You are on page 1of 39


The NidÄnasamyukta and the M lamadhyamakakÄrikÄ: understanding the Middle Way through comparison and exegesis
Mattia Salvini

hetukriyapara par ya jÂne tasya na bhotiha astinÂstibhÂv || For one who understands the succession of causes and e ects, There are in this world no such things as ‘exists’ or ‘does not exist’.1

The NidÄnasamyukta is a small fragment of a large non-MahÄyÄna literature once extant in Sanskrit. Some of its recurring themes (catu ko i, dependent arising, wrong views, and emptiness) immediately remind one of Madhyamaka thought; in one of its acceptations, the very word nidÂna is related to the causal links/limbs in the twelvefold formula of dependent arising. In this article, I will try to highlight similarities between the NidÄnasamyukta and the M lamadhyamakakÄrikÄ, suggesting a relation between the two texts; this may in turn help interpret and contextualize some of NÄgÄrjuna’s arguments.2 In particular, it will be useful in understanding the link between more generalized discussions about dependent arising and the speci c concern with rebirth (as represented in the last two chapters of the M lamadhyamakakÄrikÄ). Lastly, I will argue that Candrakîrti is aware of the non-speci cally MahÄyÄna background of the arguments, and is conscientious in representing it faithfully.

1 2

Lalitavistara 25.4cd. I am not implying that the M lamadhyamakakÄrikÄ may be related only to the NidÄnasamyukta as a non-MahÄyÄna source. Compare for example M lamadhyamakakÄrikÄ 24.8 with a passage from the EkottarÄgama: dve satye samup ritya buddhÂn dharmade an | lokasa v tisatya ca satya ca paramÂrthata || 24.8 ||; dve satye aik asya bahukare bhavata | sa v tisatya paramÂrthasatya ca || EkottarÄgama, page 201 in TripÄ hî’s edition.

Thai International Journal of Buddhist Studies II (2011): 57-95. © The International PhD Programme in Buddhist Studies, Mahidol


8/22/2011 4:11:21 PM


TõBS II, 2011 • Articles

1. ‘A Discussion with K¼ty¼yana’, in Sanskrit In M lamadhyamakakÄrikÄ 15.7 we KÄtyÄyanÄvavÄda: nd a reference to the

kÂtyÂyanÂvavÂde cÂstìti nÂstìti cobhayam | prati iddha bhagavat bhÂvÂbhÂvavibhÂvin || In the Discussion with KÄtyÄyana, ‘exists’, ‘does not exist’, or both, was negated by the Blessed one, the one who well observes existence and non-existence. Kalupahana has justly remarked that scholars have somehow disregarded this reference, and has pointed to a corresponding section of the PÄli Canon.3 His suggestions are indeed welcome, and I will try to expand upon his precious insights. I should specify at the outset, though, that I do not subscribe to his conclusions in their entirety – that NÄgÄrjuna’s work is merely a commentary on the KaccÄyanagottasutta and there is nothing speci cally MahÄyÄna-oriented in it.4 I will attempt to o er a di erent interpretation, by taking into greater consideration NÄgÄrjuna’s intended opponents. I shall not try to discredit the (to me convincing) association between NÄgÄrjuna and the MahÄyÄna. Kalupahana argues that NÄgÄrjuna chooses the KaccÄyanagottasutta over the KÄ yapaparivarta for reasons of doctrinal preference: such choice may be better explained, though, by considering his purported opponents (for whom the Ratnak a literature would have had no authority). Moreover, we cannot simply assume that NÄgÄrjuna
3 However, he was not the rst to notice this reference to the PÄli canon. See de La Vallée Poussin’s edition, page 269 note 2, and also M. Sprung, M. (tr.), Lucid Exposition of the Middle Way, Boulder: PrajñÄ Press, 1979, page 159, note 1. De La Vallée Poussin o ers useful comments and references in this regard. Kalupahana contends that Candrakîrti is bringing Buddhism closer to VedÄnta. It will be obvious that I disagree on interpretive grounds; let me add here that the actual popularity of VedÄnta at Candrakîrti’s own time is far from self-evident or easy to assess, which makes Kalupahana’s overall reconstruction slightly less convincing. 4

Kalupahana’s own position is more nuanced, since he proposes a periodization of his own for the history of Buddhist philosophy, sharply distinguishing between ‘earlier’ and ‘later’ MahÄyÄna trends. Here too, I disagree on several accounts: for example, in the VajracchedikÄ, the discussion about how to give rise (utpÂdayitavyam) to the mind (citta) seems very much related to the idea of cittotpÂda (see Bodhisattvabh mi II), meaning the arising of bodhicitta: it is understood in this way by Kamala îla. In order not to excessively burden my arguments, though, I have on occasion glossed over some ner distinctions introduced by Kalupahana. It will also be seen that this article implicitly challenges his periodization of ideas.


8/22/2011 4:11:21 PM

Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ


would have been familiar with the KÄ yapaparivarta as available to us now. If we wish to look for non-MahÄyÄna sources related to NÄgÄrjuna’s thought, the PÄli Canon may not be the best candidate: the very fact that the M lamadhyamakakÄrikÄ is written in Sanskrit suggests as much. Moreover, the background for many of the arguments is closer (although not identical) to VaibhÄ ika Abhidharma rather than TheravÄda Abhidhamma. It would therefore be more sensible to look into texts within a similar fold. Candrakîrti does as much, as he quotes from the KÄtyÄyanÄvavÄdas tra. Moreover, he ends his quote by noticing that ‘this S tra is found in all the NikÄyas’ (ida ca s tra sarvanikÂye u pa hyate). Interestingly, and against the grain of Kalupahana’s overall assessment of his thought, Candrakîrti quotes the KÄtyÄyanÄvadÄna again in the MadhyamakÄvatÄrabhÄ ya, in order to prove that rÄvakas also do realize emptiness (something that had already been pointed out by de La Vallee Poussin).5 Kalupahana’s references to PÄli Suttas are surely not out of place, considering how PÄli and Sanskrit non-MahÄyÄna sources are usually fairly close; he himself mentions the correspondence between the KaccayÄnagottasutta and its Chinese counterpart (translated from the Sanskrit). However, if a relevant non-MahÄyÄna text could be found in Sanskrit, we should give it some attention: this is the extent of my argument in terms of identifying background sources. I am primarily concerned here with how NÄgÄrjuna’s arguments may have been perceived and understood by his prospective audience – who were, after all, recipients of a Sanskrit rather than a PÄli text.6 This will surely help us understand the intended context (and content) of the M lamadhyamakakÄrikÄ. The S tra fragment quoted by Candrakîrti resembles not only the KaccÄyanagottasutta, but also another very similar text, ancient, possibly of wide acceptability amongst NÄgÄrjuna’s opponents, and extant in Sanskrit. The NidÄnasamyukta contains a S tra representing a discussion (avavÂda) between KÄtyÄyana and the Buddha (S tra 19): it is close to the PÄli KaccÄyanagottasutta mentioned by Kalupahana (SamyuttanikÄya XII.15, PTS II.16-17), but, as we will see, it contains

Page 269, note 5 of his edition of the PrasannapadÄ: ‘Dans le bhÄ ya du MadhyamakÄvatÄra (I.6) le KÄtyÄyanas. est allégué en vue de prouver que la connaissance de la ç nyat est commune aux ÇrÄvakas et aux Bodhisattvas.’
6 As pointed out by Peter Skilling, the context and mode of circulation of polemical texts in Medieval India is far from clear to us, and I believe this a ects our perception of the nature and purpose of the texts we are discussing about.



8/22/2011 4:11:21 PM

Annotated Translation of S tras from the Chinese Sa yuktÂgama relevant to the Early Buddhist Teachings on Emptiness and the Middle Way.indd 60 8/22/2011 4:11:22 PM . is itself the evidence. this world is attached to existence and non-existence […]. I am not in a position to assess whether the PÄli and Chinese versions are identical.7: yad bh yas kÂtyÂyana aya loko’stit v abhinivi a nÂstit ca […] For the most. I may have to disagree with Kalupahana’s contention: Nor is there any evidence to support the hypothesis that the “Admonition to KÄtyÄyana” (KÄtyÄyanÄvavÄda) that NÄgÄrjuna was referring to was a version di erent from the KaccÄyanagottasutta found in Pali and Chinese. Choong. The existence of a di erent version. but not identical.8 However. the KaccÄyanagottasutta (or KaccÄnagottasutta) di ers from the version in the NidÄnasamyukta. pp. as Kalupahana implies. that there is evidence to support the hypothesis that NÄgÄrjuna was referring to a di erent version – the one I am going to analyse being one such candidate. See M. 2004. 39-43. 2011 • Articles some signi cant di erences. as I do not read Chinese.7 Unfortunately. it clings to existence and to non-existence […]. clings to two things: for the most. 7 8 Kalupahana. and incidentally. suggests that the Chinese version may be closer to the Sanskrit than to the PÄli version. NidÄnasamyukta 19 presents precisely what NÄgÄrjuna describes as the content of the KÄtyÄyanÄvavÄda: the Buddha explains that both asti and nÂsti are extremes (anta) and that they are not the Middle Path. NidÄnasamyukta 19: dvaya ni rito 'ya kÂtyÂyana loko yad bh yasÂstitÂñ ca ni rito nÂstitÂñ ca […] This world. The initial passage of Candrakîrti’s quotation appears to be related. KÄtyÄyana. closer to the M lamadhyamakÄrikÄ. page 7.60 TõBS II. Johor Bahru: Persatuan Penganut Agama Buddha. This means. KÄtyÄyana. however. TIJBS FINAL VERSION. We may compare the two: PrasannapadÄ on 15. to this text. it also di ers from the KÄtyÄyanÄvavÄdas tra fragment as quoted in Candrakîrti’s commentary. For this reason too. Choong’s translation. ibidem.

Someone. And compare it with the last section of NidÄnasamyukta 19: tat kasmÂdd heto | lokasamudaya kÂtyÂyana yathÂbh ta samyakprajñay pa yato y loke nÂstit s na bhavati | lokanirodha yathÂbh ta samyakprajñay pa yato y loke 'stit s na bhavati | And why is that? Someone. page 5. It does not exist now. See in particular the last verse. explaining their incompatibility with the commonly perceived fact of change (anyathÂtvam). Perhaps this was meant as a conscious hyperbole? TIJBS FINAL VERSION. New York: State University of New York Press. KÄtyÄyana. who sees through perfect knowledge the cessation of the world as it is. does not conceive of existence in respect to the world. the contention that ‘[…] the M lamadhyamakakÄrikÄ […] is a grand commentary to the Buddha’s own KaccÄyanagotta-sutta […]’ is perhaps an exaggerated conclusion. NÄgÄrjuna’s argument in the following verses refutes asti and nÂsti in terms of a wrong view of permanence (  vata) and a view of discontinuity (uccheda) respectively. KÄtyÄyana. nor in the M lamadhyamakakÄrikÄ – the word ‘everything’ (sabbam): sabbam atthìti kho kaccÂyana eko anto | sabba dutiyo anto || 9 nÂtthìti aya David J. who sees through perfect knowledge the arising of the world as it is. however.9 I have anticipated that the PÄli version of the same Sutta contains some signi cant di erences: in fact. NÂgÂrjuna: The Philosophy of the Middle Way: the M lamadhyamakakÂrik of Nagarjuna. discontinuity is entailed.indd 61 8/22/2011 4:11:22 PM . does not not exist: this is permanence. Kalupahana.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 61 The following sections do not correspond. adding something to be found neither in the NidÄnasamyutka version. does not conceive of non-existence in respect to the world. the initial sentences appear to be quite close. 1986. 11: asti yadd hi svabhÂvena na tan nÂstìti  vatam | nÂstìdÂnìm abh t p rvam ity uccheda prasajyate || What exists with svabhÂva. it quali es the basic statements of the Buddha. It does seem like verses 8-11 may be commenting upon this part of the S tra: however. [although] it was before: thus.

I would not subscribe to such line of argumentation. Kalupahana’s ‘milder’ interpretation of NÄgÄrjuna. the analogous passage from the KÄ yapaparivarta does not have an equivalent to the additional quali er sabbam of the PÄli: Ibidem.62 TõBS II. However. and brings the PÄli version further away from the wording of the M lamadhyamakakÄrikÄ (which does not contain them either). and this may be thought to help my argument. In this case. 10 TIJBS FINAL VERSION. rests heavily on this quali ed claim as found in the PÄli KaccÄyanagottasutta.indd 62 8/22/2011 4:11:22 PM . this is the second extreme. due to the way he reads the KaccÄyanagottasutta: at least. commenting and interpreting the Sutta. To say that asti and nÂsti are extremes and bases for attachment can be read as meaning that they may be mere conceptual additions. I do not see. KaccÄyana. On the other hand. It is indeed a feature of the entire M lamadhyamakakÄrikÄ to state negations without quali ers. pages 10-20. an important theme throughout Madhyamaka literature. meaning that some things do indeed exist (but not all do) while other things (but not all) do not exist. but between two commentaries. Incidentally. any textual support for Kalupahana’s claims. we must add this additional consideration that regards coherence of exegesis. according to which he is not refuting the ultimate dharmas. one could argue that the simpler and shorter Sanskrit version may be older than the PÄli. closer to NÄgÄrjuna’s own text and allowing for a stronger interpretation of the ‘extremes’. In other words. If I may put it this way. after all. Secondly. perhaps best exempli ed in the 9th chapter of the BodhicaryÄvatÄra. I am purposely avoiding any consideration of relative chronology.10 However. I nd that such reasoning in favour of textual anteriority is weak and inconclusive. Most importantly. 2011 • Articles Everything exists’. ‘everything does not exist’. Hence. I understand him to take it as a statement about the arising and passing away of dhammas. weakens the force of Kalupahana’s arguments – allowing us to read the M lamadhyamakakÄrikÄ more in line with Madhyamaka commentarial literature. there is no assurance that NÄgÄrjuna and his audience would have necessarily favoured the older version. even if we were to accept such rationale. Kalupahana’s suggestion that we should free ourselves from the commentaries somehow masks the fact that he is also. is indeed one extreme. The existence of a Sanskrit S tra. in the Sutta itself. the additional word sabbam has no corresponding expression in the Sanskrit S tra. apart from the more obvious argument based on NÄgÄrjuna’s language of choice (Sanskrit). but I suspect such a position may carry metaphysical implications (what about the idea of an experiential ‘given’?). Kalupahana seems to consider this a ‘given’ of experience. to say that sabbam atthi and sabba nÂtthi are extremes can be interpreted as a much milder claim. both claiming to be faithful. I nd an additional di culty with Kalupahana’s exegesis. the choice is not between a commentary and a faithful representation.

Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 63 astìti k yapa ayam eko’nta nÂstìty aya dvitìyo’nta yad etayor dvayor antayor maddhyam iyam ucyate k yapa madhyam pratipad dharm  bh tapratyavek  ||11 ‘Exists’. MMK 3. MMK 12. NS 21.6. MMK (M lamadhyamakakÄrikÄ) 12. that is dependent designation.17.11. treating of dependent arising and its various limbs. KÄ yapa. MMK 10. NS 19. the genuine observation of the dharmas. 20.14. NS 18. NS 12. NÄgÄrjuna famously declares the equivalence between dependent arising and emptiness: ya pratìtyasamutpÂda nyat t pracak mahe | s prajñaptir upÂdÂya pratipat saiva madhyam || Whatever is dependent arising is what we call ‘emptiness’. The equivalence of emptiness and dependent arising in the Nid¼nasamyukta and in the M lamadhyamakak¼rik¼ In verse 24.5. this is true of the following: 1. MMK 27. MMK 27. and that itself is the Middle S S tras of the NidÄnasamyukta edited by TripÄ hî. The most extensive and (to my mind) signicant similarities are analysed in this article. Additional similarities in content can be seen between the following: NS (NidÄnasamyukta) 6.19.12 2.4. MMK 17 (especially verse 18).10. are thematically related to chapter 26 of the M lamadhyamakakÄrikÄ out of twenty. since I agree that here the t is a super uous ak ara.2. this is called. 12 Most of the S tras of the NidÄnasa yukta.3. a sentence which.indd 63 8/22/2011 4:11:23 PM . one has ‘MahÄ nyatÄ’ as its theme. MMK 15. MMK 6. KÄsyapa: this is one extreme. such a statement is found in the KÄ yapaparivarta without further quali cations. in turn. To summarize.21. the situation is as follows: the NidÄnasamyukta contains no statement about asti and nÂsti being the two extremes (anta).23. TIJBS FINAL VERSION. NS 15. NS 8.8 of the M lamadhyamakakÄrikÄ. More is to be gained by looking at the NidÄnasamyukta as a whole. Among the twenty. The middle between these two extremes. This sentence is very close to the short passage found in the PÄli KaccÄnagottasutta as quoted above. NS 14.16. NS 22. MMK 12. NS 13. Under the heading of ‘the Teaching of Great Emptiness’ (mah nyatÂnÂmadharmaparyÂya) 11 See pages 25-26 of Vorobyova-Desyatovskaya’s edition. the Middle Way. with the additional quali er sabbam. MMK 12. the PÄli version makes a similar statement. is not at all found in NidÄnasamyukta 19. ‘doesn’t exist’: this is the second extreme. NS 20. I prefer this reading to StaelHolstein bh tapratyavek Ät.22.

emptiness is seen within the context of dependent arising (udaya and vyaya) and the establishment of the Four Noble Truths.1 || If all of this is empty.64 TõBS II. it teaches dependent arising in the context of the Four Noble Truths. No. 2011 • Articles it presents nothing more than the twelve links of dependent arising. within the context of the arising of su ering (samudaya). Lamotte brought to our notice that this text considers the MahÄ nyatÄs tra as one of ‘those S tras of the Tripi aka that formally teach about dharma nyatÂ. in Bulletin of the School of Oriental and African Studies. and in fact. In Honour of Walter Simon (1973). and it teaches dependent arising as the teaching of ‘Great Emptiness’: the same equivalence brought forth by NÄgÄrjuna. 2.’ 14 Peter Skilling. A link between this S tra and the teachings of emptiness (in the Madhyamaka sense of dharmanairÂtmya as well) is found in a text ascribed (by tradition) to none other than NÄgÄrjuna himself: the *MahÄprajñÄpÄramitopade a Ästra. p. 31. 11 (2007): pp.14 13 Etienne Lamotte. The non-existence of the Four Truths of the Noble follows for you.20 || If all of this is non-empty. Vol. within a similar context. In other words. Just as in the Mah nyatÂdharmaparyÂya. To understand that the context is similar. there is no arising and no passing away. 225-247. ‘M gÄra’s Mother’s Mansion: Emptiness and the nyatÂs tras’. preserved in Chinese. Consider in particular the following section. in Journal of Indian and Tibetan Studies. and it may TIJBS FINAL VERSION. The non-existence of the Four Truths of the Noble follows for you. University of London. the path to overcome it (mÂrga). an important article by Peter Skilling has pointed out many of the links which I am here discussing. Verse 20 turns the objection against the opponent: yady a nyam ida sarvam udayo nÂsti na vyaya | cat r Âm ÂryasatyÂnÂm abhÂvas te prasajyate || 24.’13 More recently. 295 b 27-8) range le Maha nyatÂs tra du SamyuktÄgama parmi les quelques S tra du Tripi aka enseignant formellement la dharma nyatÂ. consider that NÄgÄrjuna in this chapter is responding to an initial objection that: yadi nyam ida sarvam udayo nÂsti na vyaya | cat r Âm ÂryasatyÂnÂm abhÂvas te prasajyate || 24. ‘Trois S tra du “Sa yukta” sur la vacuité’. and the nal cessation of su ering (nirodha). 36. page 319: ‘[…] l’Upade a (ch. there is no arising and no passing away. most pertinent to the present study: ‘What is the relation between dependent arising and emptiness? The two are not explicitly identi ed in the PÄli canon.indd 64 8/22/2011 4:11:23 PM .

This identi cation becomes explicit in the Mah nyatÂ-nÂmadharmaparyÂya (above. or at least codi ed or canonized the concept. Lamotte has reconstructed from the Chinese two more S tras of the SamyuktÄgama that deal with emptiness. the features of Madhyamaka philosophy that he dealt with and emphasized are quite di erent from the focus of the present article. canonical. 3. the true understanding of phenomena’ (madhyam pratipad dharm  bh tapratyavek Â).’ (Page 238). However. which are already linked in such texts as the BimbisÂrapratyudgamana-mahÂs tra and its many parallels. Here follows the central. denying four possible alternatives in respect to its production and thus applying the tetralemma in a way analogous (although perhaps not identical). Louis Gomez had also drawn parallels between certain sections of the PÄli SuttanipÄta and Madhyamaka thought. he teaches the dharma.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 65 In the same article. two short s tras unique to the SarvÄstivÄda transmission. avoiding two extremes. which in turn means dependent arising. Again. after all. Gomez. in which each a ga and its cessation is understood to be non-dual. both in content and terminology. colophon to §V) and the ParamÂrtha nyatÂ-s tra. ‘Proto-Madhyamika in the Pali Canon’ in Philosophy East and West. AcelakÄ yapa asks the Buddha about the nature of su ering. also helps understanding verse 21 of the LokÄtîtastava. 26. See: Louis O. which describes a particular interpretation of pratìtyasamutpÂda. The analysis of su ering: Chapter 12 of the M lamadhyamakak¼rik¼ compared to Nid¼nasamyukta 20 A few texts from the NidÄnasamyukta employ the well-known argumentative structure of the catu ko i. Apr. of course. for example the K yapa-parivarta. Vol. No. The identi cation of the middle path with dependent arising is.indd 65 8/22/2011 4:11:23 PM . 2. as we shall see. relevant section from that S tra: have been the SarvÄstivÄdins who rst took the step. or is it arisen without a cause – be it self-production or other-production (asvaya kÂraparakÂrahetusamutpanna )? To all four alternatives. Once again Kalupahana helpfully identi es the relevant sources in PÄli and in Chinese. is it both self-produced and produced from something else (svaya k tañ ca parak tañ ca). 1976: 137-165. In S tra 20. In particular. the Buddha does not deny su ering altogether but. to M lamadhyamakakÄrikÄ Chapter 12. TIJBS FINAL VERSION. The S tra. as for example in the KÂtyÂyana-s tra and other s tras in the NidÄnasa yukta. produced from something else (parak tam). the thematic recurrence of emptiness within the SamyuktÄgama should be taken into account when reading the M lamadhyamakakÄrikÄ. Nonetheless. together with S tra 22. which is. S tra 20 takes up the analysis of su ering. as ‘the middle path. The idea is taken up in Bodhisattva s tras. a natural outcome of the concepts of non-self and dependent arising. according to four alternatives: is it self-produced (svaya k tam). the Buddha states that he gives no answer (avyÂk tam).

is it both self-produced and produced from something else. that when you are asked ‘is su ering selfproduced’ you say ‘I don’t explain in that way’? When asked ‘ is su ering produced from something else. is it arisen without a cause.indd 66 8/22/2011 4:11:24 PM . is su ering produced from something else? By me this also is not declared. Gautama. be it selfproduction or production from something else? By me this also is not declared. Gautama. is su ering both self-produced and produced from something else? By me this also is not declared. is su ering self-produced? By me this is not declared. Gautama. KÄ yapa – that su ering is both self-produced and produced from something else. 2011 • Articles […] kin nu bho gautama svaya k ta du kham | avyÂk tam ida may k yapa svaya k ta du kham | kin nu bho gautama parak ta du kham | etad api k yapa avyÂk ta may parak ta du kham | kin nu bho gautama svaya k tañ ca parak tañ ca du kham | etad api k yapa avyÂk ta may svaya k tañ ca parak tañ ca du kham | kin nu bho gautama asvaya k»raparak»rahetusamutpanna du kham | etad api k yapa avyÂk ta may asvaya kÂraparakÂrahetusamutpanna du kham | kin nu bho gautama svaya k ta du kham iti p o ‘vyÂk tam iti vadasi parak ta svaya k tañ ca parak tañ ca asvaya kÂraparakÂrahetusamutp anna du kham iti p o ‘vyÂk tam iti vadasi | kin nu bho gautama nÂsty eva du kham | na k yapa nÂsty eva du kham iti tv asty eva du kham | sÂdhu me bhavÂn gautamas tath dharma de ayatu yathÂha du kha jÂneya du kha pa yeyam | s eva k yapa vedan sa vettìti yasyaiva syÂt svaya k ta du kham evam aha na vadÂmi | any vedan anyo vettìti yasyaiva syÂt parak ta du kham evam aha na vadÂmi | vedanÂbhibh tasyaivÂsata pare du kha samavadadhatìti yasyaiva syÂt svaya k tañ ca parak tañ ca du kham evam apy aha na vadÂmi | satsu ca pratyaye u du khÂni santìti yasyaiva syÂd asvaya kÂraparakÂrahetusamutpanna du kham evam aha na vadÂmi | ity etÂv ubhÂv antÂv anupagamya madhyamay pratipad tathÂgato dharma de ayati | yad utÂsmin satìda bhavaty asyotpÂdÂd idam utpadyate p rvavad yÂvat samudayo nirodha ca bhavati | […] Gautama.that su ering is self-produced. How is it. KÄ yapa – that su ering is produced from something else. is su ering arisen without a cause.66 TõBS II. KÄ yapa – su ering arisen without a cause. be it self-production or production from something else’ you say TIJBS FINAL VERSION. KÄ yapa . be it self-production or production from something else. Gautama.

Gautama. it is good. that exists. that su ering doesn’t even exist: su ering indeed exists. Respected Gautama. as before.1 || svaya k ta yadi bhavet pratìtya na tato bhavet | skandhÂn imÂn amì skandh sa bhavanti pratìtya hi || 12.4 || parapudgalaja du kha pare a k tv tad du kha yadi yasmai pradìyate | sa du khena vin kuta || 12. I may see su ering.indd 67 8/22/2011 4:11:24 PM . Feeling is something else from the one who feels: as for one who thinks in this way that su ering is produced from something else – I don’t say it to be so. di erent types of su ering are there: as for one who thinks in this way that su ering arises without a cause.5 || parapudgalaja du kha yadi ka parapudgala | vin du khena ya k tv parasmai prahi oti tat || 12. When the conditions are there. KÄ yapa. that arises. Feeling and the one who feels are the same: as for one who thinks in this way that su ering is self-produced – I don’t say it to be so. Not approaching either of these two extremes.6 || TIJBS FINAL VERSION. be it self-production or production from something else – I don’t say it to be so. su ering doesn’t even exist? It is not. may the Dharma be taught to me is such a way that I may know su ering.3 || svapudgalak ta du kha yadi du kha punar vin | svapudgala sa katamo yena du kha svaya k tam || 12. the TathÄgata teaches the Dharma according to the Middle Way. up to ‘arising and cession occurs’ […] Compare this with M lamadhyamakakÄrikÄ 12: svaya k ta parak ta dv»bhy» k tam ahetukam | du kham ity eka icchanti tacca kÂrya na yujyate || 12.2 || yady amìbhya ime ‘nye syur ebhyo vÂmì pare yadi | bhavet parak ta du kha parair ebhir amì k t || 12. The su ering of someone who is not even overcome by feeling descends upon another: as for one who thinks in this way that su ering is both self-produced and produced from something else – I don’t say it to be so either. from the arising of this. That is: this being there.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 67 ‘I don’t explain in that way’? Is it that.

Indeed. because it is not produced by he himself. some wish su ering to be so. 12. by whom su ering is produced? 12. who is its person.10 || na kevala hi du khasya cÂturvidhya bÂhyÂnÂm api bhÂvÂn cÂturvidhya 12.8 || syÂd ubhÂbhy par»k»r»svaya k ta k»ra du kha syÂd ekaikak ta yadi | du kham ahetuka kuta || 12. 12. produced from something else. Self-produced. or produced from no cause. 12.7 || na tÂvat svak ta du kha na hi tenaiva tat k tam | paro nÂtmak ta cet syÂd du kha parak ta katham || 12.5.2. If these were other from those. If su ering is born from some other person. that su ering would be self-produced.68 TõBS II.4. When what is self-produced is not established. where is he apart from su ering? 12. these aggregates come about in dependence upon those aggregates. it would then not have come about in dependence upon something.8.7. apart from su ering.9 || na vidyate | na vidyate || 12. who gives su ering to someone else after having produced it? 12. If it were self-produced. the one to whom that su ering is given after having being produced by someone else. 2011 • Articles svaya k tasyÂprasiddher du kha parak ta kuta | paro hi du kha yat kuryÂt tat tasya syÂt svaya k tam || 12.indd 68 8/22/2011 4:11:24 PM . su ering is not self-produced. who is the other person. or those were other from these. Then. TIJBS FINAL VERSION.3. If su ering is born from some other person. but that does make sense as an e ect. 12. apart from su ering. from both.6. the other one who would produce su ering. for him. how could su ering be produced from something else? Because. su ering would be produced from something else: these would be produced from those others.1. If su ering is produced by one’s own person.

Comparing the text of the M lamadhyamakakÄrikÄ with the S tra helps us to understand Candrakîrti’s glosses for the terms du kha and asvaya kÂraparakÂrahetusamutpanna jarÂmara am (NidÄnasamyukta 6. asvaya kÂraparakÂrahetusamutpanna sukhadu kham (22.23). asvaya kÂraparakÂrahetusamutpanna sukhadu kham (22. authorship. but always resort to nobhayak ta when denying the third ko i. twice). twice). or title. They contain the negation of ‘produced from oneself’ (svaya k ta) and ‘produced from another’ (parak ta). only twice. asvaya kÂraparakÂrahetusamutpanna du kham (20. if it could be produced from each taken severally. and there also. Su ering could be produced from both.6. asvaya kÂraparakÂrahetusamutpanna nÂmar pam (6.9. Practically the same expressions appear in what VV Gokhale called the *Madhyamaka Älistambas tra.7).’ TIJBS FINAL VERSION. two MahÄyÄna texts closely related to Madhyamaka.16b). for external things also these four modes are not found. how could su ering be produced by something else? 12. 34 (BodhicaryÄvatÄrapañjikÄ page 481).10. without a cause? 12. It is not only for su ering that these four modes are not found. asvaya kÂraparakÂrahetusamutpanna nÂmar pam (6. twice).7. refuting the fourth alternative in the catu ko i. asvaya kÂraparakÂrahetusamutpanna sukhadu kham (22.indd 69 8/22/2011 4:11:25 PM .13). it is not found even once in the Älistambas tra or the *Madhyamaka Älistambas tra.10. twice).15 On the other hand. asvaya kÂraparakÂrahetusamutpanna sukhadu kham (22. I must thank Peter Skilling for pointing out that the short text is in fact ‘a Âstra of unknown origin.9. How could there be su ering neither from other-production nor from self-production.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 69 If the other is not self-produced. asvaya kÂraparakÂrahetusamutpanna sukhadu kham (22.10). asvaya kÂraparakÂrahetusamutpanna du kham (20. in the same context as in NÄgÄrjuna’s text. twice).9 contains the expression parÂkÂrÂsvaya kÂra which seems as close as one could get (within anu ubh metre) to asvaya kÂraparakÂra: the latter compound appears 17 times in the NidÄnasamyukta.16 Even these alternative expressions appear with far less frequency than we nd them in the NidÄnasamyukta. asvaya kÂraparakÂrahetusamutpanna vôñÂnam (6. 16 15 sa cÂyam a kuro na svaya k to [na parak to] nobhayak to ne varanirmito […] Âlistambas tra.9. asvaya kÂraparakÂrahetusamutpanna du kham (20. sa ca nÂmÂr p kuro na svaya k to na parak to nobhayak to ne varanirmito […] Älistambas tra.11). 14 (see BodhicaryÄvatÄrapañjikÄ page 578). Verse 12.15).

amongst NÄgÄrjuna’s opponents. they further qualify the last ko i with the expression ‘at random’ (adhiccam).’ (bìj kuragha apa Âdaya ). Please consider how this too corroborates my overall claim that MÄdhyamikas. sprouts. On the other hand. respectively). etc. we can also notice that the very last verse brings NÄgÄrjuna’s arguments closer to the Älistambas tra. This suggests that Vasubandhu. moreover. which shows NÄgÄrjuna’s treatment to be one step removed from his non-MahÄyÄna sources and one step closer to the discussion of dependent arising in the Älistambas tra. just like other Buddhist philosophers. in part.25cd and commentary (which will be discussed later in this article). he states that ‘external things too’ refers to such things as ‘seeds. since he extends the argument to the bÂhyapratìtyasamutpÂda – usually (as far as I know) not directly discussed in the PÄli Suttas or in the NidÄnasamyukta. 17 TIJBS FINAL VERSION.2 and the nal section of the S tra. in the narrative). both texts refer to the dependent arising of the aggregates (verse 12. where the NidÄnasamyukta has ‘cause’ (hetu): The Abhidharmako abhÄ yam accepts that pratìtyasamutpÂda can either pertain to sentient beings (sattvÂkhya ) or two things that are not sentient beings (asattvÂkhya ). The PÄli version di ers from the Sanskrit in several respects (even. 2011 • Articles bahyÂnÂm api bhÂvÂnam (verse 10). seed and sprout are the initial limbs of what is elsewhere (in the Älistambas tra) termed ‘external’ (bÂhya) dependent arising. discuss dependent arising primarily within the context of the twelve limbs and hence.indd 70 8/22/2011 4:11:25 PM . Vasubandhu however mentions that ‘in the S tra we nd only the one pertaining to sentient beings’ (s tre sattvÂkhya eva) and sets to explain why. something to take into account when we wish to assess the currency that this S tra may have had a few centuries earlier.70 TõBS II. we can nd an important feature. and secondly. they rest upon an analysis of the agent of experience (yo vetti in the S tra. However. This is in line with the idea of understanding the arising of su ering as the internal (ÂdhyÂtmika) dependent arising. Candrakîrti speci es that ‘su ering’ refers to the ‘ ve aggregates of clinging’ (pañcopÂdÂnaskandh du kham ity ucyate). In his introductory commentary to verse 1. pots and clothe textiles. the equation put forth by S tra 20. in Abhidharmako abhÄ ya 3. rebirth. Two di erences are here to be noted. In this particular instance. and the pudgala in the M lamadhyamakakÄrikÄ). This lends some additional credit to Candrakîrti’s identi cation of the Älistambas tra as a primary exegetical source for the M lamadhyamakakÄrikÄ. introducing verse 10. We can further detect several analogies in the arguments employed: in both cases. in the context of the Abhidharmako abhÄ ya. did not consider the Älistambas tra as authoritative for his audience. the Buddha’s answers do not contain the expression avyÂk tam (m heva kassapÂti bhagav avoca as opposed to avyÂk ta may k yapa […]).17 I have highlighted (in bold) the terminological similarities between the two texts.

arisen fortuitously? NidÄnasamyukta: kin nu bho gautama asvaya kÂraparakÂrahetusamutpanna du kham18 Gautama. Compare this with LokÄtîtastava 21: svaya k ta parak ta dvÂbhy k tam ahetukam | tÂrkikair i yate du kha tvay t kta pratìtyajam || Self-produced. they substitute sukhadu kha for du kha. without a cause? The next two S tras (21 and 22) of the NidÄnasamyukta are also closely related to the same topic.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 71 SamyuttanikÄya: kimpanabhogotamaasaya kÂramapara kÂramadhiccasamuppannam dukkhanti | Gautama. TIJBS FINAL VERSION. it would be produced from each taken severally. 18 Leon Feer. The second point was indeed implied even in S tra 20. from both. is it that su ering is not self-produced.indd 71 8/22/2011 4:11:26 PM . however. is it that su ering is arisen from neither self-production nor production from something else as a cause? We can see how the wording of the NidÄnasamyukta is closer to NÄgÄrjuna’s own (which does not include yad ccha): syÂd ubhÂbhy k ta par»k»r»svaya k»ra du kha syÂd ekaikak ta yadi | du kham ahetuka kuta || 12.) after denying that su khadu kha corresponds to any of the four alternatives. here it is explicitly stated. wherein kÂra is dropped in case of the fourth ko i. the editor of the Pali Text Society. produced from something else.) the questions about the nature of su khadu kha are asked by non-Buddhist ‘wanderers’ (anyatìrthikaparivrÂjak ). This would bring the text even further away from NÄgÄrjuna’s verse.9 || 12. How could there be su ering neither from other-production nor from self-production.9. however. the Buddha tells Ãnanda that it is dependently arisen (pratìtyasamutpannam). If su ering were produced from both. records an alternative reading. produced from without a cause. not produced from something else. S tra 22 adds two important elements: i. ii.

There is good justi cation. Incidentally.indd 72 8/22/2011 4:11:26 PM . tatra katama p rvÂnta katamo’parÂnta katha na pratisaratìti tad vyutpattyartham Ârabhate | 19 Chapters 2 to 12 all begin with atrÂha. anyone familiar with the narrative presented in NidÄnasamyukta 22 would have recalled that narrative. this time not even using atrÂha. 2011 • Articles Su ering is thus interpreted by the speculators. Although we may wish to generalize the sense of tÂrkikai for philosophical reasons. Dependent arising and wrong views: A comparison between S tra 14 and the last two Chapters of the M lamadhyamakak¼rik¼ Between Chapter 26 and Chapter 27 of the M lamadhyamakakÄrikÄ we nd a progression from the twelve links of dependent arising to the elimination of wrong views. Chapter 13 introduces atrÂha shortly after Verse 1. I think. Even Chapter 26 begins with atrÂha. rather than an objection. 4. he a rms dependent arising. others hold on to one thing or the other.e. NÄgÄrjuna refutes them all. i. but in this case. lastly. introducing an objection. but it was explained by you to be born dependently.72 TõBS II. Chapters 14 to 25 again all begin with atrÂha. the Buddha denies them and. Chapter 27 introduces once again a question. it introduces a question about dependent arising. this is also akin to the overall structure of the M lamadhyamakakÄrikÄ. now (beginning of 26) Candrakîrti tells us that the opponent is asking about dependent arising.19 However. This may suggest a progression of sorts: that at this point the hypothetical opponent has been ‘convinced’ by the argument in favour of emptiness and now wishes to know more about its nature. this may puzzle some readers. he quotes a source wherein dependent arising and the relinquishing of views are expressly linked: ya caiva pratìtyasamutpÂda yathÂbh ta pa yati sa na p rvÂnta pratisarati nÂparÂnta pratisaratìtyadi s tre pa hyate. since it di ers from the rationale employed in all the other chapters to shift from one subject matter to the next. and lastly (chapter 26). since that source presents the same exact elements: others expect one of the four alternatives. Candrakîrti explains the transition from one topic to the next through the logic of contextually relevant objections and rebuttals. for saying that the transition from 25 to 26 and the transition from 26 to 27 are interpreted di erently from all the preceding instances. TIJBS FINAL VERSION. in this instance he suggests that the shift of topic is derived from continuity of exegesis. he a rms the workings of the twelve limbs of pratìtyasamutpÂda. Since this was earlier identi ed as dependent arising. In most instances.

past lives (atìtÂtmabhÂva). does not run after the prior limit. does not run after the ulterior limit’ etc. what is the ulterior limit. For stylistic reasons nv aham abh vam atìte ‘dhvani Âhosvin nÂbh vam atìte’dhvani ko nv aham abh vam atìte’dhvani katha nv aham abh vam atìte ‘dhvani | apar»nta v punar na pratisarati . the actual quote appears at the end of Chapter 27. I believe this should be the rst candidate. taken to be a commentary.indd 73 8/22/2011 4:11:26 PM . Moreover.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 73 ‘One who in this way sees dependent arising as it is. In that respect. which implies that the following section of the M lamadhyamakakÄrikÄ is. Kalupahana) in identifying a source other than the KaccÄyanagotta/ KÄtyÄnÄvavÄda as the background for Chapter 27. The important expression here is ‘he begins in order to explain that’ (tad vyutpattyartham Ârabhate). asatas tucchato riktato’sÂrata rogato ga ata alyato’ghato’nityato du khata nyato’nÂtmata ca samanupa yati. how does one not run after them? Thus. considering that the following would have been a very bulky introductory text: ya ka cid bhadanta Âriputra ima pratåtyasamutp»da bhagavat samyakpra ìtam eva yath»bh ta samyakprajñay satatasamitam ajìva nirjìva yathÂvadaviparìtam ajÂtam abh tam ak tam asa sk tam apratigham anÂvara a ivam abhayam anÂhÂryam avyayam avyupa amasvabhÂva pa yati. Candrakîrti explains that p rvÂnta means past identities. Candrakîrti is stating that since the topical progression in the S tra source is from the twelve limbs to the elimination of views. We may now look at possible sources which may corroborate Candrakîrti’s contention. I would side with Candrakîrti (perhaps vs. – this is found in the S tra. NÄgÄrjuna is following the same steps. Âhosvin na bhavi yÂmy anÂgate ‘dhvani. wrong views about rebirth. sa na p rv»nta pratisarati . it was quite reasonable on the part of Candrakîrti to abbreviate. what is the prior limit. hence we can read his quotation as an abbreviated form of the text immediately contiguous to the section of the Älistambas tra last quoted in Chapter 26. after all. Considering that the S tra quoted at the end of Chapter 26 is in fact the Älistambas tra. Candrakîrti may be quoting from memory and paraphrasing. The very fact that Candrakîrti may have explained this concluding section in this vein should be a good index of the intended topical emphasis in nv aha bhavi yÂmy anÂgate ‘dhvani. katha nu bhavi yÂmy anÂgate ‘dhvani | TIJBS FINAL VERSION. he begins in order to explain that. Please note: the ‘views’ to be eliminated are. ko nu bhavi yÂmy anÂgate ‘dhvani. in fact.

and the verse which it is commenting upon. 2011 • Articles Please notice how the initial sentence in Candrakîrti’s commentary matches with the highlighted sections of this part of the S tra. through perfect wisdom. I will later argue that Candrakîrti understands the rest of Chapter 27 according to the same principle. It would be di cult on the other hand to try to establish a similar relation between NÄgÄrjuna’s root text and the Älistambas tra – there is no sign of such a link anywhere else in the M lamadhyamakakÄrikÄ. wherein dependent arising and views are directly connected to the process of rebirth. at the end of section 5 of this article. by a learned Noble rÄvaka.indd 74 8/22/2011 4:11:27 PM . TIJBS FINAL VERSION. A very similar passage is quoted in Abhidharmako abhÄ ya 3.25: yata ca bhik avo bhik u  pratìtyasamutpÂda ca pratìtyasamutpann ca dharm eva yathÂbh ta samyakprajñay d  bhavanti | sa na p rvÂnte pratisarati ki nv aham abh vam atìte’dhvanìti vistara || I discuss the previous part of this commentary. Since I have shown strong similarities between the latter and the NidÄnasamyukta.74 TõBS II. then he does not run after the prior limit: did I exist in the past? Or did I not exist in the past? Who was I in the past? How was I in the past? 20 21 For an English translation of this passage. I believe that the following section of S tra 14 does: yata ca rutavatÂrya rÂvake a pratìtyasamutpÂda ca pratìtyasamutpann ca dharm yathÂbh ta samyakprajñay sud  bhavanti | sa na p rvÂnta pratisarati | kin nv aham abh vam atìte ‘dhvani | aho svin nÂham atìte ‘dhvani | ko nv aham abh vam atìte ‘dhvani | katha nv aham abh vam atìte ‘dhvani | aparÂnta v na pratisarati | kin nu bhavi yÂmy anÂgate ‘dhvani | aho svin na bhavi yÂmy anÂgate ‘dhvani | ko nu bhavi yÂmy anÂgate ‘dhvani | katha nu bhavi yÂmy anÂgate ‘dhvani |21 When dependent arising and dependently arisen dharmas are well seen.20 Candrakîrti’s commentary therefore links Chapters 26 and 27 to the Älistambas tra. according to the way things are. it may be sensible to ask whether any part of the NidÄnasamyukta shows a progression comparable to the transition from Chapter 26 to Chapter 27. page 70. see Ross Reat.

and extends its concern to the removal of all views: yÂni tÂny ekatyÂn rama abrÂhma Ân p thalloke d igatÂni tadyath ÂtmavÂdapratisa yuktÂni sattvavÂdapratisa yuktÂni jìvavÂdapratisa yuktÂni kot halama galavÂdapratisa yuktani tÂny asya tasmin samaye prahì Âni bhavanti parôñÂtÂny ucchinnam lÂni tÂlamastakavad anÂbhavagatikÂny ÂyatyÂm anutpÂdadharm i || 14.29 || Or.30 || He taught the Saddharma for the sake of relinquishing all views.12 || tat kasmÂd dheto | tath hi rutavatÂrya rÂvake a pratìtyasamutpÂda ca pratìtyasamutpann ca dharm yathÂbh ta samyakprajñay sud  suju  suvidit supratividdh || 14. NÄgÄrjuna speaks of getting rid of all views: athav sarvabhÂvÂn nyatvÂc ch vatÂdaya | kva kasya katam kasmÂt sambhavi yanti d aya || 27. Even the progression to the last two verses of Chapter 27 (verses 29 and 30) nds a parallel in the structure of S tra 24. where. similarly. the latter text starts by explaining the twelve limbs of dependent arising and then passes on to say that if one understands dependent arising one overcomes views. follows from the refutation of those views related to the p rvÂnta and aparÂnta. The S tra’s concluding section. which. would views like ‘permanent’ and so forth come about? sarvad iprah Âya ya saddharmam ade ayat | anukampÂm upÂdÂya ta namasyÂmi gautamam || 27.indd 75 8/22/2011 4:11:27 PM . for whom.13 || TIJBS FINAL VERSION.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 75 Nor does he run after the ulterior limit: will I exist in the time to come? Or will I not exist in the time to come? Who will I be in the time to come? How will I be in the time to come? The parallel between chapters 26 and 27 and S tra 14 works very well. since as in those chapters (and as in the Älistambas tra). out of compassion: I bow to that Gautama. because of the emptiness of all things. and due to what reason. After analysing the speci c views that he represented in a catu ko i format.

TIJBS FINAL VERSION. Candrakîrti’s comments and perceptive use of quotations turn out to be very helpful. Here it is important to emphasize how the sections of the Älistambas tra quoted by Candrakîrti are very close to non-MahÄyÄna materials.ÂtmavÂdapratisa yuktÂni sattvavÂdapratisa yuktÂni jìvavÂdapratisa yuktÂni pudgalavÂdapratisa yuktÂni kautukama galavÂdapratisa yuktÂni unmiñjitÂni nimiñjitÂni ca.linked to a view of self. cut at the root. are in him destroyed. Candrakîrti may have had good reasons to employ the Älistambas tra rather than the NidÄnasamyukta. judging from his use of the Älistambas tra. well known. It is even uncertain whether 22 For an English rendering. On the other hand. since the material from the Älistamba is very similar to that of the NidÄnasamyukta (and to much non-MahÄyÄna material more broadly). linked to a view of a sentient being. at that time.76 TõBS II. not only from the NidÄnasamyukta. This part of S tra 14 is in turn close (almost verbatim) to the portion of the Älistambas tra quoted at the end of Candrakîrti’s commentary. well penetrated. well practiced. but also from the PÄli (SamyuttanikÄya and MajjihimanikÄya). for whom the NidÄnasamyukta would have most likely been more authoritative. page 72. they have the quality of not arising in the future. NÄgÄrjuna’s kÂrikÂs seem to be addressed to readers. linked to a view of a soul. they will not come into existence. linked to a view of auspicious rites – those. tadyath . What is the reason for that? It is because. perfectly cognized. see Ross Reat. Nonetheless. It seems that Candrakîrti hinted at the same idea. by a learned Noble rÄvaka. 2011 • Articles Whichever views in this ordinary world may exist for rama as or BrÄhma as . Considering his intended audience.indd 76 8/22/2011 4:11:27 PM . dependent arising and dependently arisen dharmas are well seen. through perfect wisdom. tÂny asya tasmin samaye prahì Âni bhavanti parôñÂtÂni samucchinnam lÂni tÂlamastakavad anÂbhÂsagatÂni ÂyatyÂm anutpÂdÂnirodhadharm i ||22 The structure of chapters 26 and 27 of the M lamadhyamakakÄrikÄ might have been in uenced by its sources of reference. following the quote reproduced above and presenting an identical progression: yÂni eke  rama abrÂhma Ân p thagloke d igatÂni bhavi yanti. according to the way things are. like the top of a palm tree.

The next eight views are the counterpart of the rst eight. A closer analysis of chapter 27 is now necessary to that end. etc.. 5. Basic structure of M lamadhyamakak¼rik¼. when he introduces this section by saying that: TIJBS FINAL VERSION. Eight of these views. As Candrakîrti puts its.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 77 Candrakîrti himself would have considered the Älistambas tra as a speci cally MahÄyÄna text: in any case. the rst four refer to oneself (Âtmana ). while the second set of four present a generalised question (sÂmÂnyena). than Kalupahana would make him to be. Verses 14-20 analyse the four views which refer to oneself and are based on the nal limit . though. My di erences from previous scholarly interpreters of the D iparìk  partly stem from a di erent view about its structure: I will thus try to group the verses according to my understanding.indd 77 8/22/2011 4:11:28 PM .whether I will exist. unambiguously. this is to say that the commentator is in fact much more aware of the non. that in both cases the limit referred to is. Candrakîrti makes this clear. Verses 1 and 2 delineate sixteen basic wrong views (d i) regarding the continuity of the person from one birth to the next. the next or nal limit (hence. I have hinted at the fact that these comparisons may help assessing the intended context of some of NÄgÄrjuna’s arguments. future lives). or not. Again. Verses 3-13 analyse the rst four views. which are based on the prior limit. mentioned in the rst verse. when referred to the future. Notice. Chapter 27 We may gain a better understanding of the sense and the central concerns of the concluding chapter of the M lamadhyamakakÄrikÄ by outlining its basic structure. we are obviously dealing with a temporal limit. depend on a prior limit (which Candrakîrti identi es with previous lives): -did I exist in the past? ki nv aham abh vam atìtam adhvÂnam -did I not exist in the past? nÂbh vam atìtam adhvÂnam -did I both exist and not exist? abh va ca nÂbh va ca -did I neither exist nor not exist? naivÂbh va na nÂbh vam The other four extend the same question of temporal continuity to the whole world (loka). o ering an interpretive scheme that I think makes the text coherent and in line with Candrakîrti’s comments. a temporal limit.(speci cally) MahÄyÄna background of these chapters. in the future.

78 TõBS II. present and future lives. with a verse of homage to the Buddha.1 (the quotation also corroborates my contention). Verse 29 is a general statement about emptiness. Candrakîrti’s introductory remark is quite clear: idÂnìm antavattvam antÂnantÂdicatu ayam aparÂnte yath na sa bhavati. I would argue. while verse 30 concludes both the chapter. The link between the two (Âtman. 23 I am referring especially to Gar eld’s own commentary to this section of the MMK. namely the four views about the temporal limit of this world. as well as the whole work. he denies the (four views) based on the nal limit. by linking the aggregates to di erent realms of existence. Thus. TIJBS FINAL VERSION. idÂnìm aparÂnta sam ritasya prati edham Âha […] Thus. Again. Candrakîrti concludes his commentary by quoting the long passage of the Âlistambas tra which. 2011 • Articles tad eva p rvÂnta sam ritasya d icatu ayasya asambhavam udbhÂvya. In brief. as well as the continuity of one’s world. loka) is highlighted by the fact that NÄgÄrjuna examines even the views regarding the limits of the ‘world’ by referring them back to the aggregates. Rather.23 The text gives no indication of shifting away from the analysis of temporal continuity. he says […] As will be seen. one’s world (will the world/my world exist?). Verses 21-28 examine the other four views resting on a nal limit. showing how the four views of a limit or its lack are impossible in reference to the nal limit. I would read the entire chapter as analysing views speci cally regarding rebirth: the nature of one’s continuity. tath pradipÂdayann Âha […] Now.indd 78 8/22/2011 4:11:28 PM . Candrakîrti elaborates on the same lines. he had summarized in the introductory commentary to 26. this is where my reading (and. one might either ask oneself questions regarding one’s own personal continuity (did/will I exist?) or generalise the doubt by including one’s environment. I have argued. a di erent emphasis is introduced: while thinking of past or future lives. I suggest that the intended emphasis on temporality and rebirth has been (occasionally) missed by Gar eld’s exegesis. Candrakîrti’s) di ers from Gar eld’s. between past. who taught the saddharma for the sake of relinquishing all views. having in this way demonstrated the impossibility of the four views based on the prior limit.

future and present lives is highlighted by Vasubandhu in the Abhidharmako abhÄ ya. Verse 1 Views resting on the prior limit: 1a.indd 79 8/22/2011 4:11:28 PM . commenting on kÂrik 3. I existed in the past 2a. I did not 3a. 4b Discussion of views 5b. I will neither exist nor not exist 5b.24: p rvÂparÂntamadhye u sammohaviniv ttaye || For the sake of removing delusion about the prior and nal limit. I will exist in the future 2b. 2a. I will not 3b.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 79 The following table clari es the structure of the chapter as I have just explained it. The world has a prior limit 6a. TIJBS FINAL VERSION. It neither exists nor does it not exist Verse 2 Views resting on the nal limit: 1b. 4a Discussion of views 1b. I neither existed nor did not exist 5a. I both did and did not 4a. It both does and does not 8a. The world has a future limit 6b. 7b. and about the middle. 8b Conclusion Please also notice that the link between understanding dependent arising and removing delusions about the past. I will both exist and not exist 4b. It does not 7a. 2b. It neither exists nor does it not exist Verses 3-13 Verses 14-20 Verses 21-28 Verses 29-30 Discussion of views 1a. It does not 7b. 3a. 6b. It both does and does not 8b. 3b.

an imputation of either identity or di erence between one life and the next. and I believe more coherent. the Älistambas tra and the M lamadhyamakakÄrikÄ.e. about one’s existence in past and future lives. Consider furthermore that Vasubandhu introduces his verse by asking ‘for what purpose in the S tra there is [dependent arising] only as pertaining to sentient beings?’ (kim artha puna s tre sattvÂkhya eva). Even according to the interpretation of previous scholars. This section of the BhÄ ya bears many similarities to NÄgÄrjuna’s PratîtyasamutpÄdah dayavyÄkhyÄna .24 Although a few centuries divide NÄgÄrjuna from Vasubandhu. Gods. 24 tatra p rvÂntasammoho yata iya vicikits ko nv aham abh vam atìte’ dhvani Âhosvin nÂbh va katha nv abh vam iti | aparÂjtasa moho yata iya vicikits ki nu bhavi yÂmy anÂgate’ dhvanìti vistara | madhyasa moho yata iya vicikits ki svid ida ke santa ke bhavi yÂmaiti. though. and this is where I intend to support a di erent.80 TõBS II.17 run as follows: sa deva sa manu ya ced eva bhavati  vatam | anutpanna ca deva syÂj jÂyate na hi  vatam || 15 || devÂd anyo manu ya ced a  vatam ato bhavet | devÂd anyo manu ya ced santatir nopapadyate || 16 || divyo yady ekade a syÂd ekade a ca mÂnu a | a  vata  vata ca bhavet tac ca na yujyate || 17 || Gar eld translates the verses (from the Tibetan) as follows: 15. A few verses. 6. humans and the meaning of paraloka Not all the verses of this chapter are problematic. This amounts to noticing the privileged thematic link between dependent arising and the analysis of sentient processes .i. wherein the delusion to be counteracted is. it is not unlikely that they may have drawn from a similar and possibly related exegetical tradition about the twelve a gas of dependent arising.15 and 26. once again. most of them have to be explained in terms of rebirth.indd 80 8/22/2011 4:11:29 PM . 2011 • Articles Vasubandhu’s commentary quotes the very same questions found in the NidÄnasamyukta. have been interpreted without reference to rebirth. Verses 26. rebirth. If a human were a god. On such a view there would be permanence. exegesis. TIJBS FINAL VERSION.

15. it seems that the god he is referring to is not quite the impermanent. whose in nite life is to be contrasted with man’s mortality (as Gar eld points out. This way of translating would t the diachronic sense of the statement. If the human were di erent from the god. TIJBS FINAL VERSION. the di culty is between reconciling the contradicting attributes which would pertain to man (impermance) and god (permanence). Gar eld writes of the ‘divinity’ that ‘engenders its eternality’. his analysis misses the diachronic element of the argument. In verse 26. The Fundamental Wisdom of the Middle Way. It would be both permanent and impermanent.indd 81 8/22/2011 4:11:29 PM . In this reading. like ‘the same’: if the same god is that very man. That would be irrational. mode of existence intended by Buddhist authors in general and here. which could be rendered in English by using a quali cation.the same reasoning could apply to the 25 Jay L. On the other hand. If one part were divine and One part were human. the repetition of the pronoun sa is likely to have an emphatic function. In his commentary to this verses.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 81 The god would be unborn. A continuum would not be tenable. the most plausible sense of these verses is that permanence or impermanence refers to the identity or di erence between one mode of existence and the next. For any permanent thing is unborn. On such a view there would be impermanence. Moreover. I think. It may rather be the eternal God. Gar eld. which is about the impossibility of identifying the succeeding modes of birth (human and godly) as being the same person. 1995. 16. On the other hand. If a human were di erent from a god. partly due to some confusion in interpreting the word deva. etc. New York: Oxford University Press. 17. albeit favourable. by NÄgÄrjuna as well. The choice of the human and god realm is incidental . a translation with a determinate article (‘the god’ instead of ‘a god’) would better mark the fact that we are not talking of an abstract concept of divinity.25 Gar eld’s translation is not far from literal sense of the Sanskrit (apart from some additions). Gar eld’s wording turns the kÂrik into a general statement about the relationship of the human and the divine. yet I feel it misses the emphasis due to some choices of sentence structure. much like in the JudeoChristian tradition).

indd 82 8/22/2011 4:11:29 PM . for what reason? Since that. 2011 • Articles problem of dying as a human and being reborn as an animal. wherein the problem of future birth is explicitly mentioned. and I must thank him for such courtesy. Thus. after performing wholesome deeds in this world. permanence makes no which case. to read the verse without disrupting the continuity of the argument. but between identity and di erence within a continuum of di erent births. and an unborn god makes no sense. is not born. which is permanent.26 I propose.82 TõBS II. The di erence from Gar eld’s reading may be summarised in two points: deva here refers to the god 26 adhvany anÂgate ki nu bhavi yÂmìti dar anam | na bhavi yÂmi cety etad atìtenÂdhvan samam || M lamadhyamakakÄrikÄ. the god would not be born. which reads the kÂrik as referring to rebirth: iha hi ka cin manu yagatistha ku ala karma k tv devagati gacchati | tatra yadi sa eva deva sa eva manu ya iti evam ubhayor aikya syÂt tad  vata syÂt | na caitad eva yad deva eva manu yo bhaved iti | ato nÂsti ki cic ch vatam | api ca  vatavÂde sati asamutpanna ca deva syÂt | ki kÂra am? yasmÂj jÂyate na hi  vatam | yadd hi vastu  vatam tad vidyamÂnatvÂn naiva jÂyate | tata ca anutpanno deva syÂt anutpanno devo na yujyate iti | eva tÂvac ch vata na yujyate c|| Someone placed in the human realm. Therefore. 27 In all fairness and to his credit. but I believe the sample just quoted should be su cient to support the interpretation that I have proposed. instead. there would be permanence. Therefore.14. Permanence and discontinuity are the unwanted consequences stemming from attributions of identity and di erence respectively. Gar eld’s reading does not link verse 15 with the preceding one. one may want to consider Candrakîrti’s commentary to the same verse. And it is not the case that the very god could be the man. there is nothing permanent. within a view of permanence. which is permanent. That entity. the dialectic is not between permanence and impermanence as mutually exclusive attributes of man and divinity.27 Moreover. since it is already there. Candrakîrti’s commentary to verses 16 and 17 is along the same lines. if that very god is the same one who was a man. there would be identity of the two . In this context. In other words. I should note that Jay Gar eld readily admitted the inaccuracy of his rendering of deva through personal communication. TIJBS FINAL VERSION. is not born. Moreover. reaches the god realm. 28. the god could not be born.

The sense of loka. which he (correctly) interprets as ‘focusing on the temporal limits of the world’. TIJBS FINAL VERSION. I read verses 21 and 22 as a continuous argument. therefore. my world. Verse 21 reads: antavÂn yadi loka syÂt paraloka katha bhavet | athÂpy anantav loka paraloka katha bhavet || The sense of the word paraloka may be (mis)taken to refer to another world. In this case. Kalupahana too moves away from a diachronic interpretation. moreover. thus possibly bringing the M lamadhyamakakÄrikÄ in line with Kant’s antinomies. etc.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 83 realm. paraloka. which refer more speci cally to a person when the quali cation upÂdÂnaskandha is added. verse 22 becomes internally inconsistent. The rst is that the two Sanskrit terms loka. which is no other than rebirth. amutra. in the end. depending on the situation of my (re)birth. and the dialectics of permanence and impermanence refer to the problem of identity and di erence between two successive births. oscillates between ‘world’ and ‘person’. How could there be another world? If the world were unlimited. This reading. There are two major di culties with Gar eld’s rendering. How could there be another world? In his commentary. the next life (even in modern Indian languages). the latter being possibly referred to as loka. the question addressed by this verse would be one of spatial in nitude. Another term closely connected to rebirth is the Sanskrit paraloka. and the link between the two is probably that this/the next world is. as Gar eld’s translation (and commentary) suggests: If the world were limited. He also clearly distinguishes the issue at hand in this verse from the discussion introduced in verse 22. This avoids reading the three verses as a digression from the central topic of the chapter. the one I favour. Kalupahana interprets them correctly. in the Buddhist sense of an impermanent state of existence. Gar eld clari es that NÄgÄrjuna is pointing out the di culties in speaking of a world beyond the world. If we don’t understand the world loka in this way. commonly refer to this life.. like atra. A more technical explanation is that all compounded dharmas are subsumed within the ve skandhas.indd 83 8/22/2011 4:11:30 PM . in the sense of something beyond the spatial limits of the present universe. and this explains why the continuum of the aggregates is employed in verse 22 to refer to loka: the skandha-santÂna is the proper basis for imputing a person (loka) within a world.

refers to the temporal limits of this life (loka) versus the next (paraloka). the PrasannapadÄ speaks of a temporal succession from the aggregates of a human existence (manu ya-skandh ) to the aggregates corresponding to a birth in the god realm (deva-gati-upapatti-sa g hìt ). Candrakîrti’s commentary supports my interpretation: yadi hy antavÂn vin Âd paraloko na syÂt | rdhva p rvaloko na syÂt. What the reasoning implies is that the temporal continuity between one life and the next is what makes it impossible to speak of loka as having limits. then the next life could not occur. thus an occurrence to be delimited in time: this or that life. The nondestruction of the rst aggregates is the reason for the impossibility of this succession. 28 De la Vallée Poussin (page 587.indd 84 8/22/2011 4:11:30 PM . showing how neither having limits nor not having them is possible for a person. – vinÂçÂd rdhvam = zhig nas phyis. TIJBS FINAL VERSION.84 TõBS II. This is again clari ed by his introductory commentary upon verse 22: idÂnìm antavattvam anantavattva ca ubhayam etal lokasya yath na sambhavati. skandhÂn) in the verse. The expression ‘prior to its destruction’ (vin Âd rdhvam) leaves little doubt that we are dealing with temporal limits. in my understanding. notices that ‘p rva manqué dans le tibétain. It does imply that the Tibetan version of Candrakîrti’s commentary (that Gar eld may have most probably consulted) is slightly less explicit about the sense of loka and paraloka in this passage. he says […] Moreover. which would thus make the person’s world (loka) limited. note 3). When commenting upon the double occurrence of the word ‘aggregates’ (skandh . Candrakîrti takes verse 22 as the explanation of the assertions contained in verse 21 regarding the impossibility of ascertaining the temporal limits between two lives.28 This passage comments directly on verse 21. Verse 21. tad If in fact the previous life had a limit prior to its destruction. Candrakîrti’s commentary to verse 23 clari es that loka must refer to one state of existence. 2011 • Articles has the advantage of not disrupting a continuous line of argumentation. tath pratipÂdayann Âha […] Now. Once again. The presence of p rva in the Sanskrit version rules out any reading of the passage as referring to the world and the ‘world beyond the world’.’ I thank Peter Skilling for bringing this to my attention.

The rest of the passage also refers to birth (placing the semen) and the intermediate state. The pratisa dhi-vôñÂna is the consciousness corresponding to the third of the twelve links in the cycle of dependent origination (sa skÂra-pratyaya vôñÂnam). which I delineated at the beginning of this section. nevertheless. At this juncture. since we negate that we will go to another world after this life (lokÂntara-gamanakriyÂ) and that we have come to this life from another world (lokÂntara-ÂgamanakriyÂ). views are treated contiguously to the twelve limbs of dependent arising: the 29 30 Page 6 in Tatia’s Edition. A very clear instance is AbhidharmasamuccayabhÄ yam (xxv-xxix):29 nÂsty aya loko nÂsti paraloko nÂsti mÂt nÂsti pit nÂsti sattva upapÂduka iti kriyÂpavÂda . and here loka and paraloka are explicitly linked to the action (kriyÂ) of going and coming. the Älistambas tra as well as the NidÄnasamyukta o er some further clues. both sense and context link paraloka to a process between lives. which is a diachronic process. Here the BhÄ yam simply states that. disrupting its basic structure. perhaps more signi cantly. Gar eld correctly interprets them according to their diachronic sense. which in the YogÄcÄra tradition is distinguished from the consciousness of the fourth link (vôñÂna-pratyaya nÂmar pam).Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 85 The same considerations apply to all the text up to verse 28. of the action of placing semen. breaking up the ow of the argument (against Candrakîrti’s reading) and. a father does not exist. because of the denial of the action of going to and coming from another world. a mother does not exist. its present environment) is not considered in his commentary. the next world does not exist. lokÂntaragamanÂgamanakriyÂy bìjÂdhÂnakriyÂy pratisa dhibandhakriyÂy cÂpavÂdÂt | ‘This world does not exist. the intended meaning of the world loka (that it actually refers to this life. TIJBS FINAL VERSION. there is no being not born from a womb’. This in turn obscures the centrality of the question of rebirth for the whole chapter. and of the action of tying to the co unction (pratisa dhi). Just as in the M lamadhyamakakÄrikÄ. Pratisa dhi is the juncture between one life and the next. such is denial in respect to action. Thus. taken to be the ÂlayavôñÂna instead. we incur over-denial. if we deny this and the next world.30 It seems di cult to interpret the passage in any other sense other than referring to rebirth. Usages outside the M lamadhyamakakÄrikÄ con rm the interpretation of loka/paraloka as this/next life.indd 85 8/22/2011 4:11:30 PM .

when consciousness conglomerates in the mother’s womb. verse 2).indd 86 8/22/2011 4:11:31 PM . upapattya ikacitta and so forth make it unreasonable to read the text otherwise. who is it that transmigrates from this world to the next world? He says: from this world to the next world. na tat kalala kalalatvÂya sa varteta’ iti vacanÂt || Thus. tad nÂmar paprÂdurbhÂvo na syÂt. If in this realm of existence consciousness does not conglomerate. If it is beyond.thus from the utterance (of the Buddha). In brief. rather than ‘other’ in the sense of ‘beyond’ the limits of this world. 2011 • Articles latter relate speci cally to rebirth. such an indea of a world beyond this world is. as far as I know. alien to ancient South Asian cosmology and philosophy. then. TIJBS FINAL VERSION. Not only is this comprehensible within the context of the S tras – there is su cient textual evidence to show that NÄgÄrjuna himself understood them this way. not ever a subtle atom transmigrates. k arati prÂdurbhavatìtyartha | yadi iha gatau vôñÂna na sa m rchita syÂt. name and form.86 TõBS II. then the foetus would not become a foetus’. Candrakîrti is even more speci c: tad eva bimbapratibimbÂdinyÂyena mÂtu kuk au vôñÂne sa m rcchite vôñÂnapratyaya nÂmar pa ni icyate. consciousness were not to enter the mother’s womb. meaning that it pours forth or manifests. which contrast laukika. in analogy to such things as an image and its re ection. tarhi asmÂl lok»t ka paraloka sa krÂmati | Âha | asmÂl lokÂt paraloka s k mo’ ur api na sa krÂmati | If so. NÄgÄrjuna explicates the link between dependent arising and rebirth in the PratîtyasamutpÄdah dayavyÄkhyÄna. conditioned by consciousness. Ãnanda. The rst piece of evidence that I may adduce is from Chapter 26 of the M lamadhyamakakÄrikÄ. Usages like pratisandhi. It is clear that here loka and paraloka have to do with successive realms of rebirth. The short text contains in particular the following passage: yady evam. is infused. it not a world: think of the expressions lokÂtìta and lokottara. then there would be no manifestation of name and form: ‘If. the term para in the expression paraloka is always found to mean ‘next’ in the sense of temporal succession. Apart from the M lamadhyamakakÄrikÄ. ‘ sa ced Ânanda vôñÂna mÂtu kuk i nÂvakrameta. NÄgÄrjuna writes that consciousness enters a realm of existence (vôñÂna sannivi ate sa skÂrapratyaya gatau. Indeed.

indd 87 8/22/2011 4:11:31 PM .25. and quotes a S tra to corroborate his claim. This in turn has facilitated understanding the structure and intended emphasis of the last two chapters of the M lamadhyamakakÄrikÄ. even the PratîtyasamutpÄdah dayavyÄkhyÄna can be said to have the same structure. the understanding of the way things are is equated with the removal of those views. various views about past and future lives are rejected. Furthermore. and hence. by highlighting some textual and conceptual similarities that suggest NÄgÄrjuna’s familiarity with some of the S tras from the NidÄnasamyukta (or with some alternative texts that may have resembled them). one could compare this with Vasubandhu’s commentary to Abhidharmako a 3. Conclusions Lamotte had perceptively noticed a link between the NidÄnasamyukta (rather. In other words. both in relation to the analysis of dependent arising (chapter 26) and wrong views (27) is not at all odd TIJBS FINAL VERSION. the Älistambas tra and the last two chapters of the M lamadhyamakakÄrikÄ: dependent arising is explained in terms of its twelve limbs. the SamyuktÄgama as a whole) and the Madhyamaka philosophy of emptiness. the connection between comprehending dependent arising and doing away with wrong views about rebirth seems to be a recurrent feature of medieval Buddhist thought. where he very explicitly connects the teaching of dependent arising in twelve parts to the removal of wrong view in respect to past lives. or future lives. The ‘Abhidharmic’ emphasis on rebirth in the M lamadhyamakakÄrikÄ. the present life. rebirth. I have also shown where my proposed interpretation di ers from those of Kalupahana and Gar eld. This emphasis has been identi ed with linking dependent arising and the elimination of views to an analysis of rebirth and its process.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 87 We can see a very similar topical progression in the NidÄnasamyukta (S tra 14). If we restrict the views as referring to the identity or di erence between the previous life and the next. in line with Candrakîrti’s commentary but using a di erent source as a point of reference. and NÄgÄrjuna’s exposition in the M lamadhyamakakÄrikÄ is consistent with that tradition (as Candrakîrti’s commentary also con rms). I have tried to show that the connection may be signi cant. 7.

must also be taken into account . It is indeed sensible for NÄgÄrjuna to draw from texts that his own opponents would have considered authoritative. saw a similar link between NÄgÄrjuna and the non-MahÄyÄna sources. Empty words: Buddhist philosophy and cross-cultural interpretation.32 Kalupahana is perceptive 31 It should be clear by now that my position is antithetical to Gar eld’s contention that rebirth is of little importance to the MahÄyÄna. and that it constitutes the unwanted heritage of a ‘pseudo-Ätman’. See Jay L. I think this is very much in line with the interpretation I have proposed. a great Chinese scholar of Madhyamaka from the twentieth century. An important speci cation: I do not think that my arguments about the NidÄnasamyukta in any way support the idea of NÄgÄrjuna being a non-MahÄyÄna philosopher. bondage and liberation. Oxford: Oxford University Press. are an interdependent pair – hence. TIJBS FINAL VERSION.31 Lastly. since he believed that the M lamadhyamakakÄrikÄ solved the purpose of reconciling the emptiness of the PrajñÄpÄramitÄ with the teachings of the Ãgama (i. objects and cognition. such reading. I agree with Kalupahana’s remark that one should read NÄgÄrjuna more closely against the background of non-MahÄyÄna sources. that has to do with its emphasis on dependent arising and/as dependent imputation. In other words. 2011 • Articles or inconsistent with MahÄyÄna thought. I would say that textual evidence shows rebirth to be a central concern for the MahÄyÄna (think of YogÄcÄra!) and in the case of Madhyamaka. when r pa itself depends on the intersection of di erent notions. Gar eld. though. The idea that matter could pre-exist and be ontologically more fundamental than notions is rather inconsistent with Buddhist thought as a whole – not only the Madhyamaka. o ering a plausible alternative to Kalupahana’s approach.88 TõBS II. and in that respect the NidÄnasamyukta seems a much better candidate than the PrajñÄpÄramitÄs tras.indd 88 8/22/2011 4:11:31 PM . This is su cient to account for the lack of direct appeal to MahÄyÄna sources in the M lamadhyamakakÄrikÄ. I di er in regards to the rationale behind. no doubt. and therefore. 32 Stefania Travagnin (personal communication) pointed out that Yin Shun. 2002. I cannot here dwell too extensively on the possible philosophical reasons for such an emphasis on rebirth in Madhyamaka. This link. especially page 84. It may have to do with upholding that jñeya and jñÂna. and the conclusions to be drawn from. mental events are the locus of su ering.e. however. is the larger horizon for most Madhyamaka re ection. which. non-MahÄyÄna S tras). if dependent arising works for non-sentient things (bÂhya) it must also work for sentient events (ÂdhyÂtmika). The fundamental idea that prajñaptis (notions/names) weave the fabric of the whole world.since prajñaptis are after all conscious events. are bound within causal regularities. It would be contradictory for a Madhyamaka to reduce consciousness to an accident of r pa. which is again a process within consciousness. was not exclusive. and is very well understood and highlighted by Candrakîrti’s commentary.

the larger context of MahÄyÄna literature certainly helps to understand NÄgÄrjuna’s thought: take for example NÄgÄrjuna’s extension of the catu ko i to what is bÂhya – which resembles the Älistambas tra and has no parallel in the NidÄnasamyukta.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 89 in pointing out that MahÄyÄna literature (like the KÄ yapaparivarta) retains interest in the twelve limbs of dependent arising while expounding emptiness: even the Suvar aprabhÄsas tra treats of the twelve limbs in its chapter on nyatÄ. in particular. one may quote verse 27 of the LokÄtîtastava: animittam anÂgamya mok o nÂsti tvam uktavÂn | atas tvay mÂhÂyÂne tat sÂkalyena dar itam || ‘Without reaching the signless. you have shown that in its entirety. thus have you explained. there is no liberation’. as opposed to late. since even a late commentator like PrajñÄkaramati quotes extensively from the Älistambas tra in his exposition of Madhyamaka. To show that a few short texts from a large non-MahÄyÄna corpus do speak of emptiness is far from implying that therefore the MahÄyÄna S tras are redundant in that respect. Here the most important word is sÂkalyena. Therefore. TIJBS FINAL VERSION. The treatment of emptiness in the NidÄnasamyukta is cryptic and limited when compared to the MahÄyÄna S tras. This feature is not unique to early. in the MahÄyÄna. MahÄyÄna.indd 89 8/22/2011 4:11:32 PM . as I think Kalupahana implies. For those who accept its attribution to NÄgÄrjuna.

KÄtyÄyana.indd 90 8/22/2011 4:11:32 PM . S tra 19: A Discussion with KÄtyÄyana bhagavÂn nÂdikÂy viharati guñjakÂvasathe | athÂyu mÂn sandhÂkÂtyÂyano yena bhagav s tenopajagÂma | upetya bhagavatpÂdau iras vanditvaikÂnte 'sthÂt | ekÂntasthita Âyu mÂn sandhÂkÂtyÂyano bhagavantam idam avocat | samyagd i samyagd ir iti bhadanta ucyate | kiyat samyagd ir bhavati | kiyat tathÂgata samyagd ti prajñapayamÂna prajñapayati | evam ukto bhagavÂn Âyu manta sandhÂkÂtyÂyanam idam avocat | dvaya ni rito 'ya kÂtyÂyana loko yad bh yasÂstitÂñ ca ni rito nÂstitÂñ ca | upadhyupÂdÂnavinibaddho 'ya kÂtyÂyana loko yad utÂstitÂñ ca ni rito nÂstitÂñ ca | etÂni ced upadhyupÂdÂnÂni cetaso ‘dhi hÂnÂbhinive Ânu ayÂn nopaiti nopÂdatte nÂdhiti hati nÂbhinivi aty Âtm meti | du kham idam utpadyamÂnam utpadyate | du kha nirudhyamÂna nirudhyate | atra cen na k k ati na vicikitsati | aparapratyaya jñÂnam evÂsya bhavati | iyat kÂtyÂyana samyagd ir bhavati | iyat tathÂgata samyagd i prajñapaymÂna prajñapayati | tat kasmÂd dheto | lokasamudaya kÂtyÂyana yathÂbh ta samyakprajñay pa yato y loke nÂstit s na bhavati | lokanirodha yathÂbh ta samyakprajñay pa yato y loke 'stit s na bhavati | ity etÂv ubhÂv antÂv anupagamya madhyamay pratipad tathÂgato dharma de ayati | yad utÂsmin satìda bhavaty asyotpÂdÂd idam utpadyate | yad utÂvidyÂpratyay sa skÂr iti p rvavad yÂvat samudayo nirodha ca bhavati | asmin khalu dharmaparyÂye bh yam a Âyu mata sandhÂkÂtyÂyanasyÂnupÂdÂyÂsravebhya citta vimuktam | [English translation:] The Blessed One was dwelling at NÄdikÄ in the brick residence. the Long Lived SandhÄkÄtyÄyana approached the place where the Blessed One was. KÄtyÄyana. 2011 • Articles Appendix NidÄnasamuykta. right view. clings mostly to two things: it clings to existence and non-existence. he saluted the two feet of the Blessed One with his head. while conceiving of right view?’ Having been addressed thus. This world. Then. If one does not approach these appropriations of bonds. the Blessed One said the following to the Long Lived SandhÄkÄtyÄyana: ‘this world. being the mind’s resolution. it is said right view. then. he remained on one side. How is this right view? How does the TathÄgata conceive. attachment and propensity. is bound by bonds: it clings to existence and nonexistence. Placed on one side. if he does not TIJBS FINAL VERSION. Having come near. the Long Lived SandhÄkÄtyÄyana said the following to the Blessed One: ‘Bhadanta.90 TõBS II.

this arises. Just as this method of Dharma was being spoken of. I also thank Giuliano Giustarini. And why is that? Someone. this occurs. does not conceive of existence in respect to the world. the mind of SandhÄkÄtyÄyana was released from all the out ows of appropriation. who looked at the article’s draft and helped improving it. nor becomes resolved or attached in terms of ‘self’ or ‘mine’. formations have ignorance as their condition. If about this he does not speculate or doubt. who sees through perfect knowledge the cessation of the world as it is. Not approaching either of those two extremes. Part of the research connected to this article has been kindly supported by Mahidol University. surely he has knowledge independent of others. KÄtyÄyana. is in this way. this su ering arises while arising. as said earlier. TIJBS FINAL VERSION.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 91 appropriate them. does not conceive of nonexistence in respect to the world. and thus. KÄtyÄyana. KÄtyÄyana.indd 91 8/22/2011 4:11:32 PM . the TathÄgata teaches the Dharma through the Middle Path: this being there. this su ering ceases while ceasing. Acknowledgements I must thank Peter Skilling for o ering very useful suggestions and for the many pro table discussions in the past year. In this way the TathÄgata conceives while conceiving of right view. due to the arising of this. Right view. who sees through perfect knowledge the arising of the world as it is. Someone. Any mistakes or imperfections are of course my own fault. up to ‘there is arising and cessation’.

). Reinbek: Verlag für Orientalistische Fachpublikationen. 1976.indd 92 8/22/2011 4:11:33 PM . Part II: NidÂna Vagga.). 1995.). 2002. (ed. pages 18-22. Patna: K.P. (ed. 1981. (ed. Tokyo: Soka University. Varanasi: Bauddha Bharati. Sa yutta-NikÂya. Pema (ed. pages 16-18.) Abhidharmako a. The Pali Texts Society.) Abhidharmasamuccayabh yam. S. 1989. Sa yutta-NikÂya.92 TõBS II. ÁryaprajñÂpÂramitÂvajracchedikÂs tram eva ÂcÂryakamala ìlaviracit ÁryaprajñÂpÂramitÂvajracchedik ìkÂ. Jayaswal Research Institute. Oxford. The K yapaparivarta. 1994. The Pali Texts Society. M. (ed. (ed. EkottarÂgama Fragmente der Gilgit-Handschrift. Oxford. 2011 • Articles Bibliography Primary sources: Abhidharmako abh¼ ya Dwarikadass. »ryaprajñ¼p¼ramit¼vajracchedik¼ æk¼ Tenzin. TIJBS FINAL VERSION. Part II: NidÂna Vagga. K¼ yapaparivarta Vorobyova-Desyatovskaya. L.). Ekottar¼gama Tripathi. and tr. Abhidharmasamuccayabh¼ ya Tatia. Acelasutta Feer. (ed. L. N.I. 1989. Sarnath: CIHTS.). Kacc¼nagottasutta Feer. C.

BodhicaryÂvatÂrapañjikÂ.L. Fünfun wanzig S tras des NidÂnasa yukta (Sanskrittexte aus den Turfanfunden VIII). ÁcÂryanÂgÂrjunapra ìta PratìtyasamutpÂdah daya eva ÁryadharmadhÂtugarbhavivara a. Jayaswal Research Institute. Berlin: Akademie-Verlag. (ed. 1958.). M lamadhyamakak¼rik¼ (NÄgÄrjuna) and Prasannapad¼ (Candrakîrti) de La Vallée Poussin. 1961. 1997.). TIJBS FINAL VERSION. (ed. P. 1978. L. Lalitavistara. (ed.L.). (ed. Darbhanga: Mithila Insitute. 1902.). C. 1962. *Madhyamaka ¼listambas tra Vaidya. 1913. M lamadhyamakakÂrikÂs de NÂgÂrjuna avec la Prasannapad Commentaire de Candrakìrti.P.indd 93 8/22/2011 4:11:33 PM . Darbhanga: Mithila Institute. Bodhisattvabh mi Dutt. (ed.). MahÂyÂnas trasa graha .).). Pratætyasamutp¼dah dayak¼rik¼vy¼khy¼na Jamchen Namdol. P. ÃcÄrya (ed. St.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 93 Nid¼nasamyukta TripÄthî. Bodhicary¼vat¼rapañjik¼ De la Vallée Poussin. Petersbourg: Académie Impériale des Sciences. Lalitavistara Vaidya. Sarnath: Central Institute for Higher Tibetan Studies. Bodhisattvabh mi . Nalinaksha (ed. L. Patna: K. Calcutta: Asiatic Society of Bengal.

)... 1995. Jul. review: ‘Dogmatische Begri sreihen im älteren Buddhismus. 1982. 1976: 137-165. Darbhanga: Mithila Institute. 2004.. TripÄ hî’. in: Journal of the American Oriental Society. Conze. Ia. No. New York: State University of New York Press. Empty words: Buddhist philosophy and cross-cultural interpretation. 1965: 463. and tr.Sep. Johor Bahru: Persatuan Penganut Agama Buddha. Jay L. Da ottaras tra IX-X by D. 85. Suvar aprabhÂsas tra.. Gar eld. 1967..464. The Banarsidass. ‘Proto-Madhyamika in the Pali Canon’ in Philosophy East and West. Vol. New York: Oxford University Press. Kalupahana. (ed. Schlinglo Fünfun wanzig S tras des NidÄnasa yukta by C. Delhi: Motilal Secondary literature: Choong. 2.).. Christian.94 TõBS II. TIJBS FINAL VERSION. 2011 • Articles Lok¼tætastava (NÄgÄrjuna) Lindtner. 1993. Oxford: Oxford University Press. Nagarjuniana: Studies in the Writings and Philosophy of Nagarjuna. L. 1986. No. E. NÂgÂrjuna: The Philosophy of the Middle Way: the M lamadhyamakakÂrik of Nagarjuna. Apr. Copenhagen: Institut for indisk lologi.indd 94 8/22/2011 4:11:33 PM . 2002. Suvar aprabh¼sas tra Bagchi.O. David J. N. Annotated Translation of S tras from the Chinese Sa yuktÂgama relevant to the Early Buddhist Teachings on Emptiness and the Middle Way. The fundamental wisdom of the Middle Way. Gomez. 3. . 26. ¼listambas tra Ross Reat. (ed. M. Jay L. S. Vol. Gar eld. Âlistambas tra.

Skilling. P. M.). Vol. University of London. 20. ‘Emptiness in the PÄli Suttas and the Question of NÄgÄrjuna’s Orthodoxy’. 2007: 225-247. in: Bulletin of the School of Oriental and African Studies. 1979.. 2. Number 4. 1973:.indd 95 8/22/2011 4:11:34 PM . ‘M gÄra’s Mother’s Mansion: Emptiness and the nyatÂs tras’. No. In Philosophy East & West Volume 55. in: Bulletin of the School of Oriental and African Studies. 1/3. Studies in Honour of Sir Ralph Turner. Boulder: PrajñÄ Press. in Journal of Indian and Tibetan Studies.. 2005: 507-528. Sprung. A. (tr. 1957: 569-579. Waldschmidt. 313-323. ‘Trois S tra du “Sa yukta” sur la vacuité’. Ernst. University of London. 1937-57. ‘S tra 25 of the NidÄnasa yukta’.Salvini • The NidÄnasamyukta and the M lamadhyamakakÄrikÄ 95 Lamotte. E. Velez de Cea. No. TIJBS FINAL VERSION. 36. 11. Lucid Exposition of the Middle Way. In Honour of Walter Simon. Vol. Director of the School of Oriental and African Studies.