A SKETCH ON NĀGĀRJUNA'S PERSPECTIVES ON "RELATION"^*

ABSTRACT

The aim of this paper is to provide a sketch on the way Nāgārjuna deals with the idea of 'relation'. The concept of 'relation' as expressed in the Pāli sources is here theoretically systematized according to three patterns: 1. (onto)logical, 2. strictly subordinative existential, 3. non-strictly subordinative existential. After having discussed Nāgārjuna's acceptance and treatment of these three patterns, particular attention is paid to the non-strictly subordinative existential relation. This kind of relation is meant to describe the way the factors of the conditioned co-origination are linked to each other and is exemplified by Nāgārjuna by means of the father-son bond. A possible way to explain the conditioned co-origination doctrine in the light of the father-son example is here suggested by having resource to the 'Cambridge change' theory. Even if in the Pāli Canon the non-strictly subordinative existential pattern is said to apply to all the other factors of the conditioned co-origination, there is no direct evidence that it concerns also the avidyā-saṃskāras link. It will be shown how Nāgārjuna, by applying it to the avidyā-saṃskāras link, seems to introduce a new perspective in the conditioned co-origination theory.

Keywords
Nāgārjuna; Pāli Canon; Relation; pratītyasamutpāda; Cambridge change

RESUMO

O objetivo deste artigo é proporcionar um esboço da maneira na qual Nāgārjuna aborda a ideia de 'relação'. O conceito de 'relação' que encontramos nas fontes pāli é aqui sistematizado teoricamente conforme três padrões: 1. (onto)lógico, 2. existencial estritamente subordinativo, 3. existencial não estritamente subordinativo. Após ter discutido a recepção e o tratamento nagarjunianos desses três padrões, este estudo dedica atenção especial à relação existencial não estritamente subordinativa. Esse tipo de relação visa descrever a maneira na qual os fatores da cooriginação condicionada estão ligados entre eles, e é tipificada por Nāgārjuna pelo exemplo da ligação pai-filho. Para explicar a doutrina da cooriginação condicionada à luz do exemplo pai-filho, um modo possível aqui sugerido é por meio do recurso à teoria do 'Cambridge change'. Embora no cânone pāli se diga que o modelo de relação existencial não estritamente subordinativa se aplique a todos os outros fatores da cooriginação condicionada, não há nenhuma evidência direta de que ele concerna também à ligação avidyā-saṃskāras. Será mostrado como Nāgārjuna, ao aplicar esse modelo à ligação avidyā-saṃskāras, pareça introduzir uma nova perspectiva na teoria da cooriginação condicionada.

Palavras-chave
Nāgārjuna; Cânone pāli; Relação; pratītyasamutpāda; Cambridge change

If we consider such a series as that of events, and if we refuse to allow absolute time, we shall have to admit three fundamental relations among events, namely, simultaneity, priority, and posteriority.¹

1 Introduction: preliminary remarks on 'relation' within the Buddhist framework

The aim of this article is to provide an overview on the treatment of the concept of 'relation' in Nāgārjuna's (first-second centuries CE) philosophy. In particular, the discussion will focus on the kind of relation Nāgārjuna theoreticized between the members of the pratītyasamutpāda. Before doing that, I will begin by briefly taking into consideration the ideas concerning the relation as they can be inferred from the Pāli Canon.

As it is well-known, the philosophical concept of relation in Buddhism is mainly conveyed by the term paṭiccasamuppāda in Pāli, pratītyasamutpāda in Sanskrit, and in Tibetan rten ciṅ 'brel bar 'byuṅ ba (or, in short, rten 'brel), a compound that can be translated with "conditioned/dependent co-origination".² In the Pāli Canon, the concept on which the paṭiccasamuppāda relation is grounded is generally summarized by the sentence: imasmiṃ satīdaṃ hoti imass'uppādā idam uppajjati ("Being this, there is that, from the origination of this, that originates").³ Interestingly enough, there are at least two 'negative' counterparts of this positive assertion. The first, and perhaps more obvious, one is expressed in terms of: imasmiṃ asatīdaṃ na hoti imassa nirodhā idaṃ nirujjhati ("Not being this, there is not that, from the cessation of this, that ceases").⁴ The second one, on the other hand, can be inferred from sentences as the following one: upadhīnan tveva asesavirāganirodhā natthi dukkhassa sambhavo ("But from the complete desireless extinction of nescience there is no appearance of pain")-the same is repeated for avijjā ("nescience"), saṅkhāras ("pre-forming impulses"), viññāṇa ("consciousness"), phassa ("contact"), vedanā ("sensation") and all the other factors of the paṭiccasamuppāda. The cessation of these factors, one by one, leads to the non-origination of pain.⁵ We are thus in front of at least three different canonical approaches to 'relation' for as far as paṭiccasamuppāda is concerned. In order to conceptually systematize these three approaches, I propose to analyze them as follows:

1. The positive "existing this, that exists" and the negative "not existing this, that does not exist".

2. The positive "when this originates, that originates" and the negative "when this ceases, that ceases".

3. The positive "when this originates, that originates", and the negative "when this ceases, that does not originate".

Relation (1) refers to an (onto)logical and perfect bilateral bond, since it involves the aspect of existence and/or non-existence (P. and S. sat/asat, T. yod/med) of the related elements. Relations (2) and (3) describe a kind of existential bond, since they are both grounded on the idea of origination and/or cessation (P. uppāda/anuppāda, S. utpāda/anutpāda, T. 'byuṅ ba/ma 'byuṅ ba). Moreover, relation (2) depicts a strictly subordinative (hence, unidirectional) existential link, whereas relation (3) is a non-strictly subordinative existential kind of bond.

In what follows we will take into consideration Nāgārjuna's treatment of these three aspects of the Buddhist idea of 'relation'.

2 Nāgārjuna's acceptance of the canonical perspectives

In many an occasion we find Nāgārjuna expounding the three kinds of relation pointed out above. By way of example, in Nāgārjuna's works we meet with relation (1) in RĀ 1.48a: asmin satīdaṃ bhavati ("Being this, there is that"); with (2) in RĀ 1.48c: asyotpādād udetīdaṃ ("From the origination of this, that arises") and in ŚS 71ab: 'di brten 'di 'byuṅ źes bya ba'i | | 'jig rten sgrub 'di 'gog mi mdzad | | ("The mundane principle 'dependently on this, that originates' has not been abolished [by the Buddha]"); and with (3) in MMK 26.12ab: tasya tasya nirodhena tat tan nābhipravartate | ("By means of the cessation of this, that does not come forth"). But how does Nāgārjuna deal with these three aspects of relation? In what follows I will try to systematize Nāgārjuna's thought concerning this subject.

2.1 (Onto)logical, perfect bilateral relation: the case (1)

Nāgārjuna describes the case (1) in several places, adding also examples in order to clarify to what kind of relation this case can be applicable. One of his favourite examples is the couple long/short that occurs in many an occasion. For the sake of brevity, let us consider here RĀ 1.48ab,49ab: asmin satīdaṃ bhavati dīrghe hrasvaṃ yathā sati | [...] hrasve'sati punar dīrghaṃ na bhavaty asvabhāvataḥ | ("Being this, there is that, like when there is the long, there is [also] the short [...]. Moreover, if there is not the short, [also] there is not the long: [they] are not by [their] intrinsic nature").⁶ On account of this pair of opposites in AS 12cd Nāgārjuna questions: yadā nāpekṣate dīrghaṃ kuto hrasvādikaṃ tadā || ("When it is not referred to the long, how [could] then [exist] the short and so on?"). In this cases, a mutual synchronic bond is clearly involved: if there is (sati)/is not (asati) A, there is (bhavati)/is not (na bhavati) also B and if there is/is not B there is/is not also A.⁷ What, indeed, comes first, between long and short?

Interestingly enough, Nāgārjuna here makes use of the terms hrasva and dīrgha, instead of mantioning abstract concepts like hrasva-tva (short-ness) and dīrgha-tva (long-ness). This terminology suggests that Nāgārjuna intended to stress the idea that the elements involved in this kind of mutual link are not pure abstract factors, rather concrete phenomena. Hence, "long" should be understood as "something that is long", and "short" as "something that is short". In other cases Nāgārjuna has recourse to the link (1) pointing to a logical relation as, for instance, in AS 11, where we read: svatve sati paratve syāt paratve svatvam iṣyate | āpekṣikī tayoḥ siddhiḥ pārāvāram ivoditā || ("When there is own-ness, there is other-ness; when other-ness, own-ness is accepted; the proof of these two is said to be mutual, like this shore and the other shore").⁸ The presence of the suffix -tva, in sva-tva and para-tva, conveys the idea that here Nāgārjuna is speaking of pure logical (ideal) concepts. On the basis of what precedes, and if we try to read in a philosophical perspective both RĀ 1.48ab,49ab and AS 11ab, these passages suggest to us that Nāgārjuna employs the relation (1) in order to explain the bond existing between mutually related factors. These factors, moreover, are/can be considered according to different degrees of conceptualization. In other words, on (a) an objective level, Nāgārjuna speaks of relations taking place between something long and something short. On (b) a subjective level, the perceiver's position/collocation in respect to the observed things is relevant: this shore clearly indicates the shore that is nearer to the observer, whereas the other shore stands for the farther one. On (c) an abstract level, Nāgārjuna pushes the categorization in the field of logic: own-ness is indeed a category that one can apply to every thing/concept in itself, whereas other-ness is a category that does not define the thing/concept, to which own-ness belongs.⁹

Relations of the kind (1), thence, seem to be grounded on an (onto)logical comparison of the factors involved, that is to say, A is A only in relation to B, and B is B only in relation to A. The evident implication of this kind of relation is that it does not necessarily entail any originative activity: something is in a certain way in consequence of a comparison with something else that is in another different way.¹⁰ The factor A does not give rise to the factor B, it simply sheds light on some particular characteristic possessed by B, when B is compared to A. This means that no actual change or modification concerning the factors involved do actually occur: if A is shorter than B, then we say that B is long, but if C is longer that B, we should say that B is short, without however that a real modification of B did take place. We can call this kind of bond a perfect synchronic bilateral relation.

2.2 Unidirectional, strictly subordinative existential relation: the case (2)

This kind of relation is well represented by RĀ 1.48cd,49cd: asyotpādād udetīdaṃ dīpotpādād yathā prabhā || [...] pradīpasyāpi anutpādāt prabhāyā apy asaṃbhavaḥ || ("From the origination of this, that arises, like from the origination of the lamp, the light [appears]; [...] moreover, from the non-origination of the lamp, surely there is non-appearance of light"). The first thing that we can observe, in this case, is that in this kind of bond, which is focused on an utpāda/anutpāda relational pattern, the originator (the lamp) is necessarily pre-existent with respect to the originated (the light). This means that this kind of relation entails in itself a chronological subordination. Nāgārjuna himself articulates quite well this concept by having resource to the example of the seed-sprout conditional succession. In MMK 17.7-8abc, indeed, Nāgārjuna explains: yo'ṅkuraprabhṛtir bījāt saṃtāno'bhipravartate | tataḥ phalam ṛte bījāt sa ca nābhipravartate || bījāc ca yasmāt saṃtānaḥ saṃtānāc ca phalodbhavaḥ | bījapūrvaṃ phalaṃ ("That series, beginning with the sprout, proceeds from the seed up to the fruit, and without the seed this [fruit] does not proceed; and since the series arises from the seed, and the fruit from the series, [it follows that] the seed is prior to the fruit, therefore, [the seed] is neither annihilated nor eternal").¹¹ The expression bījapūrvaṃ phalaṃ ("the seed is prior to the fruit"), suggests that we are in presence here of primarily a chronological succession.¹² As the sprout comes out from the seed, the seed being prior to that sprout, so the light comes out from the lamp, the lamp being prior to that light. Thus, unilaterality can be detected from the presence of diachrony between originator and originated.

Moreover, in this kind of relation (2), besides the aformentioned chronological subordination, also a unidirectional conditioning succession between the factors is involved. Such succession, which corroborates the subordination of the originated to the originator, is grounded on an existential level. Indeed, in this case the originated factor B can exist if, and only if, the originator A exists, whereas the opposite is not only untrue, but impossible. In other terms: out of nothing but A, there is B. Such a concept is clearly stated by Nāgārjuna in MMK 10.13a, in a passage in which he deals with the fire-fuel relation: āgacchaty anyato nāgnir ("Fire does not come out from another thing [than fuel]").

2.3 Non-strictly subordinative existential relation: the case (3)

Regarding the kind of relation (3), it is worth of note that Nāgārjuna adheres-as pointed out in section 2 (see MMK 26.12ab)-to the pattern put forward in Sn, quoted above. What is interesting, in this case, is that Nāgārjuna applies explicitly this relational model to the avidyā-saṃskāras bond, that is, Nāgārjuna suggests that the relation (3) takes place between the elements of the conditioned co-origination. In MMK 26.11ab, indeed, we read: avidyāyāṃ niruddhāyāṃ saṃskārāṇām asaṃbhavaḥ | ("When nescience is suppressed, there is no occasion for pre-forming tendencies").¹³ And exactly here is where problems begin, since elsewhere Nāgārjuna describes the conditionality existing between avidyā and saṃskāras as reciprocal. In ŚS 11 he, indeed, writes:

ma rig 'du byed med mi 'byuṅ | |

de med 'du byed mi 'byuṅ bas | |

de gñis phan sthun rgyu phyir yaṅ | |

raṅ bźin gyis ni ma grub yin | |

Nescience does not originate without the pre-forming tendencies and the pre-forming tendencies do not originate without that [nescience]; moreover, since the two are cause of each other, they are not established by intrinsic nature.

Hence, how to unravel this issue? Is the avidyā-saṃskāras link a unidirectional or a bilateral relation? To begin with, let us tackle the problem by noticing that in ŚS 11 the causal relation described,¹⁴ though bilateral as in case (1), is of the utpāda/anutpāda (T. 'byuṅ/mi 'byuṅ) type, as in case (2): A arises if there is B, and B arises if there is A. This relational model, adopted by Nāgārjuna to describe the avidyā-saṃskāras bond, reminds us of the mutual conditionality between viññāṇa and nāmarūpa, of which Dīghanikāya 2.62-63 and Saṃyuttanikāya 2.114 speak. In these two canonical passages we find the explanation of the two statements viññāṇapaccayā nāmarūpan ti and nāmarūpapaccayā viññāṇan ti.¹⁵ But what is worth of note, here, is the fact that this bilateral aspect of the viññāṇa-nāmarūpa relation is the only occurrence of explicit mutual conditionality that can be found in the Suttapiṭaka (at least the only one I have been able to detect) among two subsequent elements of the conditioned co-origination.¹⁶ By applying this pattern to the avidyā-saṃskāras link, I suggest, Nāgārjuna is therefore introducing a new perspective in the condition co-origination theory. Indeed, even if Dīghanikāya 2.62-63 and Saṃyuttanikāya 2.114 could have conceptually inspired ŚS 11, the philosophical horizon of the two contexts remains quite different from each other. Dīghanikāya 2.62-63 tells us that this mutual dependence is something necessary in order to guarantee the conception of a child. Though nāmarūpa can develop in the mother's womb only after viññāṇa has entered that womb, without the presence of nāmarūpa (i.e., without a physical body) as its dwelling place, viññāṇa is not able to bring the psycho-physical body to birth.¹⁷ So, exactly in this way we should understand also Saṃyuttanikāya 2.114, where the mutual relation between viññāṇa and nāmarūpa is explained by means of the simile of two bundles of reeds, which stand by leaning on each other (dve naḷakalāpiyo aññamaññaṃ nissāya tiṭṭheyyuṃ). However, the problem is that ŚS 11 speaks of avidyā and saṃskāras (and ŚS 13, as we will see in a while, extends the treatment to all the other factors of the pratītyasamutpāda chain), whereas both Dīghanikāya 2.62-63 and Saṃyuttanikāya 2.114 deal with a ten-membered 'version' of paṭiccasamuppāda, from viññāṇa to jarā-maraṇa, and exactly avidyā and saṃskāras are let aside.¹⁸

Furthermore, Nāgārjuna provides elsewhere an explanation of the particular bilateral relation existing between vijñāna and nāmarūpa, which seems to be in accordance with the aforesaid canonical interpretation. Indeed, in MMK 26.2cd Nāgārjuna writes: saṃniviṣṭe'tha vijñāne nāmarūpaṃ niṣicyate || ("Then, after the consciousness has entered [the womb], the name-and-form is infused"), and in MMK 26.4: cakṣuḥ pratītya rūpaṃ ca samanvāhāram eva ca | nāmarūpaṃ pratītyaivam vijñānaṃ saṃpravartate || ("Conditioned by the eye and the form, and also by attention-thus, [in other words,] conditioned by name-and-form-, consciousness proceeds").¹⁹ Nāgārjuna is here repeating the canonical idea, according to which the presence of viññāṇa represents the 'occasion' for a (new) nāmarūpa (the process of rebirth being here involved), whereas nāmarūpa is the 'occasion' for viññāṇa to perform its activity (the perceptive process being intended). This consideration leads us to suppose that the particular reciprocity that takes place only between viññāṇa and nāmarūpa, does not represent the general concept of mutuality, to which Nāgārjuna hints at in ŚS 11 (and in ŚS 13, below). In order to shed more light on this aspect of the matter, I think that it will be useful here to take into consideration the commentarial tradition. Let us, therefore, begin with the following excerpt of Candrakīrti's Vṛtti on ŚS 11ab:

gal te ma rig pa 'du byed rnams kyi rgyur rnam par bźag pa ni 'di ltar 'du byed med pa'i phyir de med la rgyu gaṅ źig rgyu ñid du 'bras bu daṅ 'brel pa byas pa las yin gyi 'brel pa med na rgyu ñid nam yaṅ ma yin pa'i phyir ro | |

yaṅ na 'bras bu ñe bar ma bzuṅ bar 'ga' źig rgyu ñid du 'gyur na | de lta na 'ga' źig 'ga' źig gi rgyur mi 'gyur la | de'i phyir rgyu'i rnam par gźag pa 'bras bu med na ma yin la rgyu yaṅ ma rig pa ni 'du byed rnams kyi'o źes bya ste | ma rig pa ni 'du byed med na rgyur mi 'gyur la | rgyu 'bras kyi dṅos po'i skabs yin pa'i phyir | rgyu'i sgra med kyaṅ rgyu ñid du rtogs pa ñid kyi 'bras bu rgyu'i sgras ma gsuṅs so | |

ji ltar 'bras bu la ma ltos par 'ga' źig rgyu ñid ma yin pa'i phyir ma rig pa 'du byed med na mi 'byuṅ ba bźin 'bras bu yaṅ rgyu la ma ltos na | rgyu med pa las mi srid do źes bya ba ni | de med 'du byed mi 'byuṅ źiṅ | ma rig pa med na 'du byed mi 'thad de | rgyu med par thal ba'i phyir ro | |

If nescience is determined as cause of the pre-forming impulses, and thus through the absence of the pre-forming impulses there is [also] absence of that [nescience] because whatever cause is connected with [its] effect in a causal process (hetutve), [it follows that] when there is absence of [such a] connection, then there is never causality.

Or, if [one admits that] the effect develops in a certain causal process without being supported [by its cause]-and if it is so, something [i.e. the effect] does not come to be by means of some cause-, for that reason there is no determination of the cause when [that cause is] without [relation to] the effect; but [here] it is said that the cause of the pre-forming tendencies is nescience, [so] nescience without pre-forming tendencies [as its effect] is not a cause; and because the events of cause and effect are introduced [in this discussion], [even if] the words "effect" (phala) and "cause" (hetu) are not [explicitly] mentioned [in ŚS 11ab], notwithstanding the absence of the term "cause" [as referring to nescience, this kārikā] is to be understood as [dealing with] causality (hetutva).

Thus, because there is not any causality without reference to the effect-[indeed,] if there is absence of pre-forming impulses, nescience does not originate-, in the same way, without reference to the cause, the effect too [is not admitted]-it is said [indeed] that from the absence of the cause [an effect] is not possible. "The pre-forming tendencies do not originate without that" [means that] the pre-forming tendencies are not admitted in the absence of nescience, because this [non-admission is] the consequence of the absence of [their] cause.²⁰

In short, according to Candrakīrti's interpretation this kārikā deals with a kind of relation that is essentially unidirectional (the main idea is indeed that avidyā is the cause of saṃskāras, avidyā→saṃskāras), whereas the reciprocity between the two members is confined within a mere semantic horizon: the saṃskāras can be an effect of avidyā only if avidyā is their cause, since we cannot speak of effect without reference to a cause and vice versa. Candrakīrti, moreover, seems not to focus too much on the idea of phan tshun rgyu (which can be restored into Sanskrit as parasparaṃ hetu), that is "mutual cause".²¹ He, indeed, by glossing ŚS 11c as follows: pha rol daṅ tshu rol bźin nam riṅ po daṅ thuṅ ṅu bźin rgyu daṅ 'bras bu dag phan tshun ltos pa daṅ bcas pa'i phyir ṅo bo ñid kyis grub pa med do | | ("Since cause and effect are in mutual relation and connection, like this shore and that shore or like long and short, they are not established by means of intrinsic nature"),²² clearly interprets the mutual relation between avidyā and saṃskāsas according to our pattern (1), which has been intended by Nāgārjuna, as we have seen, to be perfectly bilateral. Candrakīrti's exegesis presents however some problem, because ŚS 11c states that between nescience and pre-forming impulses it exists a link of mutual origination-a meaning conveyed by the verb 'byuṅ ba (S. ut√pad)-and not an (onto)logical relation-which would rather depend on the "...sati ...bhavati" pattern. Incidentally, Parahita (eleventh century CE) in his commentary on the ŚS seems to stress better this idea of mutual origination, which he refers to a mutual relation of causality and dependence (Vivṛtti ad ŚS 11):

ma rig 'du byed med mi 'byuṅ | źes bya ba la sogs pa 'phags pa'i tshigs su bcad pa gñis smras so | | 'du byed mi 'byuṅ ba źes bya ba'i de ma thag tu gaṅ gi phyir lhan cig 'byuṅ ba'i rgyu ñid yin pa de'i phyir źes khoṅ nas dbyuṅ ba sbyin par bya'o | des na don 'di yin te | gaṅ gi phyir lhan cig 'byuṅ ba'i rgyu ñid yin na | 'du byed med na ma rig pa daṅ | de med na 'du byed mi 'byuṅ ba de bas phan tshun rgyu ñid de phan tshun brten pa ñid yin pa [...]| |

"Nescience does not originate without the pre-forming tendencies" and so on, says the noble stanza. Why, immediately after, is it said that "the pre-forming tendencies do not originate [without nescience]"? An elliptical statement (*adhyāhāra) is given because [here] causality involves co-existence (*sahabhāva) [of both cause and effect]. Therefore, being this the meaning, why causality involves co-existence? When there are not pre-forming tendencies, nescience [does not originate] and when there is not that [nescience], pre-forming tendencies do not originate, hence there is mutual causality and mutual dependence [...].²³

Even if it interprets the meaning of phan tshun in a different way (and somehow more adherent to Nāgārjuna's text) than Candrakīrti's commentary, it seems to me that Parahita's explanation leaves in any case open a fundamental question, whose answer represents the key for the interpretation of the avidyā-saṃskāras link in Nāgārjuna's perspective. The question could be formulated as follows: what does "mutual causality and mutual dependence" mean in this context? Indeed, if, on the one hand, mutuality is undoubtedly involved in ŚS 11 (as Parahita underlines), on the other hand, it is quite risky to suppose that Nāgārjuna is here upholding a perfect bilateral relation, since this would contravene (and Candrakīrti in his commentary on ŚS 11ab appears to be well aware of this) the rule of pratītyasamutpāda, according to which avidyā is the element which originates saṃskāras, and not the contrary.

In the following section I will try to propose a possible solution to this apparent impasse.

2.3.1 What kind of reciprocity does it exist between avidyā and saṃskāras? The example of father and son

We find Nāgārjuna explaining his own view of the kind of relation existing between nescience and pre-forming impulses in ŚS 13, where he makes use of the example of the father and the son. ŚS 13 runs as follows:

pha bu ma yin bu pha min | |

de gñis phan tshun med min la | |

de gñis cig car yaṅ min ltar | |

yan lag bcu gñis de bźin no | |

The father is not the son, the son is not the father; those two do not exist the one without the other, moreover those two do not exist simultaneously: the twelve factors are like them.

In order to understand how this example is applicable to the avidyā-saṃskāras relation, as a first step let us see what Candrakīrti writes on this regard in his Vṛtti:

pha ñid bu ma yin te | pha źes bya skyed byed la | de ji ltar bskyed byar 'gyur | bu yaṅ pha min te | bu ni bskyed bya la | de ji ltar skyed byed du 'gyur | | de ltar re źig pa ñid bu ma yin la bu ñid pha ma yin no | | tha mi dad pa ma yin pa'i phyir de gñis phan tshun med pa min te | bu med na pha 'thad pa ma yin te | skyed par byed pa'i las kyis pha ñid du 'thad pa'i phyir | bu med na ci źig skyed par byed pas pha ñid du 'gyur | pha yaṅ bu med na mi 'grub pas | pha med na bu yaṅ mi 'thad do źes bya bas | de gñis phan tshun med min źiṅ | de gñis cig car yaṅ med ltar te | de gñis cig car na ba laṅ gi rva g.yas pa daṅ g.yon pa bźin pas bskyed bya skyed byed med pa ñid du 'gyur ro | | ji ltar pha daṅ bu dag ma grub pa de bźin du yan lag bcu gñis kyi rten ciṅ 'brel par 'byuṅ ba yaṅ ma grub ste | ma rig pa ñid 'du byed ma yin la | 'du byed ñid ma rig pa ma yin źiṅ | de gñis phan tshun med min źiṅ | de gñis cig car yaṅ ma yin | źes bya bas thams cad la mtshuṅs par sbyar bar bya'o źes gsuṅs so | |

The father, indeed, is not the son. 'Father' [means] generator (*janaka): how [could] he be [identical with] the generated (*janya) [son]? Also, the son is not the father. 'Son' [means] generated: how [could] he be [identical with] the generator? Hence, the former is not the son and the son is not the father. There being no identity (*abheda) [between them], those two do not exist the one without the other: if there is not a son, a father is not admitted because the very father is admitted in consequence of the generative action, [but] when there is not a son, what generator is [to be taken] as father? Because it is said that both the father is not established when there is not a son, and the son is not admitted when there is not a father, [it follows that] "those two do not exist the one without the other; moreover those two do not exist simultaneously"; if those two [were] simultaneous, there would be non-existence of generated and generator, because of the similarity with the right and the left cow's horn. As father and son are not established [in themselves], in the same way also the twelve factors of the conditioned co-origination are not established [in themselves], that is: the nescience itself is not the pre-forming impulses and the pre-forming impulses themselves are not nescience, and those two do not exist the one without the other. Moreover, those two do not exist simultaneously; [they] are to be put in equation [with father and son] on every point, thus it is explained.²⁴

Candrakīrti, we observe, explains the father-son relation by pinpointing first of all the generative succession intervening between them. In doing so, he follows the first part of his interpretation of ŚS 11. The avidyā-saṃskāras link is indeed similar to the father-son bond when we consider this relation from the utpāda/anutpāda point of view: in both cases there is a unidirectional dependence. Nāgārjuna himself was not at all unaware of this perspective. In the Vigrahavyāvartanī we find indeed the unequivocal stanza 49:

pitrā yady utpādyaḥ putro yadi tena caiva putreṇa |

utpādyaḥ sa yadi pitā vada tatrotpādayati kaḥ kam ||

If the son originates by the father, and if by that very son that [very] father originates [as well]; speak! There, who originates who?

What is worth of note, here, is the insistence on the verb ut√pād, which makes us exclude the possibility of being in presence of the kind of relation (1). However, as we have seen, the avidyā-saṃskāras link is said to be bilateral exactly according to the utpāda/anutpāda perspective: in the absence of nescience-tells us ŚS 11-the pre-forming tendencies do not originate (mi 'byuṅ), as well as in the absence of pre-forming tendencies, nescience does not originate (mi 'byuṅ). Hence, at a first glance and despite ŚS 13, the reciprocal dependence existing between father and son, on the one hand, and between avidyā and saṃskāras, on the other hand, seems not to be of the same kind. Rather, it seems that two different kinds of dependence should be here theorized in order to explain the two theoretical contexts. But two different kinds of dependence entail two different conceptual perspectives, with the consequent risk of vitiating from the very beginning the validity of Nāgārjuna's example (ŚS 13d). Hence, in order to avoid the conclusion of inconsistency, we should ask ourselves why Nāgārjuna makes use of the father-son example, even if it apparently does not apply consistently to the avidyā-saṃskāras relation. I suggest that we could find a possible solution to this issue, on the one hand, if we change our perspective on what concerns the meaning of 'byuṅ ba in ŚS 11 and, on the other hand, if we look to ŚS 13 in order to find a clue for determining the direction in which to push this shift.

We have already seen how, in MMK 26.11ab, Nāgārjuna admits a-so to speak-sort of precedence of avidyā on saṃskāras, a precedence based on the fact that from the elimination of nescience, there is no further occasion for the origination of the pre-forming impulses. The presence of negative expressions in MMK 26.11ab (nirodhena and nābhipravartate) is not in my opinion incidental. I suspect on the contrary that Nāgārjuna has recurred to these expressions in order to describe what, between avidyā and saṃskāras, corresponds to the primary causal factor (T. rgyu, S. hetu) in an originative (utpāda) relation. In this relation, however, a certain degree of conditional reciprocity, as said in ŚS 11c, is involved. Hence, we have a situation in which the presence of the cause entails the presence of an effect that, in its turn, 'reacts' to its own cause. I suggest, then, that Nāgārjuna in MMK 26.11ab has recourse to the negative expressions mentioned above in order to underline the fact that, between the two factors involved, it is exactly avidyā that is the primary originative cause of saṃskāras, which are its originated effect. Such a 'negative' approach, indeed, makes evident that in the absence of the cause also the effect must be absent. In the light of this argument, hence, it is correct to say that avidyā gives origin to the saṃskāras like a father gives origin to a son, and that saṃskāras are originated by avidyā as well as the son is originated by the father (this is Candrakīrti's interpretation of both ŚS 11ab and ŚS 13a). To confirm the unilateral originative link between the two members, we need indeed that without avidyā, saṃskāras be not originated, as well as without a father a son is not originated, and that if saṃskāras are not originated, then avidyā must be not present, as if the son is not originated, then a father does not exist. We are here in the presence of the application of two causal models, that is, the positive A→b model (where the capital letter refers to the originator and the lowcase one to the originated), and the negative ~A→~b model (where "~" indicates non-presence). This explanation is conceptually in line with ŚS 13c, where the non-simultaneity of the two members involved in the relation is pointed out. Such a consideration corroborates the idea that, since the father must necessarily be prior to the son, because the cause, to be a cause of an effect, must precede its own effect (as MMK 17.8c directly, and Vigrahavyāvartanī 49 indirectly, outline), for that very reason also nescience (the cause) must necessarily exist before the pre-forming impulses (the effect). But if we agree with this conclusion, it follows that the reciprocity described in ŚS 11c cannot be-as Candrakīrti seems to point out by comparing the father-son bond to the long and short relation-a perfectly bilateral, i.e. synchronic, relation.²⁵

Candrakīrti, however, is not completely wrong on this point, because the father-son link has in any case a theoretical similarity with the bilateral relations mentioned above. Indeed, the fact of being a father or a son does not in itself change their-as it were-basic or general natures. Rather, what changes is only the category according to which we consider these two persons and this reminds us of the case of a thing that can be categorized as shorter in respect to another thing and/but longer in respect a third thing. If we accept such a viewpoint, then we have to conclude that in both cases we are confined within an (onto)logical framework.

However, I think this is a reductive interpretation of the avidyā-saṃskāras (and father-son) bond. In order to better clarify my point of view, I propose to explain this particular aspect of the relation between avidyā and saṃskāras by referring to the so-called 'Cambridge change'. The 'Cambridge change' defines the changes that concern the relational predicates of the things involved in some kind of link, without that any actual modification of their-so to speak-substantial existences did really take place.²⁶ This applies well also to our cases, with the exception that Nāgārjuna would surely not have accepted any idea of substance underlying the changing attributes. According to the 'Cambridge change' perspective, a thing/person is short/father only in relation to another thing/person which is long/son, and vice versa. The modification of the categorical/predicative status of the related members depends on the terms of comparison. This becomes clearer when we consider that a man, who is a father in relation to his son, is likewise a husband in relation to his wife, and he is a son in relation to his own father. In all these relations, however, he remains always the same human being. Only the logical(/social) attributes appear to change. Nāgārjuna on this issue could have had in mind a perspective very similar to-even if not identical with-the 'Cambridge change', that is, a particular aspect of Buddhadeva's thought. However, since we have no textual evidence that corroborates such an assumption, we must be cautious and say that it is possible, but not certain, that Buddhadeva's philosophy inspired Nāgārjuna.²⁷ Buddhadeva was a Sarvāstivādin and developed a particular concept of dependence in relation to the three times (past, present, future). No direct work of Buddhadeva reached us, but his viewpoint on this matter is summarized in several sources. For instance, the Vibhāṣāprabhāvṛtti commentary on Abhidharmadīpa 302 tells us that Buddhadeva had recourse to a particular example in order to clarify his approach to the theory of dependence in relation to the three times: dharmo'dhvasu pravartamānaḥ pūrvāparam avekṣyānyathā cānyathā cocyate | naivāsya bhāvānyathātvaṃ bhavati dravyānyathātvaṃ vā | athaikā strī pūrvāparam apekṣya mātā cocyate duhitā ca | ("A dharma proceeding on the way [of its existence] is said to be one or the other, in consideration of before and after. Of this [dharma] neither there is difference in [its] mode of existence, nor [is there] difference in [its] substance. Then [for instance] one woman is called mother or daughter in consideration of before and after [her giving birth to her own offspring]").²⁸

However, what makes quite reductive the assimilation of the father-son (avidyā-saṃskāras) relation to the long-short relation is the fact that Nāgārjuna in ŚS 11c affirms a principle of mutual causality between the factors involved in the relation and not, as in RĀ 1.48ab-49ab, a mere co-existence and co-dependence. This means that not only the father/avidyā is cause of the son/saṃskāras, but also the son/saṃskāras is/are cause of the father/avidyā. Our problem, then, seems to remain still unsolved. I suggest, at this point, that we need to change our perspective, in order to answer to the following question: what differentiates father and son from long and short? The answer that I will take into consideration here is: the difference lies in the fact that only father and son can actively-by 'actively' I mean by force of their own activity-interact with each other from within a precise categorical framework. The same, says Nāgārjuna, happens with avidyā and saṃskāras. Indeed, as 'father' is a person considered according to his-so to speak-social status (i.e., according to a particular category) and 'son' another person belonging to another social status,²⁹ in the same way nescience can be seen as nothing but a particular mental status,³⁰ and the pre-forming impulses, conditioned by nescience, as another mental status.³¹

Now, it remains the point of how to explain the active interactions taking place between the two members involved in the relation. I suggest that a possible explanation, consistent with the passages so far taken into consideration, can be obtained by way of an example, if we observe what happens-or can, or could happen-between father and son in a situation like the following one, which is, I think, very likely. Let us suppose that it is night, and the father is sleeping in his bed; at a certain point the son begins to cry-why is he crying is here a negligible detail-; the father then wakes up and goes to his son, trying to guess why he cries. Then he decides to hold his son in his arms for lulling him to sleep. To be brief, let us imagine that the son, reassured by the lullaby, falls again asleep. Now, what is in this illustration particularly meaningful for our discussion? On the one hand, the fact that the son, by the simple fact of crying, makes his father do something, i.e. to sing a lullaby, as a result of that input. On the other hand, the fact that the father, by singing the lullaby and rocking his baby, makes the son do something, i.e. to fall asleep again, as a result of his behavior. In other words, the son makes the father behave as father and the father makes the son behave as son. Thus, the father is 'corroborated' in his role of father exactly when the son is 'corroborated' in his role of son. What is worth of note, here, is the fact that such a reciprocity is neither an (onto)logical relation of the kind (1) tout court, nor a strictly subordinative relation of the kind (2) tout court: it rather looks like a non-strictly subordinative relation, in which an originative process is involved, and is based on reciprocal feedbacks taking place between two elements considered within a particular categorical framework. In other words, it is a relation of the kind (3). A certain person acts as a father because the general context, into which he finds himself, makes him feel himself a father, since his acting as a father is stimulated when he is in relation-or believes to be in relation-with his son. When, on the other hand, that same person is in relation, for instance, with his wife, we should expect that the general context makes him feel himself-and consequently behave as-a husband.

In the same way, I suggest, we could imagine the link between avidyā and saṃskāras described in ŚS 11. It is in this case noteworthy that Nāgārjuna, in ŚS 11a, begins with the saṃskāras→avidyā relational direction and only in ŚS 11b he proceeds with avidyā→saṃskāras, inverting thus the usual way of the conditional chain. In my opinion, this fact supports the idea that here we are not in presence of the case of 'bare' originative relations (as in SN 2.1), rather of-as it were-'feedback' relations. I propose to explain the conceptual meaning of this inversion, in simple terms, as follows: when avidyā is confirmed in its role by saṃskāras, then saṃskāras are reinforced by avidyā, which is in its turn confirmed again by them, and so on.³²

Now, a last consideration deserves to be put forward. We have seen how, in MMK 26.11ab, Nāgārjuna argues that with the cessation of avidyā there is no origination of saṃskāras. Let us then try to apply to this relational pattern the father-son example. In this case, since we are dealing with the cessational aspect of the avidyā-saṃskāras relation, we could explain it by considering that the father's death does not entail in itself the son's death, rather it entails that no new son can be originated by that very father. The already existing son continues to exist, he survives his father (probably feeling himself as the son of his father for the rest of his life, also after his father is passed away). In the light of this consideration, I propose to interpret the meaning of MMK 26.11ab as not pointing to the idea that with the elimination of nescience also the already existing pre-forming impulses are eliminated. Rather, the already existing pre-forming impulses remain present up to their 'natural' end. Only, they cannot make nescience reacting to their inputs, they are-so to speak-halved. In other words, when someone is no more under the effect of avidyā, then saṃskāras lose their retroactivity: the 'feedback' relation is thus cut off, it becomes fruitless and, being fruitless, it is also causeless. The same applies, according to ŚS 13d, to the other factors of the pratītyasamutpāda.

• *
  I am grateful to Giuseppe Ferraro for having read with insightful philosophical analysis and criticism a first draft of this paper, and to Ramkrishna Bhattacharya for his notes on the text and his help with my English.
• 1
  Russell (1903, p. 221).
• 2
  For a brief but clear sketch on the concept of paṭiccasamuppāda as fundamental topic of (early) Buddhism, see Miyamoto (1955). For a general survey on Nāgārjuna's understanding of pratītyasamutpāda, which is not the topic I am concerned with here, see among others the following specific studies: Tachikawa (1980-1981), Chinn (2001). Verhagen (1988) provides an insightful study of the Tibetan translation of the term pratītyasamutpāda, from a grammatical and semantic point of view.
• 3
  Majjhimanikāya 1.262-263; Saṃyuttanikāya 2.28; 2.70; 2.96.
• 4
  Majjhimanikāya 1.264. For a study and interpretation of the causal pattern(s) according to the Canonical sources I refer the reader to Watts (1982).
• 5
  Suttanipāta 3.12 (p. 141).
• 6
  See AS 13b (dīrghe hrasvaṃ tathā sati) and RĀ 1.92c (T. thuṅ daṅ riṅ); in RĀ 1.93b to the pair long and short, Nāgārjuna adds the pair thin (T. phra) and thick (T. sbom), and in RĀ 95ab to these couples also virtue (T. dge) and non-virtue (T. mi dge) are added. In AS 11ab own-ness (svatva) and other-ness (paratva) are considered. Other similar examples in MMK 23.10ab-11ab: anapekṣya śubhaṃ nāsty aśubhaṃ prajñapayemahi | [...] anapekṣyāśubhaṃ nāsti śubhaṃ prajñapayemahi | ("We make known that there is not impure without reference to the pure; we make known that there is not pure without reference to the impure") and ŚS 7ab: gcig med par ni maṅ po daṅ | | maṅ po med par gcig mi 'jug | | ("In the absence of one, the many [do not follow], and in the absence of many, the one does not follow").
• 7
  Taber calls this kind of relation "principle of coexisting counterparts". It is interesting to note that he arrives at the conclusion that (1998, p. 237): "The reader [of Nāgārjuna's texts] is not compelled to adopt that perspective by rigorous logic, but is invited to do so by making a paradigm shift [...]. Viewed in this way, the principle of coexisting counterparts [...] expresses Nāgārjuna's main metaphysical insight" (emphasis added). See also Taber (1998, p. 241, note 17).
• 8
  In the same way, abstract concepts are also dealt with in, for instance, AS 13a,c: astitve sati nāstitvaṃ [...] | nāstitve sati cāstitvaṃ ("If there is existence, there is non-existence [...], and if there is non-existence, there is existence").
• 9
  Own-ness’ and ‘other-ness’, as general, absolute or abstract conceptual categories (suffix -tva), are neither necessarily connected to something real, nor necessarily dependent on comparisons between real things (as ‘long’ and ‘short’ are), even if they could be derived from an abstractive process inspired by comparisons of some kind. A similar observation could be made also for the couple existence/non-existence (S. sat/asat, T. yod/med). Indeed, it is not possible to determine the existence of a thing by a mere comparison with those things that do not exist, as it is impossible, between two things, to determine tout court which thing is the same and which the other (whereas it is possible to determine which of the two is shorter, and which longer). The binomial existence/non-existence seems, thus, to be purely a way according to which we conceptually see things. I propose to interpret in this direction AS 13a,c. Moreover, the fact that AS 13b (dīrghe hrasvaṃ tathā sati) follows AS 13a as an example of the case suggests, in my opinion, the possible mental procedure that, according to Nāgārjuna, we bring into play when things are considered in a comparative way: if one thing has existence (astitva, abstractive categorical conceptualization) as longer– i.e., it is long (dīrgha)–when compare with another thing, it follows that the other thing has its existence as shorter–i.e., it is short (hrasva)–than the former one. This means that the first thing does not have existence (nāstitva) as shorter–i.e. it is not shorter–than the second thing, etc. The fundamental point, here, is the difference between to have a characteristic and to be in a certain way. Only the relationship of ‘to have’ involves a mere, pure conceptual distinction between characterized (lakṣya) and characteristic (lakṣaṇa); see MMK V (in particular kārikā 7), whose philosophical implication is that we can make a correct use–i.e., without grasping to svabhāva–of such conceptual abstractions, overimposed to things, only if we know how events actually are–that is, they are always in/by relation to other events.
• 10
  Compare with what Saṅghabhadra (V century C.E.) asserts on account of sahabhūhetu ("reciprocal cause") in his Nyāyānusāra (Taishō 1562, 419a1-2 and 419b1-8). It is nonetheless interesting to notice that for Saṅghabhadra the lamp-light relation is par excellence an example of reciprocal cause. On this and on the interpretation of the meaning of sahabhūhetu in abhidharmic context, see Tanaka (1985). Moreover: Potter (1999, pp. 704-705).
• 11
  See also Lokātitastava 17-18.
• 12
  Oetke observes (1990, p. 99): "[...] 'real conditions' are conceived as involving certain requirements regarding the temporal order".
• 13
  To compare with Sn. An excursus on the relation between avidyā and saṃskāras based on MMK 26.11ab can be found in Sasaki (1992, pp. 66-69, note 13). For a brief treatment of MMK 26, see Katsura (1997).
• 14
  For an insightful study of the concept of 'causality' in Buddhism see Kalupahana (1975).
• 15
  In this respect, it is interesting to notice, with Watts, that alongwith the 'philosophical' sense of "condition", paccaya (1982, p. 408) "can have such various meanings as 'support,' 'requisite,' 'means,' 'reason,' 'grounds,' 'motive'".
• 16
  Other Canonical passages underline this mutual relationship: in Saṃyuttanikāya 2.104, Dīghanikāya 2.32, 2.56, etc. viññāṇa and nāmarūpa are said to depend on each other. For a clear treatment and explanation of this particular aspect of the viññāṇa-nāmarūpa relational link, see Langer (2001, pp. 19-27), Gupta (1977, pp. 176-177).
• 17
  For viññāṇapaccayā nāmarūpan ti see Langer (2001, pp. 21-23) and for nāmarūpapaccayā viññāṇan ti see Langer (2001, pp. 23-27).
• 18
  Notwithstanding the preliminary digression of Dīghanikāya 2.58-61 (in which it is explained why vedanaṃ paṭicca taṇhā taṇhaṃ paṭicca pariyesanā pariyesanaṃ paṭicca lābho lābhaṃ paṭicca vinicchayo vinicchayaṃpaṭicca chandarāgo chandarāgaṃ paṭicca ajjhosānaṃ ajjhosānaṃ paṭicca pariggaho pariggahaṃ paṭiccamacchariyaṃ macchariyaṃ paṭicca ārakkho; “Desire is conditioned by sensation, search is conditioned by desire, acquisition is conditioned by seach, investigation is conditioned by acquisition, lust is conditioned by investigation, application is conditioned by lust, possession in conditioned by application, avarice is conditioned by possession, protection is conditioned by avarice”), the conceptual basis of Dīghanikāya 2.62-63 can indeed be traced back to Dīghanikāya 2.55-56. A dependent co-origination constituted by ten members only, occurs also in Saṃyuttanikāya 2.104 and Dīghanikāya 2.32-33, that is, in all the passages in which viññāṇa and nāmarūpa are said to be related in a mutual way.
• 19
  Compare with Majjhimanikāya 1.111: cakkhuñ c'āvuso paṭicca rūpe ca upajjati cakkhuviññāṇaṃ ("And conditioned by eye and forms, O friend, the visual awareness originates").
• 20
  Erb (1997, p. 250). sDe-dge bsTan-'gyur, dBu-ma, vol. Ya, ff. 293a2-6. As a general remark on the fundamental meaning of the 'dialectical' verbs used by Mādhyamikas, and for a better understanding of the philosophical extent adumbrated by Nāgārjunian terminology, we can recall here Bugault (1983, p. 24): "Going from YUJ to upa-PAD, and then to VID, one passes from logical possibility to real possibility (or logical and real possibility). And with VID, to simply effective reality".
• 21
  The term used by the Mādhyamikas to refer to 'mutual dependence' is parasparāpekṣā. See for instance Lokātitastava 8c, MMK 6.3d. An account of all the occurrences of mutual dependence and dependent co-arising in the MMK is provided by Tachikawa (1997, pp. 37-45). See also Tachikawa (1997, pp. 61-92).
• 22
  Erb (1997, p. 250). sDe-dge bsTan-'gyur, dBu-ma, vol. Ya, ff. 293a6-7.
• 23
  sDe-dge bsTan-'gyur, dBu-ma, vol. Ya, ff. 344b6-345a1. Emphasis added.
• 24
  Erb (1997, p. 254). sDe-dge bsTan-'gyur, dBu-ma, vol. Ya, ff. 294a5-294b2. On janaka (and janako hetuḥ) in Candrakīrti see Salvini (2014, p. 487).
• 25
  See Oetke (1990, p. 99): "[...] 'logical condition' is 'atemporal' with respect to the conditioning relation".
• 26
  On "Cambridge change" see Geach (1969, pp. 71-72). See also Geach (1972, pp. 321-322): "Clearly any change logically implies a 'Cambridge' change, but the converse is surely not true; there is a sense of 'change', hard to explicate, in which it is false to say that Socrates changes by coming to be shorter than Theaetetus when the boy grows up, or that the butter changes by rising in price, or that Herbert changes by 'becoming an object of envy to Edith'; in these cases, 'Cambridge' change of an object (Socrates, the butter, Herbert) makes no 'real' change in that object".
• 27
  As to my knowledge, the first scholar that suggested the hypothesis of a possible influence of Buddhadeva's thought on Nāgārjuna was Williams (1977, p. 281): "It is perhaps no coincidence that the words used for 'relation' or 'in dependence on' were probably the same in all cases, since the words 'apekṣā' and 'apekṣya' are found in every one of our Sanskrit sources for Buddhadeva's views as well as in the passage quoted from Nāgārjuna. Thus it is possible at least to suggest that Nāgārjuna's approach in constructing his refutation of time took as its starting point an approach to time which he knew already existed among philosophers of the school he was combating. It could therefore be conjectured that perhaps Buddhadeva's view was the most firmly established one at the time, or it is even possible that it was the only Sarvāstivādin view on the subject that Nāgārjuna knew about. There would, after all, be little point in constructing a refutation of time based on a view nobody held". Regarding the mutual influences between Madhyamaka and Abhidharma, as far as the concept of conditioned co-origination is concerned, we find a good assessment in Salvini (2014). Salvini in his study takes into account especially MMK 1, on which chapter see also the considerations put forward in Garfield (1994).
• 28
  See Jaini (1959, p. 260). Besides Williams (1977), more recently two other articles have been partly devoted to Buddhadeva's thought: Buswell (1997a) and Buswell (1997b). As far as the term bhāva (occurring in this excerpt) is concerned, Nāgārjuna's philosophical interpretation of it is well summarized in Priestly (1997).
• 29
  Taber explains this point as follows (1998, pp. 117-118): "a woman is a mother of a child only secondarily. First and foremost she is a woman, and it is by virtue of her properties as a woman, as well as other circumstances, that she is a mother. She does not depend on the child in order to exist as a woman" (emphasis added). The last sentence is unequivocal: categories are here intended as a product of our putting things in relation, because a woman is a mother only in dependence on her child.
• 30
  This perspective is confirmed by passages like Dīghanikāya 3.254, where avijjā is said to be a mental inclination, anusayā; Aṅguttaranikāya 1.194 defines avijjā as mental confusion, moha; etc.
• 31
  In this case we should be well aware that the term saṃskāra conveys several meanings, according to the context in which we find it. However, saṃskāras are usually explained by making reference to the concept of volition or intention (cetanā; there is a well-known passage, in Saṃyuttanikāya 3.60, where this equation is outlined). On this point see, among others, Vetter (2000, pp. 27-63). Nāgārjuna deals with saṃskāras in MMK 13. For an exegetical (Bhāviveka and Candrakīrti) account of this chapter, see Nietupski (1996).
• 32
  At the beginning of the last century Louis de la Vallée Poussin underlined that a similar relation takes place between saṃskāras and vijñāna (1913: 10): “Les saṃskāras sont le domaine [...] sur le quel la pensée (Majjh[imanikāya] III, 99), ou l’intelligence (vijñāna, Saṃ[yuttanikāya] II, 65), prend point d’appui (paṭiṭṭhā) pour durer et se developper. Ce faisant la pensée renouvelle les saṃskāras”.

Referencesa) Primary sources

• BHĀVIVEKA. "Prajñāpradīpavṛtti". Ed. Ramanath Pandeya, 2 Vols. "Madhyamakaśāstram of Nāgārjuna. With the Commentaries Akutobhayā by Nāgārjuna, Madhyamakavṛtti by Buddhapālita, Prajñāpradīpavṛtti by Bhāvaviveka, Prasannapadāvṛtti by Candrakīrti, Critically Reconstructed". Delhi: Motilal Banarsidass, 1988.
• BUDDHAPĀLITA. "Madhyamakavṛtti". Ed. Ramanath Pandeya, 2 vols. "Madhyamakaśāstram of Nāgārjuna. With the Commentaries Akutobhayā by Nāgārjuna, Madhyamakavṛtti by Buddhapālita, Prajñāpradīpavṛtti by Bhāvaviveka, Prasannapadāvṛtti by Candrakīrti, Critically Reconstructed". Delhi: Motilal Banarsidass 1988.
• CANDRAKĪRTI. "Prasannapadāvṛtti": ______. Ed. in Sanskrit by Louis de la Vallée Poussin. "Mūlamadhyamakakārikās (Mādhyamikasūtras) de Nāgārjuna avec la Prasannapadā commentaire de Candrakīrti". St.-Petersbourg: Bibliotheca Buddhica IV, 1903-1913.
• ______. Ed. in Tibetan and translated into French by Jacques May. "Candrakīrti Prasannapadā Madhyamakavṛtti. Douze chapitres traduits du sanscrit et du tibétain, accompagnés d'une introduction, de notes, et d'une édition critique de la version tibétaine". Paris: A. Maisonneuve, 1959.
• ______. Ed. Ramanath Pandeya, 2 Vols. "Madhyamakaśāstram of Nāgārjuna. With the Commentaries Akutobhayā by Nāgārjuna, Madhyamakavṛtti by Buddhapālita, Prajñāpradīpavṛtti by Bhāvaviveka, Prasannapadāvṛtti by Candrakīrti, Critically Reconstructed". Delhi: Motilal Banarsidass, 1988.
• CANDRAKĪRTI. "Śūnyatāsaptativṛtti": ______. "sToṅ ñid bdun cu pa'i 'grel pa". In: "The Nyingma Edition of the sDe-dge bKa'-'gyur and bsTan-'gyur", dBu-ma, Vol. Ya, ff. 267a1-336b7. Oakland (California): Dharma Mudranālaya, 1981.
• ______. Ed. Felix Erb. "Śūnyatasaptativṛtti Candrakīrtis Kommentar zu den "Siebzig Versen über die Leerheit" des Nāgārjuna [Kārikās 1-14]". Stuttgart: Franz Steiner Verlag, 1997.
• CANDRAKIRTI. "Yuktiṣaṣṭikakārikāvṛtti". Ed. in Tibetan and translated into French by Cristina Anna Scherrer-Schaub. "Yuktiṣaṣṭikavṛtti Commentaire à la soixantaine sur le raisonnement ou Du vrai einsegnement de la causalité, par le maître indien Candrakīrti". Bruxelles: Institut Belge des Hautes Études Chinoises (Mélanges chinois et bouddhique 25), 1991.
• "Dīghanikāya": ______. Ed. Thomas W. Rhys Davids, Joseph E. Carpenter, 3 Vols. "Dīgha Nikāya". Oxford: Pali Text Society, 1992-1995 (rep.).
• ______. Translated into English by Thomas W. Rhys Davids, Caroline A.F. Rhys Davids, 3 Vols. "Dialogues of the Buddha". Delhi: Motilal Banarsidass, 2000.
• "Majjhimanikāya": ______. Ed. V. Treckner, Robert Chalmers, 3 Vols. "Majjhima Nikāya". Oxford: Pali Text Society, 1993-1994 (rep.).
• ______. Translated into English by Isaline B. Horner, 3 Vols. "Middle Length Sayings". Oxford: Pali Text Society, 2000 (rep.).
• ______. Translated into English by Bhikkhu Ñāṇamoli, Bhikkhu Bodhi. "The Middle Length Discourses of the Buddha. A New Translation of the Majjhima Nikāya". Boston: Wisdom Publication, 1995.
• NĀGĀRJUNA. "Acintyastava". See: Nāgārjuna. "Catuḥstava".
• NĀGĀRJUNA. "Catuḥstava": ______. Ed. in Sanskrit and translated into English by Fernando Tola, Carmen Dragonetti. "The Catustava of Nāgārjuna". In: F. Tola, C. Dragonetti (ed). "On Voidness: A Study on Buddhist Nihilism". Delhi: Motilal Banarsidass, 1995. pp. 101-153.
• ______. Ed. in Tibetan and translated into Hindī by Gyaltsen Namdol. "Catuḥstavaḥ of Ācārya Nāgārjuna". Sarnath (Varanasi): Central Institute of Higher Tibetan Studies, 2001.
• NĀGĀRJUNA. "Lokātītastava". See: Nāgārjuna. "Catuḥstava".
• NĀGĀRJUNA. "Mūlamadhyamakakārikā": ______. Ed. Jan W. de Jong. "Mūlamadhyamakakārikā". Madras: The Adyar Library and Research Center, 1977.
• ______. Ed. Ramanath Pandeya, 2 Vols. "Madhyamakaśāstram of Nāgārjuna. With the Commentaries Akutobhayā by Nāgārjuna, Madhyamakavṛtti by Buddhapālita, Prajñāpradīpavṛtti by Bhāvaviveka, Prasannapadāvṛtti by Candrakīrti, Critically Reconstructed". Delhi: Motilal Banarsidass, 1988.
• NĀGĀRJUNA. "Ratnāvalī". Ed. Michael Hahn. "Nāgārjuna's Ratnāvalī" (Vol. I: "The Basic Texts Sanskrit, Tibetan, Chinese"). Bonn: Indica et Tibetica Verlag, 1982.
• NĀGĀRJUNA. "Śūnyatāsaptatikārikā": ______. Ed. in Tibetan and translated into English by Christian Lindtner. In: Ch. Lindtner (ed.). "Nāgārjuniana: Studies in the Writings and Philosophy of Nāgārjuna". Delhi: Motilal Banarsidass, 1987. pp. 31-69.
• ______. Ed. in Tibetan and translated into Hindī by Sempa Dorje. "Śūnyatāsaptatiḥ with Auto-Commentary of Ārya Nāgārjuna". Sarnath (Varanasi): Central Institute of Higher Tibetan Studies, 1996.
• ______. Translated into English by David R. Komito. "Nāgārjuna's 'Seventy Stanzas'. A Buddhist Psychology of Emptiness". Ithaca (New York): Snow Lion Publications, 1987.
• NĀGĀRJUNA. "Vigrahavyāvartanī": ______. Ed. in Sanskrit and translated into English by Kamaleswar Bhattacharya, Elgin H. Johnston, Arnold Kunst. "The Dialectical Method of Nāgārjuna, Vigrahavyāvartanī". Delhi: Motilal Banarsidass, 1990 (rep.)
• ______. Translated into English by Jan Westeroff. "Nagārjuna's Vigrahavyāvartanī The Dispeller of Disputes". Oxford: Oxford University Press, 2010.
• PARAHITA. "Śūnyatāsaptatikārikāvivṛtti". "sToṅ pa ñid bdun cu pa'i rnam par bśad pa". In: The Nyingma Edition of the sDe-dge bKa'-'gyur and bsTan-'gyur, dBu-ma, Vol. Ya, ff. 337a1-374b2. Oakland (California): Dharma Mudranālaya, 1981.
• PIṄGALA(?). "Akutobhayā". Ed. Ramanath Pandeya, 2 Vols. "Madhyamakaśāstram of Nāgārjuna. With the Commentaries Akutobhayā by Nāgārjuna, Madhyamakavṛtti by Buddhapālita, Prajñāpradīpavṛtti by Bhāvaviveka, Prasannapadāvṛtti by Candrakīrti, Critically Reconstructed". Delhi: Motilal Banarsidass, 1988.
• "Saṃyuttanikāya": ______. Ed. Léon Feer, G.A. Somaratne, 5 Vols. "Saṃyutta Nikāya". Oxford: Pali Text Society, 1975-1999 (rep.).
• ______. Translated into English by Caroline A.F. Rhys Davids, Frank L. Woodward, 5 Vols. "The Book of the Kindred Sayings". Delhi: Motilal Banarsidass, 2005 (rep.).
• ______. Translated into English by Bhikkhu Bodhi, 2 Vols. "The Connected Discourses of the Buddha. A New Translation of the Saṃyutta Nikāya". Boston: Wisdom Publication, 2000.
• "Suttanipāta": ______. Ed. Dines Andersen, Helmer Smith. "Suttanipāta". Oxford: Pali Text Society, 1990 (rep.).
• ______. Translated into English by Kenneth R. Norman. "The Group of Discourses". Oxford: Pali Text Society, 2001 (rep.).
• VIMALAMITRA(?). "Abhidharmadīpa". Ed. Padmanabh S. Jaini. "Abhidharmadīpa with Vibhāṣāprabhāvṛtti". Patna: Jayaswal Research Institute, 1959.

b) Secondary sources (studies and articles)

• BRONKHORST, J. (ed.). "Panels of the VIIth World Sanskrit Conference: Kern Institute, Leiden, August 23-29, 1987". Vol. 2: D. SEYFORT RUEGG, L. SCHMITHAUSEN (ed.). "Earliest Buddhism and Madhyamaka". Leiden: Brill, 1990.
• BUGAULT, G. "Logic and Dialectics in the Madhyamakakārikās". Journal of Indian Philosophy, Vol. 11, Nr. 1, pp. 7-76, 1983.
• BUSWELL, R. E. "Buddhadeva: Materials Toward an Assessment of His Philosophy". Journal of Indian Philosophy, Vol. 25, Nr. 6, pp. 561-587, 1997a.
• BUSWELL, R. E. "The Three Times Controversy in Abhidharma Philosophy". In: Nogwŏn sŭnim Kohŭi Kinyŏm haksul nonch'ong wiwŏnhoe (ed.), 1997b. pp. 1129-1266.
• CHINN, E. "Nāgārjuna's Fundamental Doctrine of pratītyasamutpāda". Philosophy East and West, Vol. 51, Nr. 1, pp. 54-72, 2001.
• GARFIELD, J. L. "Dependent Arising and the Emptiness of Emptiness: Why Did Nāgārjuna Start with Causation?". Philosophy East and West, Vol. 44, Nr. 2, pp. 219-250, 1994.
• GEACH, P.T. "God and the Soul". London: Routledge and Kegan Paul, 1969.
• GEACH, P.T. "Logic Matters". Berkley and Los Angeles: University of California Press, 1972.
• GUPTA, R. "Twelved Membered Dependent Origination: An Attempted Reappraisal". Journal of Indian Philosophy, Vol. 5, Nr. 1-2, pp. 163-186, 1977.
• KALUPAHANA, D. J. "Causality: The Central Philosophy of Buddhism". Honolulu: The University of Hawaii Press, 1975.
• KATSURA, SH. "Nāgārjuna and Pratītyasamutpāda". Journal of Indian and Buddhist Studies, Vol. 46, Nr. 1, pp. 497-492, 1997.
• LANGER, R. "Das Bewusstsein als Träger des Lebens: einige Weniger beachtete Aspekte des Viññāna in Pālikanon". Wien: Arbeitskreis für Tibetische und Buddhistische Studien Universität Wien, 2001.
• MIYAMOTO, SH. "A Re-Appraisal of pratītya-samutpāda". In: G.M. Nagao, J. Nozawa (ed.), 1955. pp. 152-164.
• NAGAO, G. M., NOZAWA, J. (ed.). "Studies in Indology and Buddhology Presented in Honour of Professor Susumu Yamaguchi on the Occasion of His Sixtieth Birthday". Kyoto: Hozokan, 1955.
• NIETUPSKI, P. "The Examination of Conditioned Entities and the Examination of Reality". Journal of Indian Philosophy, Vol. 24, Nr. 2, pp. 103-143, 1996.
• NOGWŎN SŬNIM KOHŬI KINYŎM HAKSUL NONCH'ONG WIWŎNHOE (ed.). "Han'guk Pulgyo ŭi chwapyo" ["The Direction of Korean Buddhism"]. Seoul: Pulgyo Sidaesa, 1997.
• OETKE, C. "On Some Non-Formal Aspects of the Proofs of the Madhyamakakārikās". In: J. Bronkhorst (ed.), 1990. pp. 91-109.
• POTTER, K.H. (ed.). "Encyclopedia of Indian Philosophies". Vol. VIII: "Buddhist Philosophy from 100 to 350 A.D.". Delhi: Motilal Banarsidass, 1999.
• PRIESTLY, L. C. D. C. "The Term 'bhāva' and the Philosophy of Nāgārjuna". In: N.K. Wagle, F. Watanabe (ed.), 1997. pp. 124-143.
• RUSSELL, B. "The Principles of Mathematics". Cambridge: Cambridge University Press, 1903.
• SALVINI, M. "Dependent Arising, Non-arising, and the Mind: MMK1 and the Abhidharma". Journal of Indian Philosophy, Vol. 42, Nr. 4, pp. 471-497, 2014.
• SASAKI, G. H. "Linguistic Approach to Buddhist Thought". 2^nd ed. Delhi: Motilal Banarsidass, 1992.
• TABER, J. A. "On Nāgārjuna's So-Called Fallacies: A Comparative Approach". Indo-Iranian Journal, Vol. 41, Nr. 3, pp. 213-244, 1998.
• TACHIKAWA, M. "An Introduction to the Philosophy of Nāgārjuna". Delhi: Motilal Banarsidass, 1997.
• TACHIKAWA, M. "Pratītyasamutpāda in the Dedication of the Mūlamadhyamakakārikā". The Adyar Library Bulletin, Vol. 44-45, pp. 639-653, 1980-1981.
• TANAKA, K. K. "Simultaneous Relations (sahabhū-hetu): A Study in Buddhist Theory of Causation". Journal of the International Association of Buddhist Studies, Vol. 8, Nr. 1, pp. 91-111, 1985.
• VALLÉE POUSSIN, L. de la. "Théorie des Douze Causes". London: Luzac & Co., 1913.
• VERHAGEN, P. C. "Tibetan Expertise in Sanskrit Grammar-A Case Study: Grammatical Analysis of the Term Pratītya-samutpāda". Journal of the Tibet Society, Vol. 8, pp. 21-48, 1988.
• VETTER, T. "The 'Khandha Passages' in the Vinayapiṭaka and the four main Nikāyas". Wien: Österreichische Akademie der Wissenschaften, 2000.
• WAGLE, N. K., WATANABE, F. (ed.) "Studies on Buddhism in Honour of Professor A.K. Warder". Toronto: University of Toronto, Centre for South Asian Studies, 1997.
• WATTS, J. D. 1982. "Necessity and Sufficiency in the Buddha's Causal Schema". Philosophy East and West, Vol. 32, Nr. 4, pp. 407-423, 1982.
• WILLIAMS, P. W. "Buddhadeva and Temporality". Journal of Indian Philosophy, Vol. 4, Nr. 3-4, pp. 279-294, 1977.

Abbreviations
• AS   Acintyastava
• MMK   Mūlamadhyamakakārikā
• P.  Pāli
• RĀ   Ratnāvalī
• S.  Sanskrit
• ŚS   Śūnyatāsaptati
• T.  Tibetan
• YṢ  Yuktiṣaṣṭikakārikā.

Publication Dates

• Publication in this collection
  Jan-Apr 2016

History

• Received
  03 Oct 2014
• Accepted
  10 Feb 2015