Chinese Heart Sutra: Dates and Attributions

Xīnjīng

One of the important conclusions of Jan Nattier's 1992 article on the Heart Sutra was that the traditional dates ascribed to its composition could not be correct and that it is more likely that it was composed in the 7th century, a time period which coincides with the life of Xuánzàng (602 – 664 CE) and his activity as pilgrim and translator. This coincidence allows Nattier to speculate that it might even have been Xuánzàng who translated the text from Chinese into Sanskrit. The speculation is bolstered by the fact that Xuánzàng has form in this area. He is known to have translated the Chinese authored 《大乘起信論》 (Dàshéng qǐxìn lùn) or Awakening of Faith in Mahāyāna into Sanskrit.

In this essay I will rehearse Nattier's arguments about chronology and attribution of the Chinese translations as a prelude to discussing the challenge to them published by Dan Lusthaus. Lusthaus (2003) draws attention to two Chinese commentaries on the Heart Sutra, one of which appears to refer to alternate versions of the Heart Sutra in a way that Lusthaus claims poses "serious problems" for Nattiers conclusions about the chronology of the Heart Sutra. As one of the few scholars to engage critically with Nattier's thesis in print, Lusthaus's article is interesting both for the new information it presents and for the test it provides for the Chinese Origin thesis.

Another reason to rehearse this aspect of Nattier's thesis, is that that the popular Zen inspired commentaries seem to struggle with it. Red Pine, Mu Seong, and Kazuaki Tanahashi all seem to be in denial about the evidence. As such, most modern readers are given the impression that Nattier's argument is weak or improbable. But this is not the case.

In this essay I favour the Pinyin Romanisation of Chinese characters. Lusthaus and Nattier both use the Wade-Giles system. Additionally, Lusthaus uses McCune–Reischauer for the name of the Korean monk 원측, and the Revised Romanisation of Korean proposed by the South Korean government is now standard. I silently emend their Romanisation to fit my own preference (and recent scholarly convention). In particular I change the names:

• 玄奘: Hsüan-tsang > Xuánzàng
• 원측: Wŏnch’ŭk > Woncheuk
• 窺基: K'uei-chi > Kuījī

Nattier's Comments on the Authorship and Dates of The Heart Sutra

The Heart Sutra exists in three short versions in the Chinese Tripiṭaka. This essays focusses on T250 and T251 attributed to Kumārajīva (334–413 CE) and Xuánzàng respectively. T256 is now thought to be attributable to Amoghavajra (705–774) and directly influenced by the Sanskrit text. The main argument for this is in Japanese, but a summary can be found in Tanahashi (2014: 68).

As Nattier points out, the attributions of T250 and T251 first appear in an 8th century catalogue of Buddhist texts called 《開元釋教錄》Kāiyuán shìjiào lù (T2154) long after both men were dead (1992:174). This raises the question of why this very popular text failed to be associated with either in their lifetime, especially when we consider the explicit links between Xuánzàng and the Heart Sutra in his biographies. The simplest answer is that neither was involved in the creation of these versions. As we will see this is also the most plausible answer.

The catalogue of Buddhist texts in China 《綜理衆經目錄》 Zōnglǐ zhòngjīng mùlù (compiled in 374), itself now lost but reproduced in 《出三藏集記》 Chū sānzàng jíjì, compiled around 515 by Seng-yu (僧祐; 445-518), records two texts considered to represent lost versions of the Hṛdaya in Chinese. The two titles mentioned are:

《摩訶般若波羅蜜神咒一巻》
Móhēbōrěbōluómì shénzhòu yī juàn
Great Perfection of Wisdom Vidyā in one scroll

《般若波羅蜜神咒一巻》
bōrěbōluómì shénzhòu yī juàn
Perfection of Wisdom Vidyā in one scroll

These titles are certainly similar to the Chinese sutra titles:

T250 《摩訶般若波羅蜜大明呪經》
Móhēbōrěbōluómì dàmíngzhòu jīng
Mahāprajñāpārami[tā]-mahāvidyā-sūtra.

T251 《般若波羅蜜多心經》
Bōrěbōluómìduō xīn jīng
Prajñāpāramitā-hṛdaya-sūtra.

T256 《唐梵翻對字音般若波羅蜜多心經》
Táng fàn fān duì zì yīn bōrěbōluómìduō xīn jīng
Tang [i.e. Chinese] Transcription of the Sanskrit Prajñāpāramitā-hṛdaya-sūtra.

However the similarity itself is suspicious, because it was Kumārajīva who introduced the transcription 般若波羅蜜 bōrěbōluómì for prajñāpāramitā. Nattier points out that earlier translations of the Large Perfection of Wisdom Sutra do not use this terminology. For example:

T221 《放光般若經》Fàngguāng-bōrě-jīng, by Mokṣala (291 CE)

T222 《光讚經》 Guāng zàn jīng, by Dharmarakṣa in 286 CE

However early translations of the Aṣṭasāhasrikā Prajñāpāramitā Sūtra use 般若bōrě for Sanskrit prajñā, and one uses 摩訶 Móhē for Sanskrit mahā. Eg.

T224 《道行般若經》Dàohéng-bōrě-jīng, by Lokakṣema (179 CE).

T226 《摩訶般若鈔經》Móhēbōrěchāo-jīng, by 竺佛念 Zhúfóniàn (382 CE).

I think this undermines the argument that the title is anachronistic. Nattier's dismissal on the grounds that the two supposed early texts containing the term 神咒 shénzhòu because "both are clearly intended to be construed as mantras based on - or at least associated with - the Prajñāpāramitā corpus." (1992: 183) is less convincing because mantras did not come into Chinese Buddhism for some centuries after the supposedly early period of the texts. On the other hand the use of the phrase 神咒 may itself be anachronistic. Mantras were non-Buddhist until after this period, but dhāraṇī and vidyā (along with Pāḷi parittas) were not. The idea that the Heart Sūtra is itself intended as a dhāraṇī is one that Nattier herself discusses (1992: 175-6). On the other hand, another early translation of Aṣṭa, 《大明度經》Dàmíngdù-jīng (T225) by 支謙 Zhīqiān (225 CE), uses the character combination 神呪 (or possibly 神祝, the editions disagree) to represent Sanskrit vidyā, and Xuánzàng apparently employs 神咒 for the same word. So the titles of the two "lost translations" are not so unusual after all. But it is possible that the catalogue was edited at a later date to include texts that could not have existed at the time, and it's also plausible that a no-longer extant text predates both T250 and T251 because of their variations (and differences between them and the Sanskrit mss.). I do not think that T250 or T251 are a plausible ur-text.

If they did exist, the two texts are now lost and we cannot draw any hard and fast conclusions about them, however ambiguous the evidence. We certainly ought not to join Red Pine in taking their existence on face value.


Kumārajīva & T250

Having decided that we must set aside non-existent texts, Nattier then turns to the ascription of T250 to Kumārajīva. This was already in doubt as Conze attributed it to Kumārajīva’s pupils (1978: 20). Nattier summarises the consensus view:

"...it seems clear that the students of Kumārajīva (in particular, Sēngzhào) read and commented on the core passage of the Heart Sūtra found in Kumārajīva's version of the Large Sūtra [ie. T223]. There is no evidence, however, that they were aware of the existence of the Heart Sūtra as a separate text, nor is there any evidence that Kumārajīva himself had any role in the production of the 'translation' associated with his name." (1992: 184)

It is precisely this consensus of informed opinion that Tanahashi (2014) rejects when he refers to T250 as the "α-version", doing his readers a disservice. There is simply no way that T250 is the ur-text for the Heart Sutra. It clearly dates from after Kumārajīva's death and has been edited by third parties unknown. It's interesting to note also that Sēngzhào (ca. 378—413 CE) is associated with the establishment of Madhyamaka, whereas Xuánzàng and his students were instrumental in establishing Yogācāra thought in China.

That the Heart Sūtra is based on Kumārajīva's translation T223, or perhaps on the version found embedded in 《大智度論》Dà zhì dù lùn (= *Mahāprajñāpāramitopadeśa; T1509), is not in doubt. The similarity between the two is too great to be a coincidence. That the Heart Sutra is based on the large Perfection of Wisdom text is also evident in some of the Nepalese Manuscript titles. For example the new Hṛdaya manuscript (EAP676/2/5) I described in 2014: is titled Ārya-pañcaviṁśatikā-pajñāpāramitā-mantra-nāma-dhāraṇī which translates as The Dhāraṇī named The Mantra of the Noble 25,000 Perfection of Wisdom.

The argument against attributing T250 to Kumārajīva is complex (Nattier 1992: 184-189). Where T250 has two passages of extra characters, these can be traced to T223. Nattier asserts, not entirely convincingly I think, that it is unlikely that the parallels would have been translated identically by Kumārajīva and that the exact correspondence argues for a plagiarism. The argument would be stronger if we had some concrete examples of this actually happening. I can supply an example of Kumārajīva's inconsistency from his translation of the Aṣṭasahāsrikā-prajñāpāramitā-sūtra, i.e. 《小品般若經》 T227. At 8.542.b5-6 Kumārajīva translates vidyā as 呪術 zhòu shù, while a little later at 8.543b25-29 he translates first as 明呪 míng zhòu, and then simply as 呪 zhòu. More examples would be needed to establish a pattern, but it lends plausibility to Nattier's assertion.

Nattier further points out that the initial equation of form and emptiness conforms not to T223, but to T1509 《大智度論》. The combination of observations leads Nattier to propose that T250 is based on, or has been made to conform to, T1509, rather than T223. Thus, the earliest possible date (terminus post quem) for T250 is the date of the translation of T1509, ca. 406 CE (1992:188).

Nattier's next step is to point out that, unlike Kumārajīva's other translations, which eclipse Xuánzàng's in popularity even to this day, T250 was never popular in China. Unlike T251, T250 is not craved into stone, copied, or printed. Not only are all the Chinese commentaries on the Heart Sutra on Xuánzàng's version, T251, but they do not date from earlier than Xuánzàng's lifetime, whereas Kumārajīva was active 250 years earlier. Thus the attribution of authorship of the Heart Sutra to Kumārajīva rings hollow. And in fact Kumārajīva is frequently apocryphally given as author or translator when it is clear that he is not.


Xuánzàng & T251

However the attribution of T251 to Xuánzàng is also problematic. Xuánzàng was a prolific translator. His compendium of Prajñāpāramitā texts (T220) takes up vols. 5-7 of the Taishō edition of the canon, each of which is thicker than Vol. 8 containing all the other Prajñāpāramitā texts translated by all the other translators. If Xuánzàng translated the Heart Sutra why was it not attributed to him in his lifetime, and why was his translation not included in T220? Why does the legend of his association with the text speak of him receiving the text from a sick man if he composed it or translated it from Sanskrit?

Curiously T251 largely sticks to the terminology found in T250 (and thus in T223/1509). But three key terms: the names Avalokiteśvara and Śāriputra, and the Sanskrit word skandha, are written in a way that is distinctive to Xuánzàng. A text containing 觀自在, 舍利子, and 蘊 can only have been completed during or after the work of Xuánzàng. Nattier concludes that Xuánzàng did indeed receive a text and made minor amendments. T250 seems also to be an amended text, which suggests to me an ur-text of which both T250 and T251 are revisions. This is supported by the Sanskrit text which is significantly different in places from either of the two Chinese versions, in particular it has no equivalent of 度一切苦厄 in the first sentence. That the Sanskrit translator would drop this phrase is less plausible than that at some later date it was added to the Chinese text. This is because everywhere we look, Buddhists add words, phrases, and chapters to their texts, but we very seldom see them subtracting. Indeed in light of recent scholarship, Conze's view that the Prajñāpāramitāhṛdaya and Vajracchedikāprajñāpāramitā represent 3rd or 4th Century condensations of the Prajñāpāramitā texts seems unlikely. Vaj is now thought to be contemporary with Aṣṭa and the character of Hṛdaya is not a condensation, but simply a quote or two.

Either way the Heart Sutra as we know it can be no older than the early 5th century, i.e. after Kumārajiva's translations of the Pañcaviṃśatisahāsrikā Prajñāpāramitā Sūtra and/or Mahāprajñāpāramitopadeśa because it is an extract of one or both of them. Another part of the text that is cited word for word from the same source is the passage with epithets of prajñāpāramitā, found at T223, 8.286b28-c7 and many other locations: see Nattier (1992 footnote 54a) and my forthcoming article for JOCBS.

So the catalogues which list earlier texts are most likely forgeries. And on this basis Nattier proposes that the Heart Sutra was composed in or near the 7th Century in China. The close association with Xuánzàng suggests that he may have been involved in the translation of it into Sanskrit, though given how botched the translation is, it was presumably well before his work on 《大般若波羅蜜多經》(T220). My view is that the translator from Chinese to Sanskrit was more at home in Chinese than in Sanskrit, and not very familiar with the Prajñāpāramitā literature in Sanskrit.

Woncheuk's Commentary.

Xuánzàng’s students 窺基 Kuījī (632–682) and 圓測 Woncheuk (613-696) produced commentaries on the Heart Sutra in the late 7th century (Nattier 1992: 173). These have both been translated into English: see Shih & Lusthaus (2006) and Hyun Choo (2006) respectively.

Lusthaus (2003) cites four passages from Woncheuk's commentary 《般若波羅蜜多心經贊》 (T1711), which he says lead us to two main conclusions: 1. that versions of the text once existed that were different from the extant versions; and 2. that these versions were older than the extant versions. And thus Nattier's preference for a later composition date is seriously challenged.

The first passage comments on Xuánzàng’s use of the form 觀自在 Guānzìzài for the name of Avalokiteśvara Woncheuk comments:

若依舊本名觀世音 (T1711, 33.543b.21)

"This is what the old text(s) named Guānshìyīn" (Lusthaus 2003: 82)

Quite a lot of Lusthaus's argument rests on his conclusion that it is "natural in this context to understand this as a reference to older versions of the Heart Sutra" (82). Hyun Choo (2006) concurs, he translates the passage "According to the old version of the translation [of the Prajñāpāramita-sūtras]" (138). However, as is well known Avaoliketśvara does not appear in any other Prajñāpāramita sūtras, so this is an unlikely interpretation. In fact, Woncheuk's commentary immediately proceeds to a discussion of the deity and the name 觀音in Buddhist literature, a discussion that does not include any Prajñāpāramitā sūtras or mention of the

Heart Sutra, but does include the Avalokiteśvara-sūtra (觀音三昧經), Avalokiteśvara-bodhisattva-mahāsthamaprapta-bodhisattva-sūtra(觀音授記經), and the Larger Sukhāvatīvyuha-sūtra (無量壽經). If we are talking about "natural" conclusions then Woncheuk's reference to 舊本 'old texts' appears to reference these other named texts.

The next passage concerns the first sentence of the Heart Sutra:

或有本曰 「照見五蘊等皆空」 雖有兩本。後本為正。撿勘梵本有等言故後所說等準此應知。(added punctuation for clarity)

There is another version of the text 或有本 which says "illuminatingly, he saw the five skandhas, and so on (等), are all empty." Although there are two versions of the text 有兩本, the latter text is correct. An examination of the Sanskrit text [梵本] shows that is has the word "and so on" (等). Hence the 'and so on' stated by the latter (text) should be understood to be the standard." (Lusthaus 2003:83, emphasis added)

By 'and so on' we can probably interpret Sanskrit ādi. T251 here simply has 照見五蘊皆空 without the extra character 等. Given that the text does list the skandhas and other lists such as the dhātus and āyatanas this interpolation is not wrong. However, as Lusthaus concedes, ādi doesn't appear in any known Sanskrit text. Nor does any extant Chinese text have 等 here. The mention of a Sanskrit text with a different wording here is interesting of course, but the manuscript tradition of the Heart Sutra is widely variable - so much so that editing it proved very difficult for Conze and led him to make several errors (See my forthcoming article in the JOCBS 7). No two manuscripts of the Sanskrit Hṛdaya are identical, even the oldest manuscript (the Hōryūji Manuscript; probably from the 8th century) is obviously corrupt in many places.

Next, Lusthaus cites this passage:

又解此經自有兩本 一本如上。一本經曰受想行識等亦復如是。所言等者準下經文有六善巧。謂蘊處界緣生四諦菩提涅槃。(T1711, 33.546.13-15)

"Further, for interpreting this sutra we have two texts (自有兩本). One text is as above 如上 (i.e. Xuánzàng's version, which says 'vedanā, saṃjñā, saṃskāras, and vijñāna are also like this'). The other text of the sutra says: 'vedanā, saṃjñā, saṃskāras, vijñāna, and so on 等, also like this.' The word 'and so on' [deng] indicates what is [discussed] below in the text of the sutra, i.e. the six skill in means, the aggregates, āyatanas, dhātus, pratītysamutpāda, the four truths, Bodhi, and Nirvāṇa." (Lusthaus 2003: 84).

From this we infer that Woncheuk has at least two texts in front of him. Possibly two Chinese texts and at least one Sanskrit text. And one of the Chinese texts again has 等 (= Sanskrit ādi) at the end of a list of skandhas, seeming to indicate the other lists that follow in the sutra. Again no extant Chinese or Sanskrit text has this additional feature, but it is not inconceivable, in the light of the manuscript tradition, that it could have been added by a scribe or editor.

Woncheuk's contemporary and rival, Kuījī, also wrote a commentary on the Heart Sutra and also seems to have a text with 等, and does not problematise it in the way that Woncheuk does, suggest that he only had the one text and it included 等. And this raises the question of why we do not find it in the text attributed to their teacher Xuánzàng. Lusthaus avoids the conclusion from Nattier's study, that the text of T251 was at best edited by Xuánzàng, or more likely by his later students, rather than being a translation he produced.

Finally in relation to Chinese versions corresponding to the Sanskrit passage "cittāvaraṇanāstitvād atrastro viparyāsātikrānto nirvāṇaparyavasānam", which in Chinese becomes:

心無罣礙；無罣礙故，無有恐怖，遠離顛倒夢想 ，究竟涅槃。(T251)

His mind is not obscured, since it is not obscured he is not afraid, far from upside-down dreamlike thinking, he finally attains nirvāṇa. (My translation).

Lusthaus observes that Woncheuk's two texts differ and that Woncheuk favours the one that says 遠離一切顛倒夢想 "far from all upside-down dreamlike thinking." And in this case the T250 has 離一切顛倒夢想苦惱. Lusthaus says "Unfortunately for Nattier's thesis, the alternate version this time is recognisable. It is Kumārajīva's version". Except that it is not. T250 does not include the character 遠 and adds two characters 苦惱. The difference Lusthaus is highlighting involves the interpolation of just two characters, 一切 (literally 'one cut'; figuratively 'all'), so having three other differences is significant. Certainly the two are similar, but then all of these Chinese texts derive have similarities. In fact we have reference to yet another version of the text here which is not the same as either T250 or T251.

One possible good to come out of this is that in looking for parallels in the wider Canon for the last passage, which to my knowledge has not previously been identified with any existing text, we now know to look for alternate readings, though a preliminary search did not turn up any parallels for any of the variants.

Conclusion

On the point about the dating of versions of the Heart Sutra referred to in Woncheuk's commentary we need first to address the issue of "older texts". Crucially, Lusthaus says earlier in his article,

"We have no dates of other background information on when or where the two commentaries were written... We don't know for certain even if these commentaries were written before or after [Xuanzang's] death, though my sense is that they were written after." (2003: 66: emphasis added)

The conjecture by Lusthaus that the commentaries he is discussing were written (i.e. composed) after the death of Xuánzàng is important in assessing his claim that the alternate readings found in them amount to a text from a much earlier period, particularly contemporary with Kumārajīva in the early fifth century.

We've seen that when Woncheuk mentioned old texts" (舊本) he was in fact directly referring to a number of other sutras in which Avalokiteśvara plays a prominent role. So Lusthaus's conclusion that it would be "natural" in this context to conclude that this referred to the Heart Sutra looks wrong. We've also seen that his attempt to connect Woncheuk's text with Kumārajīva fails. Lusthaus's challenge to Nattier's theory falls well short of its mark.

What we're left with is evidence of multiple versions of the Heart Sutra, probably around the time of, or not long after, the death of Xuánzàng. No texts with the readings evinced by Woncheuk, in either Chinese or Sanskrit are extant. Thus there is no good case for pushing back the date of composition of the Heart Sutra before Xuánzàng. On the other hand, the evidence for multiple versions at this time is intrinsically interesting in terms of the history of the text. And in drawing attention to these early commentaries. Lusthaus has made an valuable contribution.

Nattier's thesis on the origins of the Heart Sutra certainly has stronger and weaker points. However, it is beyond reasonable doubt that the Heart Sutra per se began life in China as a compilation of extracts from Kumārajīva's《摩訶般若波羅蜜經》(T233) or possible the commentary on it 《大智度論》and probably other texts including the Mahāmegha Sūtra (possible source of the dhāraṇī). And her arguments about the attribution and dates of T250 and T251 largely stand. Neither seem to be the product of authors to which they were attributed in the 8th Century.

~~oOo~~

Bibliography

Conze, Edward. (1978). The Prajñāpāramitā Literature. Tokyo, The Reiyukai.

Hyun Choo, B. (2006) 'An English Translation of the Banya paramilda simgyeong chan: Wonch'uk's Commentary on the Heart Sūtra (Prajñāpāramitā-hṛdaya-sūtra)' International Journal of Buddhist Thought & Culture. February 2006, Vol.6, pp.121-205.

Lusthaus, Dan. (2003) 'The Heart Sūtra in Chinese Yogācāra: Some Comparative Comments on the Heart Sūtra Commentaries of Wŏnch’ŭk and K’uei-chi.' International Journal of Buddhist Thought & Culture. September, Vol. 3: 59-103.

Nattier, Jan (1992). 'The Heart Sūtra: a Chinese apocryphal text?' Journal of the International Association of Buddhist Studies. 15 (2) 153-223. Online: http://archiv.ub.uni-heidelberg.de/ojs/index.php/jiabs/article/view/8800/2707

Shih, Heng-Ching & Lusthaus, Dan. (2006) A Comprehensive Commentary on the Heart Sutra (Prajnaparamita-hyrdaya-sutra). Numata Center for Buddhist Translation & Research.

Tanahashi, Kazuaki. (2014) The Heart Sutra: A Comprehensive Guide to the Classic of Mahayana Buddhism. Shambhala.

Further Problems in Karma Theory: Continuity and Discontinuity.

Vasubandhu 世親 (via IEP)

Emerging from early attempts at systematisation of Buddhism, sectarian Buddhists, especially those who developed Abhidharma, continued to think carefully about the theories they had inherited they discovered more and more problems. In my recent essays I've discussed a number of problems that these Buddhists inherited and some of the different approaches to the problems.

The picture that we have of Buddhist history was heavily skewed by the destruction of Buddhism in India with the lost of the larger proportion of Buddhist texts. For example we presume there was a Canon of early writings in Sanskrit and one in Gāndhārī and we have lost the bulk of both of them. This sometimes means that we know only a little about the various sects and what we think we know is often from polemics written by their opponents (who were far from scrupulous in representing each other). One result is that we can mistakenly see historical Buddhism as more homogeneous than it in fact was. It's common for modern writers on Buddhism to minimise the heterogeneity of early Buddhism and emphasise the commonality rather than the disputes. I'm interested in exploring just how disparate Buddhist sects were from each other. In some cases the disagreements were extreme. 

The quirks of history can also mean that a text that includes polemic like Vasubandhu's Abhidharmakośabhāṣya can be over-represented in discussions of sectarian Buddhism, or treated as an authoritative account of some long dead sect, simply because it survives in Sanskrit. The Bhāṣya only records one side of a multi-sided argument, and where we do have  more than one side, as for some of the disputes recorded in the Theravāda Kathāvatthu, they often disagree on details. However with care we can get some idea of the issues that Buddhists disputed amongst themselves and some of the parameters of these disputes. Chief amongst these issues was karma. I've already shown that early Buddhist karma theory was incoherent, in the sense that it did not work with pratītyasamutpāda: I've characterised this as the problem of Action at a Temporal Distance. This is not simply my conclusion, it was also the conclusion of one of the founding fathers of Mahāyāna Buddhism, Nāgārjuna in his magnum opus, Mūlamadhyamakārikā (see MMK Chp 17).

In this essay we will see once again that the preconceived metaphysical commitments that Buddhists bring to the dispute circumscribe how the respond to the challenges. Although some authors make a great deal of the Vinaya convention that a schism or sanghabheda is technically a dispute over interpretation of Vinaya, it's clear that doctrinal disputes also caused divisions amongst Buddhists. 

Continuity Problem

Most Buddhist sects agreed that experience consists of a series of mental events (citta) with associated mental concomitants (cetasika). A particular problem is that in the succession of cittas it's hypothetically possible that a kuśala citta might follow akuśala citta or vice versa. One of the important features of karma theory is that results are appropriate to actions. Having good follow directly from evil or vice versa seems on face value to contradict this. Thus it is generally forbidden, axiomatically, for a kuśala citta to follow an akuśala citta and vice versa. Different sects proposed different solutions to this problem.

In order to deal with this problem the Theravādins stipulated that two cittas which are different with respect to kuśala/akuśala cannot directly follow each other. As well as kuśala and akuśala the Theravādins acknowledged that some cittas were indeterminate (avyākṛta). These are cittas that are not associated directly with karma (action) but are either a vipāka (a result from a previous karma) or completely independent of karma (kiriya-citta). The latter category includes some mental states in an arahant such as any accompanied by with two or three of arāga, adveṣa or amoha; and functions such as āvajjana or advertence (turning towards) a sense object. So between two cittas which are different (kuśala/akuśala) a bhavaṅgacitta must intervene. 

Jaini (1959) notes that the Sautrāntikas reject the Theravāda solution because it makes no more sense for a kuśalacitta to be followed by an avyākṛtacitta, because how would an avyākṛtacitta act as a condition for an akuśalacitta? Axiomatically, like must follow like. Indeterminacy is only a fudge, not a solution to this problem. The axiom in fact prevents change, and since we perceive change, the axiom itself must be flawed. And the flaw comes because it is trying to extent continuity beyond the present moment to account for karma. 

Discontinuity Problem

Another problem that concerned later Buddhists is a problem of discontinuity. In the early Buddhist texts there is a state of attainment called 'cessation of mental activity and experience' (sañña-vedayita-nirodha). This state is said, in the Ariyapariyesanā Sutta, to occur after one has master all the jhānas and the arūpāyatanas (sometimes called the arūpajhānas). Not a great deal is said about this state, but if all experience has ceased then it would appear that there is an interruption in the flow of mental events.

In the two essays on Action at a Temporal Distance I covered the basic Theravāda and Sarvāstivāda responses to the problem. The former proposed a series of very short lived cittas which acted as conditions for the next, in up to 24 distinct ways (the 24 paccayas). And they filled the gap between sense-based experiences with the bhavaṅgacitta - a kind of resting state mental activity, the object of which is set at the beginning of one's life by the death moment mental activity (cuticitta) in the dying being and the subsequent relinking mental activity (paṭisandhicitta) in the new being. So a definite discontinuity would be impossible in the Theravāda view.

The Sarvāstivāda response to Action at a Temporal Distance was to propose that if the present vipāka was conditioned by a karma in the past, then the karma must still exist (in some sense) in the past. And if we expect a present karma to have a vipāka in the future then the present karma must still exist (in some sense) in the future. What 'exists' (asti) means here is uncertain, except that we need the karma to be capable of acting as a condition. Karma here explicitly implies the mental activity that motivates the action (cetanā) and thus is used almost synonymously with citta or dharma. If the dharma exists now, in the past and in the future, then it always exists (sarva-asti) hence the name of this sect is sarva-asti-vāda or 'the doctrine of always existing'.

Hence, despite creating two completely distinct metaphysics to explain karma, neither can allow for any kind of discontinuity. A discontinuity in the mental life of the being contradicts the worldview or belief system. And yet on face value sañña-vedayita-nirodha constitutes a discontinuity. And in bridging that conceptual discontinuity, Buddhists seem to be trapped in one or both of two fallacies: they end up creating a continuity with eternalist overtones; or the solution is not internally consistent. In fact there seems to be no way to connect impermanent conditions with effects that manifest long after the conditions have ceased without introducing some incoherence. Either we can have experience arising and passing away in the moment, or we can have actions connected to consequences, but not both. 

~~oOo~~

Anacker, Stefan (1972) 'Vasubandhu's Karmasiddhiprakaraṇa and the problem of the highest meditations.' Philosophy East & West. 22(3): 247-258.

Jaini, Padmanabh. (1959) 'The Sautrāntika Theory of Bīja.' BSOAS 22(2): 236-249.