Nāgārjuna and Śrīharṣa on Dialectical Closure
If philosophical debate is a game, it is not free from manipulation. The parties involved do not come with just their positions but also with the rules that can shift the game in their favor. Vitaṇḍā or frivolous argumentation, for the Nyāya logicians, constitutes a case for disqualification. Vātsyāyana identifies someone adopting this mode of purposeless wrangling as the person who (i) employs destructive criticism with an intent only to destabilize the thesis of the proponent with himself having no thesis to establish, (ii) makes destructive criticism as his thesis, (iii) while he rejects having a thesis but nonetheless makes destructive criticism of his opponent as his mission, or (iv) makes certain positive claims elsewhere while maintaining that he has no thesis of his own.[1] While these arguments are apparently directed towards Nāgārjuna, the dialectic of Śrīharṣa suffers the same criticism a millennium later. He would not even like to accept that he has a thesis, for that would require some form of justification. His argument is, just as a category needs the system of justification for it to be confirmed, so also the system of justification needs external verification. Neither Nāgārjuna nor Śrīharṣa are willing to concede the debate in light of this argument though. Both see virtue in their argumentation, not just the virtue of correct insight but also a soteriology embedded with this virtue. Both these are nonetheless in a logical impasse, and the strategy of Śrīharṣa is identical to that of Nāgārjuna in maintaining his position and not being incapacitated. A few instances from Nāgārjuna’s writings can help clear the air, although it is not possible for me to fully address this issue here. We can particularly gain insight by reading select passages from the Vigrahavyāvartanī (VV), as this will also sheds light on Śrīharṣa’s arguments.[2]
By adopting Searle’s distinction between propositional and illocutionary negations, Matilal (1986, 66-7, 88-9) argues that Nāgārjuna is not simply interested in rejecting the opponent’s proposition. Instead, he negates the very act of making a statement. The issue of how to interpret negation in the philosophy of Nāgārjuna is a thorny one, as evidenced by the Prāsaṅgika-Svātantrika debate in the classical times and Ferraro contra Siderits and Garfield controversy in our time.[3] While the objective here is not to analyze negation, even the position that Nāgārjuna does not maintain a proposition (or, negates the possibility of maintaining any proposition) evokes the same issues. Nāgārjuna proclaims:
If I had any thesis, this consequence would be mine. There cannot be a consequence in my [thesis], as I have no thesis (VV 29).[4]
The issue is, if the rejection of the intrinsic validity of a system of justification were a thesis (either for Nāgārjuna or for Śrīharṣa), the objections of having one’s own unproven thesis, the need for external verification of thesis, a need for something in existence as prerequisite for something to be negated, or similar other objections would be valid. However, as Nāgārjuna proclaims and Śrīharṣa silently adopts, the questioning of the system is not equivalent to the premise that a system relies on external justification. What has been discussed above while analyzing Śrīharṣa’s arguments against the system of justification, without a doubt, is an elaboration of the following position of Nāgārjuna:
If you [consider] that the establishment of the corresponding objects are by means of the system of justification, please explain, how is your means of justification established? (VV 31).
Nāgārjuna raises reductive arguments for both sides of the issue:
- If the position is that the system of justification does not rely on any system for its establishment, then the thesis that categories are established by means of the system of justification is itself rejected.
- On the other hand, if any additional system is introduced to justify the system, it leads to infinite regress.[5]
This chain of arguments directly touches the heart of the classical debate over the scope of the system of justification, with one party arguing that the system that justifies the validity of something also validates itself by the same act of producing veridical knowledge, while the other party making the argument that it is the second mode of justification that confirms the validity of the first mode.[6] Presented differently and for different purposes, these two are the most common arguments found in skepticism East and West. The first negative argument of Hume, for instance, that “all knowledge degenerates into probability (T 180, 1.4.1.1), or “all knowledge resolves itself into probability” (T 181, 1.4.1.4) explores the option of the knowledge system being capable of self-justification. It is not possible to infer something without a prior cognition through perception. This is to say that inference does not support an intrinsically self-justified system. Human reasoning is based on empirical experience. Perception, however, follows the same suit, as it is not free from defects. We have error and hallucination and day-dreaming and many other terms to describe the experiences that are not veridical. To resolve this, even the classical Naiyāyikas developed a two-tiered cognition, with first understanding pramāṇa as a means of veridical cognition, and inference used to confirm what is gained through pramāṇa, understanding pramāṇa as a system of justification. Hume’s argument above can be understood along the same lines, and be presented like “knowledge claims become embedded in belief claims” (Owen 1999). Descartes makes a similar observation that we might be making a mistake in demonstrative reasoning. Both Nāgārjuna and Śrīharṣa cannot agree with a positive claim, as Hume does on this ground, though, that all that could be doubted is to be treated as false. The issue for these philosophers is not to establish falsity but just to reject the validity of knowledge claims based on reasoning and experience.
If the means by which we make our judgments are extrinsically verified, as Nāgārjuna has pointed out, it leads to infinite regress. Hume’s second argument resembles this in him saying that “in every judgment, which we can form concerning probability, as well as concerning knowledge, we ought always to correct the first judgment, deriv’d from the nature of the object, by another judgment, deriv’d from the nature of the understanding” (T 181-82, 1.4.1.5). A general dialectical closure sought, in light of this objection, is that there is no purpose in constantly seeking justification. This, however, hardly resolves the issue.
Classical Hindu and Buddhist philosophical debate provides a platform for a number of justification theories to evolve. Most common among the arguments for intrinsic justification is that a judgment does not rely on another for its verification, but rather, if the knowledge a system has generated is veridical, the system is justified as valid by the same token. The metaphor commonly used is that just as fire illuminates itself while also illuminating other objects, so also do pramāṇas justify themselves while validating some other claims. Nāgārjuna finds this argument unintelligible, as he retorts: (i) there is no instance of the fire not being manifest, for one to make a claim that the fire illuminates itself (VV 34), and (ii) if the fire were to manifest itself it should also burn itself (VV 35). P is called a pramāṇa on the ground that it confirms Q. Something that justifies and the ground on which something is justified cannot be identical. Examples abound in the classical texts, such as a sword cutting itself or a finger pointing itself. Nāgārjuna raises another question on this ground that if a system of justification does not require anything to be justified, this system does not correspond to something outside of itself, turning into a self-referential system, and in effect collapsing the system itself (VV 41).[7] Śrīharṣa’s initial statement, ‘what does it mean to have a system of justification?,’[8] and the subsequent conversation raise the same issue of asking for the meta-categories for a system to exist, upon which a cognition can be considered veridical.
Unlike Hume who returns to ‘the ordinary wisdom of nature,’ pointing to ‘the fallacious deductions of our reason,’ Nāgārjuna and Śrīharṣa describe reality in conventional and absolute terms. Here again, although the two-tier truth theory might look identical, what Śrīharṣa wants to achieve by this, i.e., the singular reality of the Brahman or consciousness-in-itself, is quite different from what Nāgārjuna aims to demonstrate: the absolute truth is that entities are devoid of their self-nature. And it is in this conventional level that a dialectical practice is possible. Two common mistakes people make based on the above presentation are: (1) both Nāgārjuna and Śrīharṣa are not dedicated to a dialectical process, and (2) both these are mystics, who, while rejecting the phenomenal truth, are pointing to something mystical that cannot be grasped by the mind or explained by language. Needless to say, both these arguments are ludicrous. Both philosophers assume that the absolute position, Śūnyatā for one and the Brahman for the other, are confirmed through dialectical reasoning. Both maintain their status in a dialogical platform and engage in a hairsplitting argumentation. Their texts are composed (of course in language) accepting the norms of arguments, and consider the positions of their opponents, while categorically rejecting their claims. For both these philosophers, truth is constantly revealing and it is well within one’s reach to recognize Śūnyatā or the Brahman. This recognition is not something ‘higher’ or transcendental in any sense, and the insight one gains is not ‘hazy’ awareness of some ‘mystical’ experience. Although this truth may not be justified by reason, or the system of justification may fail to ground it, it is nonetheless confirmed through the dialectical process, and the realization of Śūnyatā or the Brahman is not something distinct from dialectical closure. For both, it is the Śūnyatā or the Brahman that provide the foundation for a dialogue.
Empiricus and Śrīharṣa on Methods
Greek philosophers did not recognize skepticism the way we understand it today. It was a way of life that helped its practitioners to suspend judgment in order to achieve an inner tranquility of mind. Skepticism did not arise in Greece as a rejection of the external world, and unlike its contemporaneous counterparts, doubt was not a central piece of skeptical practice in classical Greece (Mates 1996, 5-6). In this regard, the project of Empiricus is not radically different from that of Nāgārjuna or Śrīharṣa. Śūnyatā for Nāgārjuna and the Brahman for Śrīharṣa are not some dogmatic constructs that they defend by means of skeptical arguments. On the contrary, by means of suspending beliefs and questioning the epistemic systems, they find the foundational Śūnyatā or the Brahman unchallenged. There are parallels with “The Five Modes” of Empiricus and the arguments of Nāgārjuna and Śrīharṣa. Borrowing from earlier philosophers, Empiricus outlines that (1) we can reach an unresolvable impasse in a dialectical process due to disagreement, with both sides having an equally compelling argument. The lack of determining argument on one side, a vinigamanāviraha, is a quite common defect in argumentation, used both by Nāgārjuna and Śrīharṣa to buttress their arguments. Following the second argument, (2) infinite regress results when justifying one belief by another, which in turn requires yet another, or one system of justification by another. As has been evident in the previous section, this argument is foundational for both Nāgārjuna and Śrīharṣa in their dialectical practice. Accordingly, (3) things may appear relatively different to different subjects. Although this argument does not come in the sections examined above, it is commonly found in other sections of the works of Nāgārjuna and Śrīharṣa, that entities appear differently for different subjects. Accordingly, (4) when they failed to demonstrate a convincing argument, dogmatists incline to agree on a hypothesis that they deem worthy of accepting without justification. Both Nāgārjuna and Śrīharṣa categorically reject the self-justification of the system of justification. Nyāya philosophers are inclined to accept the pramāṇa system without scrutiny, a hypothesis that is not acceptable to either Nāgārjuna or Śrīharṣa. Eventually, (5) circularity ensues when pramāṇa requires the very pramāṇa for its justification. Śrīharṣa’s opening sentence questions the axiomatic argument that rests on accepted precepts, or that claims to be the bedrock assumption. Although the parallels abound, this is not to argue that the presuppositions on which Nāgārjuna and Śrīharṣa raise similar objections to those of Empiricus are identical. On the contrary, this is only to demonstrate that their methods are similar in kind.[9]
If Śrīharṣa’s methods are after all skeptical, how would he respond to some of the contemporary criticisms? Hilary Putnam, for example, has given an anti-skeptic argument in his chapter, “Brains in a Vat,” which can be paraphrased as:
P1: I do not know that I am not a brain in a vat in an otherwise empty world.
P2: If I do not know that I am not a brain in a vat in an otherwise empty world then I do not know that I am currently drinking water.
C1: So, I do not know that I am currently drinking water (Warfield 1999, 77).
To not propose a thesis, for both Nāgārjuna and Śrīharṣa, does not mean either (i) to maintain doubt, or (ii) to propose a negative thesis. If these two philosophers had to respond to the above arguments, I believe their argument would be something like this:
[Response 1]: I do not have a thesis. [So it is Putnam who is superimposing arguments onto Nāgārjuna and Śrīharṣa. So, there is neither P1 or P2, nor C1].
[Response 2]: The thesis, ‘This is a park,’ grounds on our convention or relational reality (vyavahāra or saṃvṛti). Neither Nāgārjuna nor Śrīharṣa denies that there is such a convention or empirical experience. Nāgārjuna demonstrates that this convention is relational, is a linguistic and cultural construct, and leads to the conclusion that truth is a mere construct, devoid of its own nature. Śrīharṣa, on the other hand, argues that this experience must be grounded on some metaphysical truth, but the way it is experienced and the way it is described cannot be determined by means of justification (anirvacanīya).
[Response 3]: If you say that ‘you are drinking water’ you could not be drinking water, as speaking and drinking are not possible at the same time.
Classical Indian polemics were brutal, and Śrīharṣa could actually say:
[Response 4]: What a moron!
Doubt is not the foundational ground of reasoning for Śrīharṣa. He never says he has a doubt. He is simply demanding justification for the beliefs that his opponents have. His is only the position that ‘since there is no reason for presenting a hypothesis, I have no hypothesis.’[10] And in this regard, his is not a different position from that of Nāgārjuna. This utter restraint from declaring a position, however, has not deterred Śrīharṣa from entering the ring of debate. The argument that a dialectical practice is not possible in the absence of affirming the system of justification is self-defeating because even this very proposition is used in a dialectical process in order to refute the opponent’s rejection of the system.[11]
As is well known, adopting a skeptical method does not make one a skeptic, and not all skeptics are alike. In the case of Śrīharṣa, there is a great resemblance in his arguments with those of Nāgārjuna, and for this reason it is tempting to compare further similarities in Śrīharṣa’s methods with his Buddhist counterpart in particular and also with the Pyrrhonian skepticism for a broader understanding. Śrīharṣa’s project is fundamentally to demonstrate that the world of convention is not determinable (anirvacanīya, not even that it is indeterminable), and for it to be not determinable there is something foundational, sat which also is cit, that is not challenged by the above arguments, as this does not stand as a thesis to be established but is a consequence of a logical reduction. While there is no doubt that he has exploited all the arguments against Nyāya dogmatism, he does distance himself from the Śūnyatā of Nāgārjuna, here, making emptiness as an unfounded hypothesis. Śrīharṣa’s methods, needless to say, are enriched by the insights of Nāgārjuna, and KhKh is filled with instances where he seems more comfortable with the Mādhyamika dialectics than the dogmatic approach of Nyāya. And for this matter, both these philosophers are on the same boat, as far as their methods are concerned.
Bibliography
Annas, Julia. 2011. Intelligent Virtues. Oxford: Oxford University Press.
Barnes, Jonathan. 1990. The Toils of Skepticism. New York: Cambridge University Press.
Berger, Douglas. 1998. “Illocution, No-Theory and Practice in Nagarjuna’s Skepticism: Reflections on the Vigrahavyavartani.” Online Proceedings of the 20th Annual World Congress of Philosophy. see http://www.bu.edu/cp/Papers/AsiaAsiaBerg.htm.
Bhattacharya, Kamaleswar, E. H. Johnston, and Arnold Kunst 2002 [1978]. The Dialectical Method of Nāgārjuna: Vigrahavyāvartanī. Delhi: Motilal Banarsidass.
Chatterjee, Dipankar. 1977. “Skepticism and Indian Philosophy.” Philosophy East and West, (27:2), pp. 195- 209.
Cohen, Daniel H. 2013. “Skepticism and Argumentative Virtues,” Cogency Vol. 5, N0. 1 (9-31).
DeRose, Keith, and Ted A. Warfield. 1999. Skepticism: A Contemporary Reader. New York: Oxford University Press.
Ferraro, Giuseppe. 2013a. “A Criticism of M. Siderits and J. L. Garfield’s ‘Semantic
Interpretation’ of Nāgārjuna’s Theory of Two Truths,” Journal of Indian Philosophy (41) 195-219.
————. 2013b. “Outlines of a Pedagogical Interpretation of Nāgārjuna’s Two Truths Doctrine,” Journal of Indian Philosophy (41) 563-590.
Ganeri, Jonardon. 2001. Philosophy in Classical India: An Introduction and Analysis. London: Routledge.
Granoff, Phyllis. 1978. Philosophy and Argument in Late Vedānta. Dordrecht: Reidel.
Khaṇḍanakhaṇḍakhādya (KhKh). With the Vidyāsāgarī of Ānandapūrṇa and the Khaṇḍanapañjikā of Yogīndrānanda. Varanasi: Chaukhambha Vidyabhawan, 1992.
————.. With Khaṇḍanabhūṣāmaṇi. Ed. Brahmadatta Dvivedi. Varanasi: Sampurnananda University, 1990.
Mates, Benson. 1996. The Skeptic Way: Sextus Empiricus’s Outlines of Pyrrhonism. New York: Oxford University Press.
Matilal, Bimal Krishna. 1986. Perception: An Essay on Classical Indian Theories of Knowledge. New York: Oxford University Press.
————————. 1995. Logic, Language and Reality. Delhi: Motilal Banarsidass.
Mohanty, Jitendra Nath. 1966. Gaṅgeśa’s Theory of Truth: Containing the Text of Gaṅgeśa’s Prāmāṇya (jñapti) Vāda. Delhi: Motilal Banarsidass.
Owen, David. 1999. Hume’s Reason. New York: Oxford University Press.
Ram-Prasad, Chakravarthi. 1993. “Knowledge and the ‘Real’ World: Śrīharṣa and the Pramāṇas,” Journal of Indian Philosophy (21:169-203).
————. 2002. Advaita Epistemology and Metaphysics: An Outline of Indian Non-Realism. London: Routledge.
Siderits, Mark, and Jay Garfield. 2013. “Defending the Semantic Interpretation:
A Reply to Ferraro,” Journal of Indian Philosophy (41) 655–664.
Vākyapadīya. 1966. Vākyapadīya of Bhartṛhari. With the Vṛtti and the Paddhati of Vṛṣabhadeva. Ed. K. A. Subramania Iyer. Poona: Deccan College.
Vigrahavyāvartanī (VV). See Bhattacharya Johnston and Kunst 2002 [1978].
Warfield, Ted A. 1999. “A Priori Knowledge of the World: Knowing the World by Knowing Our Minds.” See DeRose and Warfield 1999.
Westerhoff, Jan. 2009. “The No-Thesis View: Making Sense of Verse 29 of Nāgārjuna’s Vigrahavyāvartanī,” Pointing at the Moon: Buddhism, Logic, Analytic Philosophy, (Eds) Mario D’amato, Jay L. Garfield, and Tom J. F. Tillemans, New York: Oxford University Press.
Notes
[1] For discussion, see Ganeri 2001, 11; Matilal 1995, 16-17.
[2] For a closer analysis of Nāgārjuna’s skeptical position, see Berger 1998.
[3] See Ferraro 2013a, 2013b, Siderits and Garfield 2013.
[4] For analysis of this verse, see Westerhoff 2009.
[5] yadi tāvan niṣpramāṇānāṃ pramāṇānāṃ syāt prasiddhiḥ, pramāṇato ’rthānāṃ prasiddhir iti hīyate pratijñā | . . . yadi punar manyase pramāṇaiḥ prameyāṇāṃ prasiddhis teṣāṃ pramāṇānām anyaiḥ pramāṇaiḥ prasiddir evam anavasthāprasaṅgaḥ | VV, auto-commentary on verses 31 and 32ab.
[6] For the classical Indian system of prāmāṇya, see Mohanty 1966.
[7] While the conversation in VV 41 is primarily regarding the means of cognition and the objects to be cognized by those means, the same argument applies to the system of justification and therefore I have read this verse along the lines of the system of justification.
[8] pramāṇādīnāṃ sattvaṃ yad abhyupeyaṃ kathakena tat kasya hetoḥ? KhKh 6:1-2.
[9] For the Five Modes, see Outlines of Pyrrhonism I.15 (Mates 1996, 110-12). For a detailed analysis, see Barnes 1990.
[10] … kathāprayojakābhāvāt kathā mana na saṃbhavatīti tair bodhanīyam | Raghunātha’s Khaṇḍanabhūṣāmaṇi, 4:17 (see the Dvivedi edition).
[11] pramāṇādisattānabhyupagame kim iyaṃ kathā nopapadyate? kathāntaraṃ vā? nādyaḥ, ārabdhatvād eva | na dvitīyaḥ, vimatāḥ kathāḥ sattānabhyupagamapurassarāḥ kathātvād ārabdhaitatkathāvad iti bhāvaḥ | Vidyāsāgarī on KhKh 12:12-13.
The paper was first published in The Journal of Indian Philosophy (2016) 45:313–329 2014;7:6–24 and has been republished with author’s permission.
Featured Image: Sanskar Hamari