A History of Indian Philosophy Volume 2

by Surendranath Dasgupta | 1932 | 241,887 words | ISBN-13: 9788120804081

This page describes the philosophy of the dialectic of nagarjuna and the vedanta dialectic: a concept having historical value dating from ancient India. This is the nineteenth part in the series called the “the shankara school of vedanta (continued)”, originally composed by Surendranath Dasgupta in the early 20th century.

Part 19 - The Dialectic of Nāgārjuna and the Vedānta Dialectic

The dialectic of Śrīharṣa was a protest against the realistic definitions of the Nyāya-Vaiśeṣika, wrhich supposed that all that was knowable was also definable. It aimed at refuting these definitions in order to prove that the natures of all things are indefinable, as their existence and nature are all involved in māyā. The only reality is Brahman. That it is easy to pick holes in all definitions was taught long ago by Nāgārjuna, and in that sense (except for a tendency to find faults of a purely verbal nature in Nyāya definitions) Śrīharṣa’s method was a continuation of Nāgārjuna’s, and an application of it to the actual definitions of the Nyāya-Vaiśeṣika. But the most important part of Nāgārjuna’s method was deliberately ignored by Śrīharṣa and his followers, who made no attempt to refute Nāgārjuna’s conclusions.

Nāgārjuna’s main thesis is that all things are relative and hence indefinable in themselves, and so there is no way of discovering their essences; and, since their essences are not only indefinable and indescribable, but incomprehensible as well, they cannot be said to possess any essences of their own. Nāgārjuna was followed by Āryadeva, a Ceylonese by birth, who wrote a separate work on the same subject in 400 verses. For about two centuries after this the doctrines of Nāgārjuna lay dormant, as is evidenced by the fact that Buddhaghoṣa of the fourth century A.D. does not refer to them.

During the Gupta empire, in the fifth century a.d., Asaṅga and Vasubandhu flourished. In the sixth century A.D the relativist philosophy of Nāgārjuna again flourished in the hands of Buddhapālita, of Valabhl in Surat, and of Bhavya, or Bhāvaviveka, of Orissa. The school of Bhavya was called Mādhyamika-Sautrāntika on account of his supplementing Nāgārjuna’s arguments with special arguments of his own. At this time the Yogācāra school of Mahāyāna monism developed in the north, and the aim of this school was to show that for the true knowledge of the one consciousness (vijñāna) all logical arguments were futile.

All logical arguments showed only their own inconsistency[1]. It seems very probable that Śrīharṣa was inspired by these Yogācāra authors, and their relativist allies from Nāgārjuna to Bhavya, and Candrakīrti, the master commentator on Nāgārjuna’s Mādhyamika-kārikā. Buddhapālita sought to prove that the apprehension and realization of the idealistic monism cannot be made by any logical argument, since all logic is futile and inconsistent, while Bhāvaviveka sought to establish his idealistic monism by logical arguments. Candrakīrti finally supported Buddhapālita’s scheme as against the scheme of Bhāvaviveka and tried to prove the futility of all logical arguments. It was this Mādhyamika scheme of Candrakīrti that finally was utilized in Tibet and Mongolia for the realization of idealistic monism.

In taking up his refutation of the various categories of being Nāgārjuna begins with the examination of causation. Causation in the non-Buddhistic systems of philosophy is regarded as being production from the inner changes of some permanent or abiding stuff or through the conglomeration (sāmagrī) of several factors or through some factors operating upon an unchangeable and abiding stuff. But Nāgārjuna denies not only that anything is ever produced, but also that it is ever produced in any one of the above ways. Buddhapālita holds that things cannot arise of themselves, since, if they are already existing, there is no meaning in their being produced; if things that are existing are regarded as capable of being produced again, then things would eternally continue to be produced.

Bhāvaviveka, criticizing Buddhapālita, says that the refutation of Buddhapālita should have been supplemented with reasons and examples and that his refutation would imply the undesirable thesis that, if things are not produced of themselves, they must be produced by other factors. But Candrakīrti objects to this criticism of Bhāvaviveka and says that the burden of proof in establishing the identity of cause and effect lies with the opponents, the Sāṃkhyists, who hold that view. There is no meaning in the production of what already exists, and, if that which is existent has to be produced again, and that again, there will be an infinite regress. It is unnecessary to give any new argument to refute the Sāṃkhya sat-kārya-vāda view; it is enough to point out the inconsistency of the Sāṃkhya view.

Thus Āryadeva says that the Mādhyamika view has no thesis of its own which it seeks to establish, since it does not believe in the reality or unreality of anything or in the combination of reality and unreality[2]. This was exactly the point of view that was taken by Śrīharṣa. Śrīharṣa says that the Vedāntists have no view of their own regarding the things of the world and the various categories involved in them. Therefore there was no way in which the Vedānta view could be attacked. The Vedānta, however, is free to find fault with other views, and, when once this is done and the inconsistencies of other positions are pointed out, its business is finished; for it has no view of its own to establish.

Nāgārjuna writes in his Vigraha-vyāvartanī thus:

When I have these (of my own to prove),
I can commit mistakes just for the sake (of proving);
But I have none. I cannot be accused (of being inconsistent).
If I did (really) cognize some (separate) things,
I could then make an affirmation or a denial
Upon the basis of these things perceived or (inferred).
But these (separate) things do not exist for me.
Therefore I cannot be assailed on such a basis[3].

Candrakīrti thus emphasizes the fact that it is not possible for the Mādhyamikas to offer new arguments or new examples in criticizing any view, since they have no view of their own to support. They cannot even prove their own affirmations, and, if their affirmations contain any thesis, they quarrel with it also themselves. So the Mādhyamika scheme of criticism consists only in finding fault with all theses, whatever they may be, and in replying to the counter-charges so far as inconsistencies can be found in the opponents’ theses and methods, but not in adducing any new arguments or any new counter-theses, since the Mādhyamikas have no theses of their own. In an argument one can only follow the principles that one admits; no one can be defeated by arguments carried on on the basis of principles admitted only by his opponents.

Things are not produced by any conglomeration of foreign factors or causes; for, were it so, there would be no law of such production and anything might come from any other thing, e.g. darkness from light[4]. And, if a thing cannot be produced out of itself or out of others, it cannot be produced by a combination of them both. Again, the world could not have sprung into being without any cause (ahetutaḥ).

The Buddhist logicians try to controvert this view by pointing out that, whatever a view may be, it must be established by proper proof. So, in order to prove the thesis that all existents are unproduced, the Mādhyamikas must give some proofs, and this would involve a further specification of the nature of such proofs and a specification of the number of valid proofs admitted by them. But, if the thesis that “all existents are unproved” is a mere assertion without any proof to support it, then any number of counterassertions may be made for which no proof need be showrn; and, if proofs are not required in one case, they cannot be required in the other. So one could with equal validity assert that all existents are real and are produced from causes. The Mādhyamika answer to such an objection, as formulated by Candrakīrti, is that the Mādhyamika has no thesis of his own and so the question whether his thesis is supported by valid proof or not is as meaningless as the question regarding the smallness or the greatness of a mule’s horn.

Since there is no thesis, the Mādhyamika has nothing to say regarding the nature of valid proofs (pramāṇa) or their number. But it may well be asked why, if the Mādhyamika has no thesis of his own, should he hold the proposition that all existents are unproduced (sarve bhāvā anutpannāḥ)? To this the Mādhyamika replies that such propositions appear as definite views only to ordinary people, not to the wise. The proper attitude for the wise is always to remain silent. They impart instruction only from a popular point of view to those who want to listen to them. Their arguments are not their own or those which they believe to be right, but only such as would appeal to their hearers.

It is not out of place here to mention that the Mādhyamika school wishes to keep the phenomenal and the real or the transcendental views wide apart. In the phenomenal view things are admitted to be as they are perceived, and their relations are also conceived as real. It is interesting to refer to the discussion of Candrakīrti with Diṅnāga regarding the nature of sense-percep-tions. While Diṅnāga urges that a thing is what it is in itself (sva-lakṣaṇa), Candrakīrti holds that, since relations are also perceived to be true, things are relational as well. Phenomenally substances exist as well as their qualities. The “thing in itself” of Diṅnāga was as much a relative concept as the relational things that are popularly perceived as true; that being so, it is meaningless to define perception as being only the thing in itself.

Candrakīrti thus does not think that any good can be done by criticizing the realistic logic of the Naiyāyikas, since, so far as popular perceptions or conceptions go, the Nyāya logic is quite competent to deal with them and give an account of them. There is a phenomenal reality and order which is true for the man in the street and on which all our linguistic and other usages are based. Diṅnāga, in defining perception, restricts it to the unique thing in itself (sva-lakṣaṇa) and thinks that all associations of quality and relations are extraneous to perceptions and should be included under imagination or inference.

This however does violence to our ordinary experience and yet serves no better purpose; for the definition of perception as given by Diṅnāga is not from the transcendental point of view. If that is so, why not accept the realistic conceptions of the Nyāya school, which fit in with the popular experience? This reminds us of the attitude of the Vedāntists, who on the one hand accepted the view-point of popular experience and regarded all things as having a real objective existence, and on the other hand considered them as false and unreal from the transcendental point of view of ultimate reality. The attitude of the Vedāntists on this point seems to have been directly inspired by that of the Mādhyamikas. The attempts of Śrīharṣa to refute the realistic definitions of the Nyāya were intended to show that the definitions of the Nyāya could not be regarded as absolute and true, as the Naiyāyikas used to think.

But, while the Mādhyamikas, who had no view-points of their own to support, could leave the field of experience absolutely undisturbed and allow the realistic definitions of the Nyāya to explain the popular experience in any way they liked, the Vedānta had a thesis of its own, namely, that the self-luminous Brahman was the only reality and that it was through it that everything else was manifested. The Vedānta therefore could not agree with Nyāya interpretations of experience and their definitions. But, as the Vedānta was unable to give the manifold world-appearance a footing in reality, it regarded it as somehow existing by itself and invented a theory of perception by which it could be considered as being manifested by coming in touch with Brahman and being illusorily imposed on it.

Continuing the discussion on the nature of causation, Nāgārjuna and Candrakīrti hold that collocations of causal conditions which are different from the effect cannot produce the effect, as is held by the Hīnayāna Buddhists; for, since the effect is not perceived in those causal conditions, it cannot be produced out of them, and, if it is already existent in them, its production becomes useless. Production of anything out of some foreign or extraneous causes implies that it is related to them, and this relation must mean that it was in some way existent in them. The main principle which Nāgārjuna employs in refuting the idea of causation or production in various ways is that, if a thing exists, it cannot be produced, and, if it does not exist, it cannot be produced at all. That which has no essence in itself cannot be caused by anything else, and, having no essence in itself, it cannot be the cause of anything else[5].

Nāgārjuna similarly examines the concepts of going and coming and says that the action of going is not to be found in the space traversed, nor is it to be found in that which is not traversed; and apart from the space traversed and not traversed there cannot be any action of going. If it is urged that going is neither in the space traversed nor in the space untraversed, but in the person who continues to go, since going is in him in whom there is the effort of going, then this again cannot be right. For, if the action of going is to be associated with the person who goes, it cannot be associated with the space traversed. One action cannot be connected with both; and, unless some space is gone over, there cannot be a goer. If going is in the goer alone, then even without going one could be called a goer, which is impossible. If both the goer and the space traversed have to be associated with going, then there must be two actions and not one; and, if there are two actions, that implies that there are also two agents.

It may be urged that the movement of going is associated with the goer and that therefore going belongs to the goer; but, if there is no going without the goer and if there is no goer without going, how can going be associated with the goer at all? Again, in the proposition “the goer goes” (gantā gacchati) there is only one action of going, and that is satisfied by the verb “goes”; what separate “going” is there by virtue of association with which a “goer” can be so called? and, since there are no two actions of going, there cannot be a goer. Again, the movement of going cannot even be begun; for, when there is the motion of going, there is no beginning and when there is no motion of going, there cannot be any beginning. Again, it cannot be urged that “going” must exist, since its opposite, “remaining at rest” (sthiti), exists; for who is at rest? The goer cannot be at rest, since no one can be a goer unless he goes; he who is not a goer, being already at rest, cannot be the agent of another action of being at rest. If the goer and going be regarded as identical, then there would be neither verb nor agent.

So there is no reality in going. “Going” stands here for any kind of passage or becoming, and the refutation of “going” implies the refutation of all kinds of passage (niṣkarṣaṇa) as well. If seeds passed into the state of shoots (añkura), then they would be seeds and not shoots; the shoots neither are seeds nor are different from them; yet, the seeds being there, there are the shoots. A pea is from another pea, yet no pea becomes another pea. A pea is neither in another pea nor different from it. It is as one may see in a mirror the beautiful face of a woman and feel attracted by it and run after her, though the face never passed into the mirror and there was no human face in the reflected image. Just as the essenceless reflected image of a woman’s face may rouse attachment in fools, so are world-appearances the causes of our delusion and attachment.

It is needless to multiply examples and describe elaborately Nāgārjuna’s method of applying his dialectic to the refutation of the various Buddhistic and other categories. But from what has been said it may be possible to compare or contrast Nāgārjuna’s dialectic with that of Śrīharṣa. Neither Nāgārjuna nor Śrīharṣa is interested to give any rational explanation of the world-process, nor are they interested to give a scientific reconstruction of our world-experience. They are agreed in discarding the validity of world-experience as such.

But, while Nāgārjuna had no thesis of his own to uphold, Śrīharṣa sought to establish the validity and ultimate reality of Brahman. But, it does not appear that he ever properly tried to apply his own dialectic to his thesis and attempted to show that the definition of Brahman could stand the test of the criticism of his own dialectic. Both Nāgārjuna and Śrīharṣa were, however, agreed in the view that there was no theory of the reconstruction of world-appearance which could be supported as valid. But, while Śrīharṣa attacked only the definitions of the Nyāya, Nāgārjuna mainly attacked the accepted Buddhistic categories and also some other relevant categories which were directly connected with them.

But the entire efforts of Śrīharṣa were directed to showing that the definitions of the Nyāya were faulty and that there was no way in which the Nyāya could define its categories properly. From the fact that the Nyāya could not define its categories he rushed to the conclusion that they were intrinsically indefinable and that therefore the world-appearance which was measured and scanned in terms of those categories was also false. Nāgārjuna’s methods differ considerably from those of Śrīharṣa in this, that the concepts which he criticized were shown by him to have been intrinsically based and constructed on notions which had no essential nature of their own, but were understood only in relation to others. No concept revealed any intrinsic nature of its own, and one could understand a concept only through another, and that again through the former or through another, and so on. The entire world-appearance would thus be based on relative conceptions and be false.

Nāgārjuna’s criticisms are, however, largely of an a priori nature, and do not treat the concepts in a concrete manner and are not based on the testimony of our psychological experience. The oppositions shown are therefore very often of an abstract nature and occasionally degenerate into verbalism. But as a rule they are based on the fundamentally relative nature of our experience. They are never half so elaborate as the criticisms of Śrīharṣa; but at the same time they are fundamentally more convincing and more direct than the elaborate roundabout logical subtleties of Śrīharṣa’s dialectic. It cannot be denied that, based on the dialectical methods of Nāgārjuna, Buddhapālita and Candrakīrti, Śrīharṣa’s criticisms, following an altogether different plan of approach, show wonderful powers of logical subtlety and finesse, though the total effect can hardly be regarded as an advance from the strictly philosophical point of view, while the frequent verbalism of many of his criticisms is a discredit to his whole venture.

Footnotes and references:


The Conception of Buddhist Nirvāṇa, pp. 66—67. Published by the Academy of Sciences of the U.S.S.R. Leningrad, 1927.


sad asat sad-asac ceti yasya pakṣo na vidyate
upālambhaś cireṇāpi tasya vaktuṃ na śakyate.
p. 16.


anyat pratītya yadi nāma paro ’bhaviṣyat
jāyeta tarhi bahulaḥ śikhino ’ndhakāraḥ
sarvasya janma ca bhavet khalu sarvataś ca
tidyam paratvam akhile ’janake ’pi yasmāt.
p. 36.


Mādhyamika-vṛtti, p. 36. See also Stcherbatsky’s The Conception of Buddhist Nirvāṇa, to which the author is indebted for the translation and some of the materials of the last two paragraphs.


Mādhyamika-vṛtti, p. 90, 1 . 6.

Like what you read? Consider supporting this website: