1. Introduction
3.1 Nāgārjuna's Madhyamaka Worries about Absolute Causes in Relation to His Project of
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
3.1 Nāgārjuna's Madhyamaka Worries about Absolute Causes in Relation to His Project of Practice
The middle way, as Nāgārjuna himself characterized in MMK 15.76, is to forsake the side of existence (bhāvānta) and the side of non-existence (abhāvānta) not for the middle, modest position between the two ends but for another different dimension. In the passage of Saṃyuktāgama 12.301 (T02n0099_012; cf. Saṃyuktāgama 10.262), which Nāgārjuna referred to in MMK 15.7, the different dimension concerns how we come to cognize, not how the world exists:
The Buddha told Kātyāyana: “There are two [extremes] that [our conception of] the world relies on: [the idea of] existence and [the idea of] non-existence, which are to be grasped at the contact [of cognitive faculty, object and
consciousness]. Grasped at the contact, the world is thus known via [the idea of] existence or [the idea of] non-existence. Without grasping, the dependent relation between mind and object ceases itself to grasp [the idea], to dwell [in the idea] or to persist in [the idea of] the self. [Without grasping,] let it be when suffering comes; let it cease when suffering goes, and to this one holds no doubt because this is known not via the other. This is thus said to be the right view, and this is called the right view that is established by the Buddha.
Why is that? When we view the arising of the world as how it is, then we do not form the idea of the non-existence about the world. When we view the cessation of the world as how it is, then we do not form the idea of the
existence about the world. That is called the middle way which avoids the two sides....” (P. 630, Vol. 2, Zangyao; P. 85-Down, Vol. 2, No.99, Taisho
6 Luetchford (2002): “When he taught Kātyāyana, the Buddha used the power of existence and non-existence to deny both views: that an innate essence exists and that it does not exist.” The story about the Buddha teaching Kātyāyana is found in Saṃyuktāgama, vol. 12 in Taisho Tripitaka. Cf. Kumārajīva: “佛能滅有無，於化迦旃延，經中之所 說，離有亦離無.” Siderits and Katsura (2013:159): “In “The Instructing of Kātyāyana” both “it exists” and “it does not exit” are denied by the Blessed One, who clearly perceives the existent and the nonexistent”; Siderits and Katsura attributed the passage to the reference in the Kaccāyanagotta Sutta (S II.17, III.134-35).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Tripitaka)7
On the one hand, via the contact of the three parties of cognition, namely, via the realization of cognition, the world arises and ceases; on the other hand, without the realization of cognition, grasping (forming a consciousness) is impossible and hence not only does grasping cease but also existence and non-existence cease. When the arising and cessation of a phenomenon by itself is impossible, i.e., when the world that arises and ceases is viewed correctly as restricted just in our grasping the contact, then “the non-existence of the world by itself (the impossibility of the world by itself without our grasping the contact)” and “the existence and non-existence of the world that are grasped via the contact” entails each other. The middle way is thus
understood as a view to strictly confine the changes (arising and cessation) within the realm of phenomenon, where everything is only possible after the contact has been grasped, i.e., everything is only possible insofar as it is the object of the realized cognition. Meanwhile, the very same view also bears an awareness that without grasping the contact, i.e., without the reflexive realization of cognition, arising and cessation is not possible. In other words, the right view of the middle way demands an introduction of the epistemological consideration (a difference between grasping and not grasping the cognitive contact) into the simple phenomenal duality between existence and non-existence, thereby revealing the lack of the non-phenomenal support for the phenomenal dualism. However, although this passage obviously introduced the other perspective of epistemology – a view with which Nāgārjuna would agree too (MMK 18.4-5), Nāgārjuna's elaboration of this introduced a view that nonetheless struggles only within the phenomenal world for the most part.
Nāgārjuna attempted to highlight this struggle within the phenomenal world in various parts of MMK (representatively MMK 1 and MMK 15) by showing that the
7 佛告[跳-兆+散]陀迦旃延：「世間有⼆種依：若有、若無，為取所觸。取所觸故，或依有、或依無。若無此 取者，⼼境繫著使不取、不住、不計我，苦⽣⽽⽣，苦滅⽽滅，於彼不疑、不惑，不由於他⽽⾃知，是名 正⾒，是名如來所施設正⾒。所以者何？世間集如實正知⾒，若世間無者不有，世間滅如實正知⾒，若世
間有者無有(cf. Saṃyuktāgama 10.262, 如實正觀世間集者，則不⽣世間無⾒；如實正觀世間滅，則不⽣世間
有⾒)，是名離於⼆邊說於中道⋯」
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
concept of phenomenal changes (arising and cessation) and the concept of
non-phenomenal reality deny each other. He almost always adopted the logical apparatus of catuṣkoṭi/tetralemma, a four-cornered exhaustion of logical possibility consisting of “p, -p, both p & -p, neither p nor -p” in Indian logic, to elucidate his argument. By showing that a thesis leads to contradictions in each corner of the catuṣkoṭi, he proved that the thesis could be exhaustively rejected. Next, he could thus show that any possible ontological assertion has to be so rejected. For example (MMK 1.18), given a thesis that things exist, the four corners are that they come into existence either (a) from itself, (b) from others (not from itself), (c) from both itself & others and (d) from neither itself nor others. According to Pingala's explanation,9 first, (a)
contradicts the fact of dependent origination that everything must come into existence in certain multiple conditions, and second, self-origination would invite the problem of infinite regress – infinite repeats of the self's producing itself. Since (a) collapses, (b) collapses as well, because the others have to come into existence first and then their coming into existence fails as we have seen in the case of (a). (c) entails (a) and (b), so (c) cannot stand, either. It is also bizarre to say that things come into existence from nothing, because the uncaused existence would be eternal existence, which does not fit in with the idea of dependent origination and contradicts the idea of “coming into existence.” So (d) falls. Thus, the thesis that things exist is rejected. This method is tricky, because the rejection of the thesis does not imply the automatic establishment or rejection of the anti-thesis. With this method, Nāgārjuna struggled with the phenomenal predicaments and displays whereby every possible ontological assertion (assertions about the non-phenomenal reality) must contradict the
phenomenal reality. Hence, that the concept of dependent origination and the
concept of non-phenomenal reality deny each other is true; it follows then that we do not accept the concept of non-phenomenal reality, insofar as we want the phenomenal
8 Kumārajīva (T30n1564_001): “諸法不⾃⽣，亦不從他⽣，不共、不無因，是故知無⽣.” Luetchford (2002:24):
“Things do not come into existence from self or from others, nor from a combination of both. Yet things are not without cause.” Siderits and Katsura (2013:18): “Not from itself, not from another, not from both, nor without cause:/ Never in any way is there any existing thing that has arisen.”
9 The commentary of Pingala is now only preserved in Chinese translation by Kumārajīva (T. 1564, vol. 30).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
world to be real (conditionally real as the resulting cognition).
We can thus understand Nāgārjuna's middle way to leave the two sides in three steps.
First, existence and non-existence are both recognized and restricted to be
phenomenal (dependent origination) – only as a phenomenon could we know and proclaim whether or not something exists. Second, the phenomenon is not
recognized to possess any non-phenomenal foundation; otherwise, the phenomenon would be unable to change and flux (between existence and non-existence), and this is simply counterintuitive. Third, the first and the second claim one identical truth and entail each other: reality is only phenomenal (“conventional” in Nāgārjuna's term), and thus, ultimately speaking, the phenomenon is empty (without any non-phenomenal foundation) in nature.
From Nāgārjuna's philosophical workings, we can learn that in general, the concerns of the Madhyamaka thinkers about the realism maintained both by the Hināyāna school of existence and by the non-Buddhists center around (1) the worry about the absolute cause of the independent object – together with both of the affected subject and the causal relation in between, implied in the concept of the absolute objectivity, (2) the worry about the absolute cause of the independent subject – together with its productive relation to the resulting cognition and (3) one hidden difficulty with regard to causality itself. Through the texts in Nāgārjuna's early Madhyamaka
criticisms of the svabhāva maintained by Sarvāstivāda in Hīnayāna Buddhism and by Vaiśeṣika in Vedism in the Mūlamadhyamakakārikā (MMK 1&15), not to mention his criticism of the Nyāya epistemology in the Vigrahavyāvartanī, we may better appreciate these Madhyamaka worries in general.
Vaiśeṣika and Sarvāstivāda regarded atoms (paramāṇu-s) with their own-nature (svabhāva) as the indivisible, uncaused and therefore eternal element with which a dharma is constructed (MMK 15.1–2 ), and thus considered them to be the ultimate,
‧
non-phenomenal reality. On the one hand, Nāgārjuna's challenge against svabhāva indicates that one cannot logically maintain both that atoms are real – though we have no way to perceive them – and that the constructions of atoms are real – as we can easily experience them too. On the other hand, this suggests that one cannot maintain both the non-phenomenal reality and the phenomenal reality, given that Nāgārjuna rejected the ultimate reality of atoms (by denying the concept of self-nature,
svabhāva), rejected their serving as the ultimate real objective cause of all dharmas (MMK 1.2 & 15, 20), and then concluded that there is ultimately no arising
(anutpanna, MMK 1.1). Nāgārjuna believed that the concept of the indivisible and eternal self-nature and the concept of dependent origination (pratītyasamutpāda) deny each other (MMK 1.210). With the absolute objective cause being rejected, the realistic causal relation and the affected subject are naturally destined for their own rejection as well (MMK 1.3-14), because these three (the absolute objective cause, the realistic causal relation and the affected subject) imply and define one another and any of them is unconcerted with the concept of dependent origination and the fluxing phenomenal reality.11 Thus the absolute objective cause, together with its causal relation and the affected subject, is altogether found worrisome by Nāgārjuna and among Mādhyamikas. This worry was quite evident and non-controversial, and even shared by Yogācāra.
10 “如諸法⾃性，不在於緣中 (T. 1564, vol. 30).” The self-nature (svabhāva) of all dharmas cannot be found in the conditions. Luetchford (2002:24, presented as verse 3): “These conditions do not attribute an enduring nature to an entity.” Siderits and Katsura (2013:19): “The intrinsic nature of existents does not exist in the conditions, etc.”
11 Assuming the absolute objective cause of all dharmas, a realistic causal relation between the objective cause and the affected cognitive subject are at once assumed as well. We can find both assumptions in the Nyāya epistemology and Sarvāsti-vādins. These two assumptions were rejected by Nāgārjuna with the argument that they contradict the concept of dependent origination. Although Sarvāsti-vādins maintain the idea of no-self (anātman, 無我), the argument for the idea should have been found very inappropriate in the views of Nāgārjuna as well. Sarvāsti-vādins maintain the idea because although there must be some real causal relations among the perceptual faculty (根), consciousness (識) and object (塵) as well (the Abhidharma-kosa 20), the constitution of these elements does not amount to the reality of the self. On the contrary, it is exactly because the self is such a constitution that the self is not an indivisible and eternal ultimate reality (the Abhidharma-kosa 30). However, the causal relation and the affected subject should make no sense, when the cause itself does not stand. The better argument for no self in Mādhyamakas should be the argument that the whole over-simplified model of causation contradicts the idea of the concept of dependent origination and phenomenal reality. One should be aware that Mādhyamakas should not accept the Sarvāstivāda argument for no self. One may refer to MMK 18 for the detailed argument.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Nāgārjuna's rejection of the Nyāya epistemology in Vigrahavyāvartanī12 manifests itself with another particular Madhyamaka worry about the subject as the absolute cause. Unlike the metaphysical-ontological analysis in the aforementioned atomism, the epistemological approaches in Naiyayika focus on the causal relation between the subject (or, to rephrase it in a better term, the faculty of cognition) and the resulting cognition, with the assumption of certain realistic causal relations between the object and the subject and the guaranteed correspondent relations between the object and the resulting cognition via causality. The concept of pramāṇa, which refers to the means of cognition, the faculty of cognition, the cognitive power or the criterion via which the valid cognition is produced, and even the valid cognition itself in the Indian context, takes on a significant focus. With the assumption of realistic atoms (paramāṇu-s) and its realistic causal affection on the cognitive powers, the
Naiyāyikas maintain a real causal relation between pramāṇa-s and their resulting cognition, claiming that the object of cognition (artha, prameya) is proved by/via pramāṇa-s. Hattori (1974:13 406-07) posited that the Naiyāyikas maintain, though there is no real permanent existence of the composed objects of cognition and the composed cognitive faculty, that the real causal account with regard to the
combination (saṃyoga) of the objective atomic elements and the elementary self-nature of cognitive faculties should be admitted, so that the resulting cognition can account for the external existence of object. Here, we do not have to repeat the
rejection of the non-phenomenal reality of the atomic elements, as Nāgārjuna himself did not. Nāgārjuna's other argument against the idea of production in MMK 8, that the producer-product relation is only possible in relativity while the permanent reality of a producer, the permanent reality of a product and the permanent reality of a
process of production can all be rejected just as causation or atoms are rejected in
12 According to Lin (2006: 167), most contemporary scholars maintain that Nāgārjuna's main opponent in
Vigrahavyāvartanī is believed to be Naiyāyikas – Sarvāstivāda may be involved too; according to the Nyāyasūtra (II.1.8 - 20) and Vatsyayana's Nyāya-bhasya, the debates between Nāgārjuna and Naiyayikas indeed took place in history. Here, I follow the English translation from Sanskrit by Kamaleswar Bhattacharya (1978/1998) and the recension by Keiichi Miyamoto (1999) for the Chinese canonic text. For details concerning the development of the text in modern scholarship, refer to Lin (2006:165-6).
13 Translation of Yu-Kwan Ng (1983).
‧
MMK 1, 15, etc., is not difficult to understand. Here, what draws our attention in Nāgārjuna's criticism of Naiyāyikas is his argument on the impossibility of the concept of pramāṇa itself by means of catuṣkoṭi in the Vigrahavyāvartanī . He argued that each pramāṇa cannot be established (a) by itself, (b) by other(s) – no matter how be it by other pramāṇa(s), by the aggregate of the pramāṇa-s or by the result of pramāṇa-s), (c) by both itself and others or (d) by nothing (the
Vigrahavyāvartanī 11),14 so that we should not maintain this idea. The worry here concerns the very definition of pramāṇa. Since pramāṇa is something via which every other possible object is established, it cannot be established by itself. It cannot be established by others because the problem of infinite regress is invited – that which establishes pramāṇa requires further other pramāṇa. Regardless of whether it is established by itself or by others, contradiction of the definition ensues. When the self-establishment cannot stand, needless to say, the establishment by others and the establishment by both itself and others cannot stand either. Saying that pramāṇa is established neither by itself nor by others means that establishment itself is rejected.
Thus, Nāgārjuna concluded that we should not use the concept of pramāṇa; we do not say that cognition is established by pramāṇa because the concept of pramāṇa does not stand.15
An arresting and salient element in Nāgārjuna's analysis of the problem of infinite regress here is the metaphor of the self-illumination of fire – later, we will see that
14 Lin (2006: 169 – 170) contributed a detailed analysis of Nāgārjuna's argument against pramāṇa, including materials from both the Chinese version and Sanskrit version (Johnston & Kunst 1978/1986). The verse (translated in Chinese by Vimokṣaprajñā-ṛṣi and Gautamaprajñāruci): “量非能自成，亦非自他成，非是異量成，亦無因緣成.”
Pramāṇa cannot be self-established, nor be established from the combination of itself and others, nor from other pramāṇa-s, nor from no condition.
15 We can still refer to Yang (1988)'s analysis of the three critics against the realistic pramāṇa theory in the Vigrahavyāvartanī. The first criticism: one cannot criticize the thesis of emptiness upon the realistic position, because the assumption of that criticism is not assumed in the criticized; on the other side, what Nāgārjuna did in his criticism against the realists was simply to point out that the assumption of realism does not fit in with the
phenomenal reality that the realists also intend to admit. The third criticism: the producer (pramāṇa) and the product (prameya) are the conditioned reality (i.e., that which arises and ceases) in relativity (they define each other and are not self-defining); hence, the idea that pramāṇa can produce (establish) cognitions does not stand. The first criticism is important with regard to Nāgārjuna's philosophical methodology but less relevant to our discussion here.
The third can be explained in the second and suggests the same philosophical difficulty. Hence, we only deal with the second criticism here.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
this is a very core issue in Dignāga's epistemology and in the thesis of this chapter:
self-awareness (svasaṃvedana). Nāgārjuna's own fabricated opponent responded to the difficulty of infinite regress by entertaining the metaphor that cognition can cognize itself, just as fire or light can illuminate itself when illuminating others, and this does not invite the difficulty of infinite regress. Nāgārjuna rejected this metaphor (aside from the Vigrahavyāvartanī , also MMK 7 & 10) by pointing out that the metaphor is counterintuitive. If one accepts that fire or light can illuminate itself, the fire or light which has not yet been illuminated must be assumed, but the not-yet-illuminated light itself is a self-contradictory concept. Nāgārjuna thus reiterated that cognition cannot take itself as its own object. We cannot assume the
yet-cognized cognition prior to cognition itself, as much as we cannot assume the
yet-cognized cognition prior to cognition itself, as much as we cannot assume the