notes that prefixing ārya to a name is a relatively late practice:

"The prefix ārya appears in the opening salutation of Indian and Tibetan sūtras and is a late Indian Practice that begins to occur, as far as I can currently tell, in the works of [6th Century Mādhyamika] Bhāvaviveka." (Apple 2015: 4-5 n.5).

At this point in Recension One, Avalokiteśvara is practising the deep practice of Prajñāpāramitā. This is what gives him the essential insight into the nature of experience that he then communicates to Śāriputra, using the words from the opening paragraph of the standard sūtra; in other words he sees that the five skandhas are all absent. Except that in Sanskrit he sees them as svabhāvaśūnya "lacking a essence". This interpolation of svabhāvaśūnya is discussed further below.

We can infer that Avalokiteśvara is practising the yoga of nonapprehension (anupalambha-yoga). This involves using meditative techniques to cut oneself off from the world of sensory experience by deliberately and systematically withdrawing attention from the senses. The result is the cessation (nirodha) of sensory experience and this leaves the meditator dwelling in the absence of sense experience (śūnyatā). I refer to this state of the absence of sensory experience as Absence (capitalised). In English we more often encounter the word translated as "emptiness" and emptiness is frequently reified as "reality" or, worse it is hyper-reified as "ultimate reality". The result is the received strain of nihilistic anti-rational metaphysics of Madhyamaka.

Using my hermeneutic, we can look at Absence (śūnyatā) from two angles. Firstly, in Absence (śūnyatāyām) there are no mental events rising into awareness and passing away. There is some form of conscious awareness but it has no object (in philosophical jargon it is not intentional). Hence the Heart Sutra goes on to say that "in emptiness there is no colour, sound, smell, etc ... through the yoga of nonapprehension". In Buddhist jargon: in Absence dharmas do not arise or cease. If we take this state to be reality then we may be tempted to say that no dharmas ever arise in reality. But this causes us problems because when we are not in this state, dharmas constantly arise and cease (except in deep sleep).

The second angle is to say that although all mental states arise in dependence on conditions, this state is only attained when all conditions for the arising of mental states have ceased (though withdrawing attention). Absence itself then is not a conditioned mental state. It is without conditions. A conditioned mental state is saṃskṛta and thus śūnyatā is asaṃskṛta. That is to say śūnyatā is a synonym of nirvāṇa "extinction".

Something to keep in mind is that where the Heart Sutra says Shì zhū fǎ kōng xiāng, bù shēng bù miè 是諸 法空相 ，不生不滅 "All dharmas are marked with emptiness; not born, not dying" this is based on a mistranslation by Kumārajīva. The original Sanskrit says: yā śūnyatā na sā utpadyate, no nirudyate "that Absence does not arise, does not cease". In other words the original subject of this passage was not all dharmas, but the absence of all dharmas. Absence does not arise because it occurs only when all sensory experience has ceased. 

However, the story in Rencension Two is very different. Here, Avalokiteśvara has a dialogue with the Buddha. He announces his desire to give a teaching on the Heart of Prajñāpāramitā (般若波羅蜜多心), the Bhagavan praises this urge to teach and then takes a back seat, but not in samādhi. Now Avalokiteśvara enters a samādhi in which he "sees" the absence of essence in the five skandhas (照見五蘊自性皆空。). And then he addresses Śāriputra.

Quirks in this part of the text

Influence

T 253 and 254 have an extra phrase: lí zhū kǔ è 離諸苦厄 "apart/removed from all suffering and misery" means much the same as dù yīqiè kǔ è 度一切苦厄 "transcended all suffering and misery" in standard text. This phrase is absent from all extant Sanskrit manuscripts and the Tibetan canonical versions. It is present in the Fangshan Stele (the earliest physical evidence of the Heart Sutra) so cannot be a late edition. So it seems likely that it was simply left out of the Sanskrit translation, but whoever translated T 253 and T 254 from Sanskrit added it (back) in.

Look/See

Zhàojiàn 照見 doesn't really make sense as two standalone characters: zhào 照 means "shining, radiant; illuminate, make visible; reflection" and jiàn 見 is the usual verb "to see". This has led to some very odd translations such as "illuminatingly sees", where zhào 照 functions as an adverb. This is far from satisfactory and there is no consensus on how to translate it. If the two characters are a binomial reflecting the Sanskrit vyava√lok then we still have a problem because the conclusion doesn't fit the premise.

I would like to propose a solution (suggested by the Sanskrit translation), which is that zhàojiàn wǔyùn jiē kōng 照見五蘊皆空 is, in fact, two phrases: zhàojiàn wǔyùn 照見五蘊 "[he] examined the five skandhas" and jiē kōng 皆空 "all void". The latter is minimal and requires us to interpolate much that we would naturally spell out in English or Sanskrit. Firstly, it is implied that the skandhas were void or absent. Secondly, the verb "to see" is implied by the context of the initial zhàojiàn 照見. If one looks, then one sees something. That something was seen as a result of looking can remain implicit in written medieval Chinese. Knowing that Sanskrit did not have this kind of flexibility, the Translator has to specify both the action "he examined" (vyavalokayati sma) and the result "he saw" (paśyati sma).

Note that the Sanskrit translation of the standard text substitutes what should be śūnyatā with svabhāva-śūnya. I will say more about this below.

T 255 complicates matters somewhat: guānchá zhàojiàn wǔyùn tǐ xìng xījiē shì kōng 觀察照見五蘊體性悉皆是空. Guānchá 觀察 would seem to be a synonym of zhàojiàn 照見, while jiē kōng 皆空 is expanded out to tǐ xìng xījiē shì kōng 體性悉皆是空 "self-nature, without exception, is absent". Here tǐ xìng 體性 conveys svabhāva. T 254 and T 257 also pick up the Sanskrit idea that the five skandhas are void of svabhāva. It's possible T 255 intended this but a character got dropped because the text only has characteristic (xìng 性) but there are no text critical notes in the Taishō so this is speculative. T 253 has the same text as the standard version, i.e. jiē kōng 皆空.

Practising

The present participle caramāṇa is not used in either Aṣṭasāhasrikā or Pañcaviṃśatisāhasrikā, probably because it is the ātmanepada ("middle voice") form of the present participle and √car is usually parasmaipada (indicative), so that the expected present participle is carant (nominative singular caran) and this is the form we find throughout the two main Prajñāpāramitā sūtras. Although note that caramāṇa is found in the Ratnaguṇasamcayagāthā (not translated into Chinese till the 11th Century), the Mahāvastu, and in many Pāli suttas (although the use of the ātmanepada conjugations generally in Pāli is minimal).

svabhāvaśūnyan

We can now come back to the issue of the substitution of svabhāvaśūnyan for śūnyatā. In fact, the phrase svabhāvaśūnyan is absent from the Pañcaviṃśatisāhasrikā as a whole. If one is in the śūnyatā samādhi one does not (cannot) see skandhas because they are absent (śūnya). They cannot be examined. This is explicit in the core passage of the Heart Sutra:

是故，空中—無 色、無受、想、行、識... —以無所得故。
Shì gù, kōng zhōng—wú sè, wú shòu, xiǎng, xíng, shì... —yǐwúsuǒdégù
Therefore: in emptiness—there is no form; no feeling, no thought, volition, awareness... —through the yoga of nonapprehension

In the list of eighteen kinds of śūnyatā we find the following terms

16. Absence of non-being (abhāva-śūnyatā).
17. Absence of being (svabhāva-śūnyatā).
18. Absence of non-being and of being (abhāvasvabhāva-śūnyatā).

Here svabhāva is not being used in the Madhyamaka sense but as a contrast to abhāva "non-being". This is to say that svabhāva appears to be used in the sense of sabhāva "with bhāva" and abhāva means "without bhāva". In another list of four kinds of śūnyatā, svabhāva "self-being" is contrasted with parabhāva, literally "other-being"

But Pañcaviṃśatisāhasrikā is constantly telling us not to take these as metaphysics; for example:

abhāvasvabhāvaśūnyatā abhāvasvabhāvaśūnyatāyāṃ na saṃvidyate nopalabhyate, abhāvasvabhāvaśūnyatāpi yāvad adhyātmaśūnyatāyāṃ na saṃvidyate nopalabhyate. (PvsP1-2: 146)
"... the absence of being and non-being does not perceive or apprehend the absence of being and non-being nor are the other kinds of absence perceived or apprehended."

In other words, whether or not something exists cannot be distinguished when all sense experience has stopped because all information about the world outside brain has stopped registering. Although the verbs saṃvidyate and upalabhyate are routinely given an ontological gloss in translation (especially by Conze) - the former, in particular, is often translated as "exists" - both are (based on etymology) epistemic terms related to the use of one's sensory apparatus. The reason that something is na saṃvidyate "not discovered" or nopalabhyate "not apprehended" in this context is that sense experience has ceased, not because they don't exist. But on the other hand, if they did not exist, how would we apprehend them in the first place? 

Similarly with respect to the list of four kinds of absence, the Upadeśa (T 1509) says (using Lamotte's reconstruction of the Sanskrit):

• bhāvo bhāvena śūnyaḥ "being is absent from being"
• abhāvo ‘bhāvena śūnyaḥ " non-being is absent from non-being"
• svahāvaḥ svabhāvena śūnyaḥ "one's own being is absent from one's own being"
• parabhāvaḥ parabhāvena śūnyaḥ "the being of others is absent from the being of others"

It is apparent that by introducing an element of Madhyamaka metaphysics into the Heart Sutra, the Translator (Ch → Skt) has hit a bum note. This is an important consideration since it is normally assumed that Prajñāpāramitā and Madhyamaka are intimately associated. In fact, the two sects have very different ideas and there is almost no crossover.

This also means that the translation of 照見五蘊皆空 as vyavalokayati sma pañcaskandhāṃs tāmś ca svabhāvaśūnyān paśyati sma is incorrect. The first clause seems about right but what Avalokiteśvara saw was more like tāṃś ca sarvaśūnyān paśyati sma "... and he saw that they were all absent". This is consistent with the practice of Prajñāpāramitā being synonymous with the yoga of nonapprehension (anupalambhayoga) and both with the cessation of sense experience. 

Śāriputra's Question

The puerile attitudes towards arhats in some Mahāyāna sūtras are generally absent in Prajñāpāramitā. The principle representative of the Prajñāpāramitā perspective is Elder Subhūti. Elder Śāriputra is deeply interested in Prajñāpāramitā and raises the kinds of questions that we might expect someone trained in another Buddhist tradition to ask. In the Large Sūtra, just before the teaching from which the core passage is drawn, Śāriputra asks the Buddha:

kathaṃ yujyamāno bhagavan bodhisatvo mahāsatvaḥ prajñāpāramitāyāṃ yukta iti vaktavyaḥ. (PvsP1-1: 61-2)
Engaging in what way, Bhagavan, is a bodhisatva mahāsatva to be called 'engaged in prajñāpāramitā?'

This generic or abstract use of the word bodhisatva is typical of the Large Sutra. In creating the Heart Sutra, the Author has placed the teaching in the mouth of Avalokiteśvara who, in Chinese Buddhism, is a cult figure particularly associated with saving people from disasters. Hence, in the introductory extension, rather than the usual abstract questions, Śāriputra addresses his question to Avalokiteśvara whose answer then segues into the core passage of the standard Heart Sutra.

The question Śārputra asks in the Sanskrit Heart Sutra is

yaḥ kaścit kulaputro vā kuladuhitā vā asyāṃ gambhīrāyāṃ prajñāpāramitāyāṃ caryāṃ cartukāmas tena kathaṃ śikṣitavyaṃ?
That son or daughter of the community desiring to practice this deep perfection of gnosis, how should they train?

As we will see, one of the Chinese texts mistakes kulaputra for a vocative, i.e. that Śāriputra is addressing Avalokiteśvara as kulaputra. Also note that in Sanskrit the question is framed with both male and female communards (kulaputro vā kuladuhitā vā)

Again Recension Two has a different approach. Avalokiteśvara has already announced his intention to speak. In this passage he tells Śāriputra to get ready, and Śāriputra says “Indeed, Great Purifier, I am ready, preach it. Now is the right time" (唯，大淨者！願為說之。今正是時。) .

Names

We begin to notice that the form of names used in the texts is variable. The conventions established by Kumārajīva were:

• Avalokitasvara: Guānshìyīn 觀世音 
• Śāriputra: Shèlìfú 舍利弗,

Kumārajīva's translation is based on the older form of the name, i.e. Avalokitasvara (see Nattier 2007 and Karashima 2016). Using the new form, Avalokiteśvara, Xuanzang translated Guānzìzài 觀自在. He also modified Śāriputra: Shèlìzi 舍利子. Even so, Guānshìyīn 觀世音 or Guānyīn 觀音 are still the most common forms of the name in China, just as Kumārajīva's translations are still in use and were not superseded by Xuanzang's.

T 252 and T 253 have 舍利弗, the Kumārajīva spelling of Śāriputra, while T 257 has the Xuanzang spelling, 舍利子. All Chinese texts have the Xuanzang spelling of Avalokiteśvara 觀自在, except for T 254 which has Guānshìyīnzìzài 觀世音自在 which is a hybrid of Kumārajīva's 觀世音 and Xuanzang's 觀自在.

In the Tibetan texts we see some variation with the name Śāriputra, i.e. TibA shā ri'i bus; TibB shā ra dwa ti'i bus. The form Śāradvatīputra does occur in many Mahāyāna texts including the Gilgit manuscript of the Large Sutra. I'm not sure why.

Kulaputra

All texts use the term kula-putra, "a son of the community" (Tib. rigs kyi bu; Ch. shàn nán zǐ 善男子) and some include the feminine kula-duhitṛ "a daughter of the community" (Tib. rigs kyi bu mo. Ch. shàn nǚrén 善女人). There is a tendency to see this as a generic term for a high status person. The Chinese translation shàn nán zǐ 善男子 literally means "good male child", where shàn 善 is the standard translation of kuśala.

In a Buddhist context kulaputra is often translated along the lines of "good", "gentle", or "nobly born". A kula is any coherent collection of animals: "herd, troop, flock, etc"; or people "race, family, community, tribe, caste, clique, fraternity, etc." Context suggests that the community in question is the Buddhist community made up of four saṃghas: bhikṣu, bhikṣūnī, upasaka, and upasikā. Such definitions as I can find emphasise that the term is used for one's social inferiors: a teacher calls a pupil kulaputra, but not the other way around. It means "one of the flock". And the prestige derives from Buddhists' attitude of Buddhist exceptionalism. That said the use of it does seem to drift. In my notes I adopted the translation devotee.

As I said above, there is a mistake in Śāriputra's question in T 253:

善男子！若有欲學甚深般若波羅蜜多行者，云何修行？
Kūlaputra, if there is a desire to genuinely learn the deep Prajñāpāramitā practising, how should they study the practice?

The passage is punctuated in CBETA as though kulaputra is a vocative, i.e. 善男子！meaning Śāriputra is addressing Avalokiteśvara as kulaputra. What we see more often is what we find in Sanskrit and Tibetan, that the two of them discuss the kulaputra in the third person, in the abstract. Or what we see in T 252, the teacher addressing the pupil as kulaputra. This appears to be a translation error, in which the translator has misunderstood their source, most likely a Sanskrit text and encoded this misunderstanding in Chinese.

Avalokiteśvara Preaches

With respect to kulaputra, R1 texts now answer the question as if Śāriputra had asked about both sons and daughters (kuladuhitṛ) of the community (Ch. shàn nánzǐ 善男子, shàn nǚrén 善女人). The only text that did phrase the question that way was the Sanskrit. Recall that R2 doesn't use kulaputra/kuladuhitṛ in this way and that Śāriputra doesn't ask a question. Variations in the names continue. T 253 in particular has both of the two different ways of writing Śāriputra in succession: Shèlìfú 舍利弗 and Shèlìzi 舍利子.

Avalokiteśvara opens his teaching by repeating the phrase in the question (that is also found in the opening paragraph of the standard version), i.e.

yaḥ kaścic chāriputra kulaputro va kuladuhitā vā asyāṃ gambhīrāyāṃ prajñāpāramitāyāṃ caryāṃ cartukāmas tenaivaṃ vyavalokitavyam. 
"Whichever son or daughter of the community who desires to perform this profound paragnosistic practice, therefore he should observe in this way.

The problem that I pointed out in Attwood (2012) holds here as well, i.e. vyava√lok is a transitive verb, so what is the devotee supposed to observe. The answer is obvious from the context, they are supposed to observe the five branches of experience (pañca skandhāḥ). Conze failed to grasp this in his edition of the standard text and does not see it here either. 

There are, however some major variations that are important to helping to make sense of the Tibetan versions. The first notable variant is that Ce (from Feer's 1866 polyglot edition) has śikṣatavyam "he should train" instead of vyavalokayitavyam "he should observe". 

• Ce śāriputra kulaputro vā kuladuhitā vā asyāṃ gambhirāyāṃ prajñāpāramitāyāṃ varttakāmas tenaiva śikṣitavyaṃ || yaduta pañcaskandhāḥ svabhavaśūnyāḥ || katham svabhāvaśūnyāḥ ||

Words based on √śikṣ are common in Prajñāpāramitā. In T 257 the character xué 學 suggests that the translator might have had a text from the same lineage as Ce.

The other thing is that Nb, Ce, and Jb (Hasedera ms.) both add a phrase concerning the five skandhas. 

• Nb yaḥ kaścit kulaputrā vā kuladuhitā vā asyā gambhirāyāṃ prajñāpāramitāyāṃ caryā catrukāma tenaiva vyavalokayitavyaṃ || pañca skandhān svabhāva śunya vyavalokatitavyaṃ ||
• Ce śāriputra kulaputro vā kuladuhitā vā asyāṃ gambhirāyāṃ prajñāpāramitāyāṃ varttakāmas tenaiva śikṣitavyaṃ || yaduta pañcaskandhāḥ svabhavaśūnyāḥ || katham svabhāvaśūnyāḥ ||
• Jb yaḥ kaścic chāriputra kulaputra vā kuladuhitā vā gambhīrāyāṃ prajñāpāramitāyāṃ caryā cartukāmas tenaivaṃ vyavalokayitavyam | pañcaskandhāḥ | tāṃś ca svabhasūnyān samanupaśyati sma |

The Chinese and Tibetan texts all have this additional clause, but in each case it appears to be better integrated than the Sanskrit, where the tense of the verb frequently clashes with the rest of the phrase. E.g.

TibB. shā ri'i bu rigs kyi bu 'am / rigs kyi bu mo gang la la shes rab kyi pha rol tu phyin pa zab mo spyad pa spyod par 'dod pa des / phung po lnga po de dag ngo bo nyid kyis stong par yang dag par rjes su mthong ba de ltar blta bar bya ste /

“Śāriputra! Whichever gentle son or gentle daughter desires to practise the practice the profound perfection of wisdom, he [sic] remarks that those those five aggregates are inherently empty, and should observe thus:”

However, the Tibetan texts are almost certainly translated from a Sanskrit source. Note the "he remarks" in Silk's translation contrasting with the subject "Whichever gentle son or gentle daughter". This could be a quirk of Tibetan.  

Jb has just tacked on the final part of the first paragraph of the standard text without alteration, leaving the final verb in the past tense so that it clashes with the rest of the sentence, particularly the future passive participle: "it should be observed by him... he perceived...". 

Where the Sanskrit adds this phrase it separates the verb (vyavalokayitavyam) from its object (pañcasakandha) by giving pañcaskandha in the wrong case (nominative plural instead of accusative plural) and often by adding a daṇḍa between them. I showed in "Heart Murmurs" (Attwood 2015) that this was an error. The phrase is vyavalokayitavyam pañcasakandhām "he should examine the five branches of experience." In order to be consistent with this (and with the intent of the Xīnjīng) the final phrase should have to be altered to tāṃs ca sarvaśūnyān dṣṛṭvāyam "and should see that they are  all absent".

None of the extant Sanskrit texts manages anything sensible and there is a great deal of variation. This suggests that scribes who could read Sanskrit were trying to make sense of the text as they copied it and took different approaches, but none looked to the Xīnjīng. However, in this passage all of the Chinese versions have replaced "he examined the five branches of experience" (zhàojiàn wǔyùn, jiē kōng 照見 五蘊皆空) with the Madhyamaka-inspired "they should contemplate the Five Skandhas as empty of self-nature" (Yīng guān wǔyùn xìng kōng 應觀五蘊性空). Also note the change in the verb from "examines" (zhàojiàn 照見)  to "should contemplate" (yīngguān 應觀; 應 having a similar sense to the future passive participle). This change argues against the extensions being made in Chinese, since it is the Sanskrit standard text (Hṛdaya) that introduces the svabhāvaśūnya where we expect śūnyatā from the Xīnjīng. 

 

This brings us to the end of the opening extension. The text now settles into the standard Heart Sutra, although the extra negations are almost universal in the Skt texts. And absent from all of the Chinese translations. We come back to the text after the dhāraṇī.

The Buddha's Endorsement

There is a small anomaly in T 253

T 253 般若波羅蜜多行，應如是行
T 254 般若波羅蜜多行，應如是學
Skt. prajñāpāramitāyāṃ caryāyāṃ śikṣitavyaṃ

The second xíng 行 is probably a scribal error for xué 學, perhaps an eye-skip. Although note that TibA and TibB leave out "the practice of" completely:

TibA de ltar... bslab par bya'o "in that way... he should train"
TibB de ltar bslab par bya'o // "he should train in that way"

But in Tibetan, the question was asked in these terms (in Paragraph G).

T 257 adds a whole extra phrase: "If able to recite this Prajñāpāramitā dhāraṇī" (ruò néng sòng shì bōrěbōluómìduō míngjù 若能誦是般若波羅蜜多明句). Having studied the many ways that Chinese authors translated vidyā in a Prajñāpāramitā context I guess that míngjù 明句 literally "bright verse" is yet another way of translating vidyā or dhāraṇī, or even vidyā-dhāraṇī. This phrase has no parallel in any other version of the Heart Sutra.

T 253:

即時世尊 從廣大甚深 三摩地起
jíshí shìzūn cóng guǎngdà shènshēn sānmódì qǐ
The Bhagavan having arisen from the vast and profound samādhi...

This expression with the verb in the final position seems to be a Chinese idiom (Chinese is typically SVO) rather than a Sanskritism (Sanskrit is usually SOV). In T 223: cóng zuò qǐ 從座起 "he rose from his seat" literally "from seat rising" (8.229c08); also jiē cóng zuò qǐ 皆從座起 "all rose from their seats" (8.230b12). And numerous other examples.

It is slightly peculiar that the Sanskrit suggests that the teaching would be approved of by all the tathāgatas and arhats (anumodyate sarva-tathāgatair arhadbhiḥ). The arhats, generally speaking do not approve of Mahāyāna teachings, especially in the 8th Century, by which time the distinction has become somewhat schlerotic. However, as we have seen, the chief proponent of Prajñāpāramitā is Elder Subhūti, the arhat. Many of the Sanskrit manuscripts omit reference to the arhats as do all the Chinese and Tibetan translations.

T 257 also inserts a phrase—"it is real, supreme, and final" (shì jí zhēnshí zuìshàng jiùjìng 是即真實最上究竟)—which combines some common Buddhist superlatives that are also used in the standard Heart Sutra: jiùjìng 究竟 is from 究竟涅槃 translated in the Sanskrit Heart Sutra as niṣṭhanirvāṇa "final extinction" but probably more like nirvāṇa-paryavasāna "concluding in extinction"; zhēnshí 真實 is from the expression at the end of the epithets passage zhēnshí bù xū 真實不虛 "[Prajñāpāramitā] is truly real and not in vain."

Nothing parallel to this material is found in Recension Two (T 252).

Rejoicing

The final paragraph in both recensions sees the audience rejoice at the teaching, employing a pericope that is similar in both R1 and R2. The phrase, "together with everyone in the gathering" (ca sarvāvatī parṣad), is a little unusual but is followed by the utterly stock phrase about the world with its various kinds of beings (sadeva-mānuṣa-āsura-gandharvaḥ lokaḥ). Most of the mss include the bodhisatvas (te ca bodhisatvā mahāsatvāḥ ) but this phrase is omitted in Jb, TibA, TibB, and Nh.

It was notable that the opening mentioned the bodhisatva-saṃgha, but they are not mentioned here.

There are some differences between the Conze edition and the Hasedera Manuscript (Jb).

Conze: āttamanā āyuṣmāñc Chāriputraḥ āryāvalokiteśvaro bodhisatvo mahāsatvas te ca bodhisatvā mahāsatvāḥ sā ca sarvāvatī parṣat sadeva-mānuṣāsura-garuda-gandharvaś ca loko bhagavato bhāṣitam abhyanandann iti.

Jb: ānandamanā āyuṣmān śāriputraḥ āryāvalokiteśvaraśca bodhisatvaḥ sā ca sarvāvatī pariṣat sadeva-mānuṣāsura-gandharvaś ca loko bhagavato bhāṣitam abhyanandan||

Jb has ānandamanā आनन्दमना for āttamanā आत्तमना, which looks like a simple scribal error. Jb also omits te ca bodhisatvā mahāsatvāḥ, and omits garuda from the list of beings. Jb also omits the sandhi that affects āyuṣmān śāriputraḥ: the rule here is complex but there are two possible outcomes here: āyuṣmāñcchāriputra āryāvalokiteśvaro or the more comprehensible āyuṣmāñ śāriputra āryāvalokiteśvaro. 

Unfortunately, the notation in Conze's (1967) edition completely falls apart at this point. E.g. the citation labelled "na" doesn't exist in Conze's notes. Some of his lettered citations are both the beginning and the end of a passage, some are adjacent to words but the notes deal with phrases. Numbered citations go up to 61, but the notes only go up to 58. I have found at least one variant omitted (note d, Nb tasmā tahi). Also Nb and Ce omit garuḍa but this is not noted by Conze.

Conclusion

The most striking thing, for me, is the sheer scale of variation in various documents instantiating the extended Heart Sutra text. The later manuscripts are overburdened with an accumulation of scribal errors but most also show signs of being deliberately edited. In a text of around 280 words, even small changes can be very important (even if it can take years of examination to see them). Passages have been added, omitted, and changed quite freely. So describing the extended Heart Sutra is no easy task. And while a broad outline of the stemma is not so difficult, including all the variations and horizontal influences (I'm trying to avoid the jargon "contamination" here) is beyond me at this stage. This is where computer-generated phylogenetic diagrams would be of use. 

T 252 is substantially different in every paragraph of the extensions despite also containing the standard Heart Sutra at its core. The protagonists are still the Buddha, Avalokiteśvara, and Śāriputra, but these are forced on any redactor by the conventions of Buddhist sūtra composition and the content of standard Heart Sutra. The differences between T 252 and the rest make it appear that this is a separate, possibly prior effort to extend the text (the traditional date is certainly earlier). The idea that the Heart Sutra was extended twice and no one noticed till now would be entirely in keeping with the history of the text and the history of scholarship on the text. I have found no evidence of who might have created T 252, but it may well have been the man credited with translating it, i.e. Fǎyuè 法月 or *Dharmacandra.

Lacking any working knowledge of Tibetan grammar I have not done a detailed comparison of the Tibetan recensions with the extant Sanskrit sources. But they are clearly a better fit with some than with others. Paragraph I of TibA and TibB suggests the hands of two different translators or redactors. The two read the conclusion of the extended opening in quite different ways. There is also the (to date) informal suggestion, from Ben Nourse, that different Tibetan manuscripts found at Dunhuang correspond somewhat to the two recensions.

These notes are preliminary to any attempt to tackle a critical edition. I also await a definitive study of the Dunhuang Heart Sutra manuscripts.

Language of Composition

If we are to establish what the language of composition was, what criteria would we use? With the Heart Sutra the task of deciding was relatively easy because of the copied passages - we could compare the Heart Sutra versions of the passage with the Large Sutra versions (which is what Jan Nattier did in 1992). The extensions, however, are a patchwork of pericopes. This is not unusual, since all Buddhist texts are to some extent modular (See Work, Text, Document), but it doesn't give us much purchase.

We could argue that the text which is more coherent and has the fewest grammatical errors is likely to be the original. On the other hand, most of the texts we use in this comparison have been edited and standardised for publication and/or canonisation. And all of them have been copied multiple times by scribes. Copying of manuscripts in greater India and its cultural sphere was often rather careless. Even after writing was widely used, sūtras were memorised and recited rather than read. It's likely the early transmissions of sūtras to China were memorised. Buddhists texts were often objects of religious devotion, never intended to be read or studied and thus careless copying didn't matter that much. In China, at least, scribes were literate and excellent handwriting was highly prized. The downside of this is that the educated scribe is more likely to casually "correct" a manuscript that they don't understand.

The Buddha's dharma teaching and samādhi is a point of departure for the R1 texts. It seems to me that the Sanskrit text which names the dharmaparyāya and leaves the samādhi unnamed is likely to be original. The Chinese texts don't mention a dharmaparyāya and this would be an odd detail to add in the circumstances. Or we could say that, since the Chinese R1 texts are in agreement, that the work that does not feature the Buddha giving a dharmaparyāya, and the presence of one in Sanskrit, is the oddity. However, the Tibetan texts, especially TibA, do seem to have both dharmaparayāya (Tib. chos kyi rnam grangs kyi) and samādhi (Tib. ting nge 'dzin). 

The Chinese texts, except for T 253, pick up on an editorial blunder that occurs in the Sanskrit standard text where svabhāvaśūnyan (zì xìng jiē kōng 自性皆空) is found where we expect śūnyatā (jiē kōng 皆空) based on the Xīnjīng. If one were to extend the Chinese text in Chinese this is not so likely. But note that the very next phrase in T 253 and 254—lí zhū kǔ è 離諸苦厄—is drawn from the Xīnjīng dù yīqiè kǔ è 度一切苦厄. 

T 253 treats kulaputra in Śāriputra's question to be a vocative, addressing Avalokiteśvara. This seems a very unlikely mistake to make if one were composing in Chinese. It looks like a translation error, a misreading of a Sanskrit text. Also with respect to the question in Para E, only Sanskrit asks in terms of both sons (putra) and daughters (duhitṛ) of the community (Ch. shàn nánzǐ 善男子, shàn nǚrén 善女人). Chinese and Tibetan texts phrase the question in terms only of sons. But they all have Avalokiteśvara answer the question in Para I in terms of both. Thus only the Sanskrit text is consistent in this case.

Another argument for a Sanskrit original for R2 is an awkwardness that occurs because of the use of the verb vyavalokayati. This works well enough in the standard Heart Sutra but when the Redactor tries to recast this verb in the standard form of a Prajñāpāramitā question, i.e. "how should the bodhisatva go about his business", where the activity is phrased using a future passive participle, the transitivity of vyavalokayati trips them up. 

For example, if the bodhisatva was to train (śikṣati*) in some form of Buddhist practice then the question would be kathaṃ śikṣitavyaṃ "how should he train?". And after the explanation Avalokitesvara might say "for this reason he should practice this way" tenaiva śikṣitavyaṃ (Ce, Feer's polyglot edition). Unfortunately, most of the Sanskrit manuscripts finished with "for this reason he should examine in this way" tenaivaṃ vyavalokitavyam. The reason it sounds so awkward is that the verb is transitive (Conze makes this mistake throughout his edition); that is, one cannot simply examine in the absence of something to examine. We saw above that the Translator filled out a Chinese phrase involving the implication that looking involves seeing. This is just the kind of adjustment a translator has to make moving between very different languages. But because the Translator chose a transitive verb (vyaavalokayati) this left the Redactor in a bind. In order to maintain consistency they had to have Avalokiteśvara use the awkward phrase tenaivaṃ vyavalokitavyam, something that would not have happened if the Sanskrit text were a translation of an idiomatic Chinese phrase. In this case the infelicitous Sanskrit also indicates that the Redactor of R1 was working in Sanskrit. 

*śikṣati is the desiderative of √śak "be able" and therefore literally means "the desire to be able" but it is used to indicate the training that a Buddhist practitioner (cārin) undertakes in Buddhist practices (cārya). 

So my first impression, which needs further scrutiny, is that the standard Heart Sutra was extended twice. The first time produced the text T 252 and since there is no evidence of it in any other language we can conjecture that it was made in Chinese (just like the Xīnjīng). This fits well with my revised history of the standard Heart Sutra as a Chinese digest text (chāo jīng 抄經) which was "authenticated"  based on the misperception that it was a translation from Sanskrit by Xuanzang and a forged Sanskrit "original" (actually a translation from Chinese). 

The standard Heart Sutra was extended a second time, probably in Sanskrit, although with varying influence from the Chinese (especially in T253), which was then translated into Chinese (T253, 254, 257) and Tibetan (including some Dunhuang manuscripts and the canonical versions), and from Tibetan into Chinese (T 255).

I see two new conjectures emerging from this study. 1. The extended Heart Sutra exists in two recensions; and 2. analysis of the language of the documents and editions strongly suggests that R1 was redacted in Sanskrit, while the lack of evidence for other versions suggests that R2 was redacted in Chinese. Also, since T 252, the sole representative of R2, is the earliest dated version of the text, I have got the nomenclature wrong. T 252 represents Recension One, an early attempt to create a more authentic Heart Sutra (by 7th Century Chinese Buddhist standards of authenticity). However, R1 never caught on. Recension Two is a second, probably later, perhaps unrelated attempt at a more authentic Heart Sutra, this time produced from the Sanskrit standard Heart Sutra and gave rise to all the other Chinese and Tibetan translations as well as a line of Sanskrit copies. 

~~oOo~~

Bibliography

Apple, James B. 2015. "Redaction and Rhetoric in Mahāyāna Sūtras: The Case of the Jayamatiparipṛcchāsūtra." Indo-Iranian Journal, 58(1): 1-25.

Attwood, J. 2019. "Xuanzang’s Relationship to the Heart Sūtra in Light of the Fangshan Stele." Journal of Chinese Buddhist Studies, 32: 1–30. http://chinesebuddhiststudies.org/previous_issues/jcbs3201_Attwood(1-30).pdf

Attwood, J. (forthcoming). "The History of the Heart Sutra as a Palimpsest." Pacific World.

Conze, Edward (1948). "Text, Sources, and Bibliography of the Prajñāpāramitā-hṛdaya." Journal of the Royal Asiatic Society, April 80(1-2): 33-51.

Conze, Edward. (1967). "The Prajñāpāramitā-Hṛdaya Sūtra." In Thirty Years of Buddhist Studies: Selected Essays, Bruno Cassirer, pp. 147-167. Modified version of Conze (1948).

Karashima, Seishi. (2016) "On Avalokitasvara and Avalokiteśvara." In Annual Report of The International Research Institute for Advanced Buddhology at Soka University (ARIRIAB), vol. 20 (2017): 139-165.

Lopez, Donald S. (1988) The Heart Sūtra Explained: Indian and Tibetan Commentaries. State University of New York Press.

Lopez, Donald S. (1996) Elaborations on Emptiness: Uses of the Heart Sutra. Princeton University press.

Nattier, Jan. (2007) "Avalokiteśvara in Early Chinese Buddhist Translations: A Preliminary Survey." In Magee, W and Huang, Y.H. (Eds). Bodhisattva Avalokiteśvara (Guanyin) and Modern Society. Proceedings of the Fifth Chung-Hwa International Conference on Buddhism, 2006: 191-212. Taiwan: Dharma Drum Publishing.

Schopen, Gregory. 2004. "If you can’t Remember, How to Make it up, Some Monastic Rules for Redacting Canonical Texts." In Buddhist Monks and Business Matters, Still More Papers on Monastic Buddhism in India. Honolulu: University of Hawai’i Press. 395-407.

Silk, Jonathan A. (1994) The Heart Sūtra in Tibetan: a Critical Edition of the Two Recensions Contained in the Kanjur. Arbeitskreis für Tibetische und Buddhistische Studien, Universität Wien.

Jonathan A. Silk (2015) "Establishing/Interpreting/Translating: Is It Just That Easy?" Journal of the International Association of Buddhist Studies. Vol. 36/37: 205-225.