The Pragma-Dialectics of Dispassionate Discourse: Early Ny¯aya Argumentation Theory

: Analytic philosophers have, since the pioneering work of B.K. Matilal, emphasized the contributions of Ny¯aya philosophers to what contemporary philosophy considers epistemology. More recently, scholarly work demonstrates the relevance of their ideas to argumentation theory, an interdisciplinary area of study drawing on epistemology as well as logic, rhetoric, and linguistics. This paper shows how early Ny¯aya theorizing about argumentation, from V¯atsy¯ayana to Jayanta Bhat.t.a, can fruitfully be juxtaposed with the pragma-dialectic approach to argumentation pioneered by Frans van Eemeren. I illustrate the implications of this analysis with a case study from Jayanta Bhat . t . a’s satirical play, Much Ado about Religion ( ¯ Agamad . ambara).


Introduction
Among the philosophies of South Asia still making their mark on contemporary philosophy, the tradition known as "Nyāya" is perhaps one of the most influential. Due in part to the pioneering work of B.K. Matilal, the voices of these philosophers, who made an indelible impact on Indian philosophy from premodern times up to the present day, are increasingly made equal interlocutors to epistemologists and philosophers of mind in the analytic tradition. Yet, despite appreciation of their contributions to what in Sanskrit was termed pramān . avāda (the study of ways of knowing, pramān . a), and in modern English, "epistemology", less philosophical attention is directed towards their work on what today is known as "argumentation theory". This is unfortunate. Firstly, the seminal aphoristic text, the Nyāyasūtra, devotes significant portions to the methodology of debate discourse (kathā): out of its five chapters (ādhyāyas), each sub-divided into two daily lessons (āhnikas), this includes the entirety of the fifth chapter and the second daily lesson in the first. Further, insofar as Nyāya sees itself as a kind of "critical inquiry" (āvīks . ikī) whose written literature engages with interlocutors according to the rules of debate, an appreciation of argumentation and its interrelationship with epistemology is necessary for understanding Nyāya philosophy. Finally, while Nyāya epistemology focuses on the ways in which an individual agent can come to acquire knowledge, the tradition as a whole prizes collaboration, emphasizing social aspects of knowing: from the traditional presumption that learning occurs in debate between student and teacher, to the focus on inference as a set of spoken expressions leading to an interlocutor's coming to know.
In this paper, I argue that early Nyāya debate theory should be understood as concerned with both the pragmatic effects of speech, including persuasion, as well as epistemic aims, such as knowledge of the truth. I show how it overlaps with a modern theory, pragmadialectics, in its way of approaching argumentation, but with an important difference: it is explicitly concerned with truth. Further, Nyāya debate can bolster pragma-dialectics in a key area where the latter is criticized: their focus on rational acceptability which does not take into account justification (or epistemic aims). On the other hand, pragma-dialectics can be a useful framework for understanding Nyāya debate in theory and in practice. I apply pragma-dialectic analysis to an act of Jayanta Bhat . t . a'sĀgamad . ambara (Much Ado misleading objections, and clinchers: from knowledge of these, there is attainment of the supreme good (nih .ś reyasa). NS 1. 1.1. 2 The relationship between discussion, disputation, and wrangling is subject to some controversy, in particular whether wrangling is a kind of disputation or not. However, Table 1 summarizes the basic commitments in our early Nyāya commentaries. Discussion is a kind of debate in which two opposing claims are evaluated by an exchange of reasons, in contrast with the other types, in which "tricks" or deceitful uses of fallacious moves may be deployed in order to acquire victory over one's interlocutor. The NS itself, and generations of subsequent commentators, make explicit that the opening sūtra is not merely a disjointed list of things to know, but an interrelated set of topics. Understanding and deploying the cognitive tools enumerated here depends on an appreciation of the parts as well as the study of Nyāya as a whole. Thus, from the outset, epistemic instruments (also called "knowledge sources" or "ways of knowing", pramān . as) are paradigmatically ways to know objects with a moral valence: what is to be avoided or obtained, in particular, in regard to ethical living and the cessation of suffering in the cycle of rebirth, by obtaining liberation (apavarga). Jayanta Bhat . t . a differed from earlier commentators such as Vātsyāyana about Nyāya's relationship to the other disciplines (śāstras) and human aims such as liberation. As (Kataoka 2006) showed, he diverged from the tradition by arguing that Nyāya's role was primarily a defensive one, providing argumentative and epistemic tools for removing doubts about Vedic authority, rather than being a unique way of knowing the objects listed in NS 1.1.1. Still, all of these philosophers discuss the relationship between knowingand its contrary, misapprehension (mithyājñāna)-and ending the cycle of rebirth which Naiyāyikas accept, of one living self into another body, such as at NS 1.1.2. 3 Reasoning together is one way in which misapprehension can be lifted-whether this happens by hearing another person's inference without challenging it (as in the case of "inference-foranother",) or by two agents taking up position and counter-position, then engaging in debate. While discussion may be paradigmatic for the removal of misapprehension (see also NS 4.2. [47][48], it is possible to employ other means to remove faulty ideas. Any of the ways of knowing, such as perception, can lead to knowledge which then corrects faulty ideas. For instance, a person misapprehending a tall object in the distance as a post, comes to know it is a person from genuine perception at a closer position (NBh 4.2.35). However, for controversial philosophical issues, such as the existence of an eternal, substantial self (ātman), discussion and inference are crucial.
It may be true, as Nicholson (2010) said, that there is "sedimentation" in the NS (p. 76) of a prior textual layer of an agonistic debate manual. As a result, perhaps there is tension between the rhetorical emphasis on norms of debate performance and the epistemic emphasis on norms of debate rationality. However, both at the level of the sūtra text and the commentators, there is a serious attempt to bring these together. One interpretation of the original sūtra structure was, with Potter, that the material on debate is mainly useful as a way to "illustrate" theories of inference treated elsewhere (Potter 1977, p. 208). However, this leaves unexplained why debate and its main types are introduced as early as the first chapter. With (Matilal 1998), rather than seeing an incoherence presupposed by the structuring of the sūtra, we might see the compiler as introducing the types of debate early, "in its natural home, in the context of the discussion of the pramān . as, means of knowledge" (p. 58). 4 The separation of this from the fallacious "sophistries" in the fifth book is then a way to keep readers focused on the work's main focus: "the acceptable and sound method for philosophical discourse" (p. 58). However, the historical origins of the sūtra's organization will not be of concern. Rather, the focus is how a later philosopher, Jayanta Bhat . t . a, understood the epistemological project of Nyāya as bound up with its project of setting up reasonable rules of debate.
This brings us to Nicholson (2010, p. 78) argued that later commentators "find it increasingly difficult to justify the relevance of the rhetorical techniques discussed in the older parts of the NS to the epistemological cum soteriological project" (p. 78). 5 Here, by "rhetorical techniques", he is referring to those used in purely agonistic forms of debate, in which two interlocutors deploy tricks in order to be counted as winner, whether or not their argument has genuine merit. His thesis was that the paradigmatic example of a truth-seeking discussion is didactic debate, in which a performative "debate" is staged between a teacher and a student (p. 78). On his view, there is an inherent instability to discussion for Naiyāyikas. They allow for some points of defeat in discussion. Points of defeat (nigrahasthanas) are situations in which an interlocutor is judged as failing to meet some expectations in the content or expression of her reasoning. Forcing an opponent into a point of defeat could be a strategy for winning a debate, as would using them without being caught (if one's use of such a point is identified by an opponent, that in itself is a point of defeat, as is an opponent's failure to so identify such a point). For instance, stating something irrelevant (arthāntara, NS 5.2.7) is a point of defeat which would not be allowed in discussion. In a disputation, in contrast, if such an irrelevant remark suited one's pragmatic goals of winning, it should be used, likewise for a range of other fallacies. However, the points of defeat are characterized as flaws in the sūtra text. A debater ought to avoid incurring them. So, how can a subset of flaws be acceptable in discussion if these are suboptimal for coming to the truth?
As Nicholson (2010, p. 86) observed, "a debate without the possibility of defeat is no longer a debate, but something else", and so the "basic adversarial nature of debate" is preserved by early Naiyāyikas by allowing some select points of defeat. However, as noted earlier, on his view, Nyāya discussion is paradigmatically non-agonistic for historical reasons. From a conceptual perspective, though, in considering debate, we should distinguish two questions:

1.
Is it possible to justify the use of flawed reasoning (e.g., points of defeat) in argumentation which has truth as its aim? 2.
Is it definitive of argumentation aiming at truth, that it be non-adversarial?
Answering negatively to (1) does not require doing so for (2). To put it explicitly, it is possible to engage in agonistic debate in pursuit of the truth, provided that there are rational norms in place to which both parties assent. This is a broadly Gricean point: communication is a cooperative endeavor, even if two parties are at odds in other ways. Todeschini (2010) made this point clearly in his discussion of the points of defeat, citing Searle: "In order to be fighting at one level we have to be cooperating in having a fight at another level" (p. 71).
If agonistic truth-seeking debate is conceptually possible, then the question becomes whether any Naiyāyikas defended such an understanding of discussion. Jayanta Bhat . t . a's work provides an interesting case study for this question, as in his Nyāyamañjarī (NM), he seemed to align with the requirement that truth-seeking argumentation be didactic, not adversarial, and yet in his play, theĀgamad . ambara (ĀD . ), he included a debate explicitly presented as discussion (vāda), which is not between a teacher and a student. Rather, it is an adversarial debate, before a watching audience, between members of opposing traditions-a Buddhist and a Mīmām . sāka. As we will see, tensions still remain in the relationship between adversariality and truth-seeking, though these are implicit in the play.

Discussion's Goals and Methods
Jayanta, in his opening remarks in the NM on the purpose of Nyāya, related suppositional reasoning (tarka) and certainty (nirn . aya), two categories mentioned in NS 1.1.1., to discussion (vāda). 6 On his understanding, ascertainment of the truth of a position, or certainty (nirn . aya), is the goal of reasoning (nyāya). Certainty (nirn . aya) is defined at NS 1.1.41 as "the ascertainment of a fact (artha) by both standpoint and contrary standpoint, after having deliberated". 7 An agent desires that her inquiry ceases with a settled position on a subject. Thus while certainty has a subjective epistemic element-one might characterize it as a person's maximally strong confidence in the truth of a given claim-it is supposed to be responsive to objective features in the world. 8 In the context of discussion, certainty should be an attitude a person has towards settled views, those which have been positively supported by good reasons and whose contraries have been defeated in the same manner. Yet it is not a kind of Cartesian indubitability, objectively immune to future doubt, since a person can take up debates again, in new contexts, with new speakers.
Further, although inference is a tool for coming to know, it is not equivalent to, nor does it guarantee, certainty. After all, many doubts arise about our reasoning in conversation with others. It is only when doubts have been responded to, and the debate is over, that an inquirer attains certainty. 9 Again, while we must be cautious not to impute necessity to debate for Nyāya-it is possible to ascertain the truth without engaging in debatecontroversy is likely to arise given the reality of opposing viewpoints in the world, and debate practices grounded in epistemic norms, which distinguish between good and bad reasoning, become crucial. I will illustrate this with Jayanta's play, in which debate plays a defensive role against threats to Vedic authority which might otherwise cause ordinary people to have doubts. First, I show how the persuasive and rational elements of Nyāya debate are brought together in Jayanta's NM:
Discussion's persuasive strategies are context-sensitive, thereby subject to pragmatic analysis, but also rationally constrained, thereby subject to epistemic norms.
The first claim focuses on the importance of persuasion to the epistemic goals of discussion. In contrast to argumentation that aims only at consensus or resolving a dispute, Nyāya aims for participants to acquire knowledge. The second claim focuses on the importance of pragmatic analysis of speech acts in context. While the tension between what Greek and other traditions call "rhetoric" and "dialectic" appears also in Nyāya thought as we have seen above, I argue that Naiyāyikas seek to incorporate rhetorical methods within what is often called an "epistemic approach" to argumentation (e.g., Biro and Siegel 2006). 10

Discussion and Other Debate Categories
What are these debate practices, then, and how are they related to the epistemological project of Nyāya? Below is a brief summary of Jayanta's account. A wealth of further details are found in (Solomon 1976;Tripathi 2021).
Discussion (NS 1.2.1) is taking up a standpoint and a contrary standpoint, a taking up whose supporting and criticizing is by epistemic instruments and suppositional reasoning, which does not contradict settled claims, and which is obtained by the five members of inference. 11 Discussion (NM) is dispassionate discourse which results in certainty regarding a matter. 12 The kind of discourse which is paradigmatic for Nyāya philosophers is "discussion", also translated as "debate for the truth". It is defined as having both a positive position or standpoint (paks . a) as well as an opposing position (pratipaks . a), which are subject to evaluation. It involves both inference, an epistemic instrument, and suppositional reasoning, which is not a way of knowing but an auxiliary useful for critical investigation into the entailments of competing claims. Jayanta added that discussion is a kind of "dispassionate" (vītarāga) discourse. This term is loaded with soteriological connotations, though it might be glossed as "sober-minded" and "impartial", insofar as it characterizes an agent whose concern is the truth and not personal victory over an opponent. Finally, Jayanta emphasized the result of certainty-that is, ascertaining in a definitive way, against the backdrop of doubt, what one ought to believe on a matter.
While dependence on discussion, implied by the reference to "standpoint and contrary standpoint" is paradigmatic for acquiring certainty, it is not a necessary condition, since from Vātsyāyana onward, epistemic instruments such as perception are also declared to be ways of gaining certainty. However, when some matter is doubtful and two positions are at issue, then the goal of discussion is to acquire certainty, and once certainty is acquired by one of the participants, the discussion ends-Uddyotakara made this explicit in his commentary. 13 Jayanta even added that, sometimes, two people who enter a discussion without themselves having doubts will come to doubt midway through the conversation, since the reasoning of their opponent may be strong. 14 In order to acquire certainty about some matter by discussion, certain procedures must be followed, in particular the use of rational means to support one's own position and to criticize the opponent's position. Paradigmatically, this occurs through the "great sentence" (mahāvākya) that presents an inference, an expression which is composed of five components (defined at NS 1.1.33-39). 15 Further, these five components are together necessary and sufficient for coming to know, as evidenced by the existence of two points of defeat known as "insufficiency" (nyūna) and "excess" (adhika). Lacking one of the parts constitutes insufficiency and would force defeat, and likewise for adding additional components, such as extra supporting examples or reasons. 16 Epistemically speaking, a person who understands an inference which is well-formed and consists in genuine reasons will come to know the inferential conclusion, or the thing to be proven (sādhya). Further constraining the inference is that it ought not have a "pseudo" or "bogus" reason, known as a hetvābhāsa. The reason given ought to have a genuinely supporting relationship with the thesis being set forward. 17 However, within the broadly externalist epistemic framework of Nyāya philosophy, that one comes to know p does not entail that they also know that they know p, so that the mere presence of a genuine inference in a debate may not end the discussion. It proceeds until the participants in the debate are no longer in doubt about whose position is correct. Therefore discussion aims for not just first-order knowing (knowing that p) but higher-order knowing, knowing that one knows p.
While not an epistemic instrument itself, suppositional reasoning (tarka) is an auxiliary; it is especially useful for criticizing an opponent's position by considering whether the entailments that follow are in conflict with accepted facts. Jayanta Bhat . t . a explained suppositional reasoning's purpose as being to bring about clarity of mind. 18 As Kang (2010, p. 16) pointed out, Jayanta characterized suppositional reasoning as a means of "clarification of the object" (vis . ayapariśodhana) insofar as it rules out one of two options. Essentially, suppositional reasoning excludes an option on the basis of reasoning, but not by further understanding of the facts of the matter themselves, which is the role of epistemic instruments. This process of exclusion by suppositional reasoning must have an end, as Jayanta ironically notes, in a section focusing on why this term is treated separately in NS 1.1.1.
When two positions are considered equal-making determination of the topic doubtful-by making one of two positions weak, suppositional reasoning favors abandoning the position opposite from the one which is free from problems, the content belonging to the other position is understood to be an epistemic instrument (pramān . a). Therefore suppositional reasoning is treated separately. And suppositional reasoning is employed in discussion to make clarity of mind appear. Thus suppositional reasoning does not stop without certainty of the authority of one of the two positions, since certainty's function is causing the completion of reasoning. On the other hand, without a limit, its result remaining unaccomplished, what person would undertake reasoning? 19 Another important restriction on discussion is that it does not contradict settled claims (siddhāntāviruddha). Precisely what this amounts to is a matter of controversy among early commentators. Vātsyāyana argued that its mention excluded from a discussion a kind of fallacy known as "contradiction" (viruddha) introduced at NS 1.2.6 (See G. Jha 1984, vol. 1, p. 482). This kind of error, which is specifically a failure of the reason component of the five-member inference, occurs when someone puts forward a thesis that some a is an S and then, in their reason, explicitly claims that a is not an S or (more likely) claims something that entails such a commitment. Essentially, he takes the sūtra's statement to preclude putting forward reasons which have such a flaw. In contrast, Uddyotakara argues that the restriction is associated with a point of defeat known as "in opposition to what is settled" (apasiddhānta). This point of defeat, found at NS 5.2.23, is broader than the fallacy of contradiction, as it is defined as an interlocutor's engaging in debate without qualifying their comments so as to avoid contradiction with something which has been established (See G. Jha 1984Jha , vol. 4, p. 1769. Uddyotakara distinguished this from contradiction by noting that in this case "one is defeated because they give up an idea that has been settled, with the exception of the idea expressed by the thesis". 20 Jayanta noted this controversy and sided with Vātsyāyana, stating that Uddyotakara's interpretation was inapt (anupapanna). 21 He argued that not all the points of defeat should be used to defeat another person in discussion. For instance, an opponent being momentarily embarrassed or without an idea (apratibhā) is not a reason to declare victory over them. Instead, only those particular points of defeat which are relevant to the argument itself should be employed. Further, in the context of discussion, accidental errors do not bring about the end of the conversation. Jayanta noted that "pointing them out is necessary, because when they are not pointed out, the subject matter would not be completely justified". 22 Thus the relevant question is not whether an interlocutor in a discussion ought to strive for consistency-they should-but whether this results in an end to the debate. Jayanta seemed to think that only when a person had abandoned their thesis, which is inconsistency in the narrow sense, that debate should end, but not when one failed in mere performative matters. 23 Otherwise, pointing out inconsistencies which arise in the course of discussion is useful for truth-seeking, but need not end the conversational exchange. The goal is not to defeat an opponent due to irrelevant flaws, but to come to know the truth.
Finally, in discussion, along with the general avoidance of forcing points of defeat, participants are to avoid chala and jāti. Chala is sometimes translated as "equivocation" (Dasti and Phillips 2017) or "casuistry" (G. Jha 1984), and jāti as "misleading objections" (Dasti and Phillips 2017) or "futile rejoinders" (G. Jha 1984). In his opening remarks, echoing its definition at NS 1.2.10, Jayanta defined chala as "opposition to an expression by an alternative meaning". 24 Whether intentional or unintentional, a chala is an argumentative flaw insofar as one person misascribes a meaning to another person's speech, and then proceeds to raise an objection to that other meaning. A jāti, in contrast, is "an objection which has some similarity with or difference from the reason". 25 While the original NS at 1.2.18 (and indeed, Jayanta's initial gloss) does not state that the similarity or difference is faulty, this is the understanding of jāti in subsequent tradition. 26 The objection depends on some irrelevant comparison for its force. Both of these-chala and jāti-are excluded from discussion as a matter of definition.

Frans van Eemeren's Pragma-Dialectic Argumentation Theory
Turning from 9th century India to the Netherlands in the twentieth century, another theory of argumentation known as "pragma-dialectics" begins with the work of Rob Grootendorst and Frans van Eemeren. 27 It is within the purview of argumentation theory, an interdisciplinary effort to analyze argumentative discourse in both its normative and descriptive aspects. Participants in this effort include philosophers, logicians, commu-nication theorists, linguists, lawyers, and psychologists. Pragma-dialectics understands argumentation as, paradigmatically, a "critical discussion", or a discussion which aims to resolve of a difference of opinions. Grootendorst and van Eemeren's work analyzed the strategic maneuvers that participants can use in order to persuade their interlocutors within reasonable restrictions. Strategies are context-dependent and will vary depending on conversational purposes, audience, and the interlocutors themselves.
Put succinctly, their definition of argumentation was: Argumentation is a communicative and interactional act complex aimed at resolving a difference of opinion with the addressee by putting forward a constellation of propositions for which the arguer can be held accountable in order to make the standpoint at issue acceptable to a rational judge who judges reasonably.
(van Eemeren 2018, p. 21) As a theory, pragma-dialectics aims to analyze actual arguments between interlocutors, and takes as its starting point differences of opinion-whether occurrent conflicts or those which are anticipated (p. 21). In contrast to approaches which abstract from contexts of utterance, pragma-dialectics focuses on understanding the argumentative functions of language (broadly construed to include linguistic, visual, and other systems of signs). This descriptive component of the approach is coupled with a normative aspect, an analysis of the rules for the speech acts made in a particular argumentative context, which they call "reasonableness conditions" (p. 29). These rules constrain the speech acts to those which resolve the difference of opinion on the merits. Importantly, there are two levels of constraints: Problem-validity. A property of a set of rules which are jointly necessary and sufficient for excluding fallacious argumentative speech acts.

Conventional validity.
A property of a set of rules which are intersubjectively acceptable to the participants in a critical discussion. 28 Already, some overlap is observable between Nyāya debate theory and pragmadialectics. Rather than concern with an abstracted set of propositions, both are interested in complex speech acts which unfold over time between two parties. Naiyāyikas characterize inference as a "large sentence" (mahāvākya), suggesting a similar conception of argumentation. Further, they are concerned with having all and only the necessary components of this speech act, evidenced in the flaws of insufficiency and excess. At the same time, they recognize that argumentation is more than competing inferences. The points of defeat, chala, and jāti, attest to the existence of other speech acts in the context of the debate, which are explicitly illustrated in Jayanta's play. Take the point of defeat ananubhās . ana, inability to reiterate an opponent's statement. It presupposes that reiteration is part of the debate process (Todeschini 2010, p. 63). While in Nyāya discussion, such a failure would not close the debate, since points of defeat are generally excluded. This does not mean that reiteration was not part of the expected discourse. There are also suggestions that speech acts other than argumentative assertions are expected in Nyāya discussion. Another point of defeat is when an opponent fails to point out what is objectionable, or does not point out that their interlocutor has committed a fallacy (p. 65). This attests to the expectation of "commissives", or a speech act that commits a speaker to an action (p. 40).
Fallacies (in Nyāya terms, points of defeat, pseudo-reasons, chala, and jāti) are excluded by pragma-dialectics because they violate a set of agreed-upon rules which are problemvalid and conventionally-valid. Pragma-dialecticists generate a set of rules which they find appropriate for resolving a difference of opinion "on the merits", or in a reasonable manner. The aim in such a discussion-the above-named "critical discussion"-is for both parties to agree on the "acceptability of the standpoint at issue by finding out whether the one party's standpoint is tenable against the other party's doubts and criticisms in light of the mutually accepted starting points" (p. 35). Therefore they formulated ten rules for reasonable resolution of a discussion: 29 1.
Freedom rule: discussants may not prevent each other from advancing standpoints or from calling standpoints into question; 2.
Obligation to defend rule: discussants who advance a standpoint may not refuse to defend this standpoint when requested to do so; 3.
Standpoint rule: attacks on standpoints may not bear on a standpoint that has not actually been put forward by the other party; 4.
Relevance rule: standpoints may not be defended by non-argumentation or argumentation that is not relevant to the standpoint; 5.
Unexpressed premise rule: discussants may not falsely attribute unexpressed premises to the other party, nor disown responsibility for their own unexpressed premises; 6.
Starting point rule: discussants may not falsely present something as an accepted starting point or falsely deny that something is an accepted starting point. 7.
Validity rule: reasoning that is in an argumentation explicitly and fully expressed may not be invalid in a logical sense; 8.
Argument scheme rule: standpoints defended by argumentation that is not explicitly and fully expressed may not be regarded as conclusively defended by such argumentation unless the defense takes place by means of appropriate argument schemes that are applied correctly; 9.
Concluding rule: inconclusive defenses of standpoints may not lead to maintaining these standpoints and conclusive defenses of standpoints may not lead to maintaining expressions of doubt concerning these standpoints; 10. Language use rule: discussants may not use any formulations that are insufficiently clear or confusingly ambiguous, and they may not deliberately misinterpret the other party's formulations.
Todeschini (2010) has already done significant work showing how Nyāya points of defeat can be intelligible as violations of Gricean maxims, in particular the first: "Make your contribution as informative as is required (for the current purposes of the exchange)" (p. 68). But in the pragma-dialectic approach, we find a narrower analysis of the ways in which, as Todeschini said, the "debater fails to contribute as required". For instance, the language use rule (10) speaks to the manner of one's contribution, in particular Grice's maxim: "Avoid obscurity of expression" (Todeschini 2010, p. 68). We may note, however, some ways that Nyāya discussion goes beyond pragma-dialectics: chala is a fault whether one deliberately or accidentally misinterprets what has been said when attempting an objection, in contrast to the deliberate misinterpretation in rule 10. Since misinterpretation, whatever its source (ignorance or strategy), obstructs a resolution on the merits, the Nyāya version of (10) would strengthen the sufficiency of the set of rules in excluding fallacies.
In the same spirit, the starting point rule (6) corresponds to the Naiyāyika emphasis on consistency with a debater's settled claims or siddhānta, which are typically conceived of as tradition-or school-wide. The Nyāya version is also stronger: a debater may not contradict their settled claims. As I argued earlier, in Jayanta's discussion of "non-contradiction with what is settled", these starting points need not play a particular role within the inference (the thesis), nor should an inconsistency put an end to the discussion. However, once such a rule-violation has been discovered, it ought to be addressed, again, to keep the argumentative discourse aimed at the merits of the standpoint under investigation. This paper will not attempt to align pragma-dialectic rules with every Nyāya fallacy (broadly conceived, so including pseudo-reasons, points of defeat, chala, and jāti). This can be left to future research. 30 For now, the main contours of theoretical overlap between the two methods are sufficiently clear: both have normative and descriptive elements, and both are concerned with argumentation as the unfolding of a complex communicative act in time. Certainly Nyāya debate is narrower than the argumentative discourses which are the focus of pragma-dialectics (as the latter does not only focus on formal debates). This difference notwithstanding, the two approaches could be mutually supportive. Before concluding with a case study from Jayanta's play, I will identify two further aspects of pragma-dialectic theory that are important for this analysis: its identification of argumentative stages and its distinction between truth and acceptability.
Pragma-dialectic argumentation which is "on the merits"-that is, following the ideal model and set of rules described above-is analyzed in terms of stages. These may not be entirely sequential. Discourse analysis employing pragma-dialectic approaches reconstructs the elements of a discourse which correspond to these stages: confrontation, opening, argumentation, and concluding (discussed in van Eemeren 2018, pp. 36-38). See Table 2 for a summary. For instance, at the confrontation stage, there is some difference of opinion which becomes apparent. This may include two people taking up opposing positions on a single proposition (p, not p), which is the standard catalyst for Nyāya discussion. However, pragma-dialectics distinguishes further between single and multiple differences of opinion: "single" includes disagreement about one proposition, whereas "multiple" ranges over more. It also distinguishes between mixed and non-mixed differences, where a "mixed" difference of opinion is opposition, or two standpoints (p, not p), and a "non-mixed" includes only one standpoint (p, is it true p?). The Nyāya category of "wrangling" (vitan . d .ā ) would be a kind of non-mixed difference of opinion, since it includes a proponent of a standpoint and an interlocutor who only challenges that standpoint but does not commit to its opposite. Unlike in pragma-dialectics, this is a deviant form of debate, one which not only lacks a counter-standpoint, but which is understood to allow non-rational means of resolution. Table 2. Stages of argumentation in pragma-dialectics.

Confrontation
The existence of a difference of opinion, doubt, or contradiction regarding standpoint(s).

Opening
The assignment of participant roles (protagonist arguing for p, antagonist arguing not p).

Argumentation
The deployment of moves with the goal of convincing, defending.

Concluding
The protagonist and antagonist determine resolution to difference of opinion.
The subsequent stages include (2) opening, taking particular roles vis-à-vis the standpoint(s), (3) argumentation, engaging in argumentative moves, and (4) conclusion, coming to an agreement about the tenability of the standpoints under discussion. The existence of stages in pragma-dialectical analysis emphasizes not only the temporality of argumentation as a communicative act-in contrast to an abstracted set of propositionsbut also the necessity of certain components for resolving an argument on the merits (van Eemeren 2018, p. 36).
For instance, the confrontation is the required trigger for critical discussion, just as the existence of opposition (pratyanīka) is necessary for triggering Nyāya discussion. 31 However, as I have shown, opposition is narrower than confrontation, as it requires two standpoints (p, not p). Naiyāyikas argue that even in so-called "destructive" debate or wrangling (vitan . d .ā ), where no counter-position is explicitly stated, one is implied by the questioning of p. In Jayanta's words, "What belongs to the destructive debater, who is the speaker of the later standpoint, as it is dependent on putting forward an earlier standpoint, by the maxim of 'one hand and the other hand', is said to be a counter-standpoint", concluding that the counter-standpoint is implied. 32 As with pragma-dialectic analyses, Naiyāyikas are interested not only in what is explicitly stated, but what is hinted or suggested (Jayanta used the verb aks . ip, meaning "imply") in their analysis of what is actually occurring in a debate context. 33 Unlike pragma-dialectics, though, the combination of a merely implied position not-p and explicit criticism of p is associated with irrational means for its resolution, a point that Solomon (1976, p. 115) lamented, andNicholson (2010, p. 92) explained as a hangover from earlier textual material being incorporated into the tradition. 34 This is because wrangling is defined as a subtype of disputation, which is characterized by its methodology: win at whatever cost, even if this involves fallacies. However, as Nicholson pointed out, at least one early Nyāya philosopher (Sānātani, mentioned in Udayana's commentary on the NS) accepted the possibility of a truth-seeking discussion where the contrary position is implicit (p. 92). Jayanta himself said little about wrangling, merely clarifying that there is an implicit position, but that there is no genuine support (sādhana) for it. 35 He accepted that it was a subtype of disputation, which aimed at victory not truth-and thus its participants would employ whatever rhetorical means were at their disposal, regardless of their rationality. 36 The other three stages are also necessary. The opening state in which interlocutors assign roles and identify starting points ensures the possibility of resolution, since, without some shared common ground, engaging in critical discussion would be pointless (van Eemeren 2018, p. 37). Argumentation is the persuasive portion of the discourse, in which argumentative speech acts are exchanged. These are necessary in order for rational persuasion on the merits to occur, not mere agreement by some other means, such as drawing straws. Finally, without a concluding stage, the difference of opinion remains unresolved.
As noted above, a pragma-dialectic approach distinguishes between problem-valid and conventionally-valid rules, arguing that "in argumentative discourse more often than not it is not the truth of standpoints . . . that is at issue, but their acceptability . . . not a final justification . . . " (p. 29). On this understanding, the conventionally-valid rules contribute to the intersubjective agreement on whether the difference of opinion has been satisfactorily resolved, and the problem-valid rules are a way to ensure that the agreement has some rational basis-that it occurs in virtue of "appealing to [the listener's or reader's] reasonableness" (p. 30). However, by explicitly eschewing "final justification", pragmadialectics leaves itself open to the charge that the approach is unmoored from rationality in a robust sense. This approach is in striking contrast to Nyāya discussion which aims at certainty (nirn . aya) and truth (tattva).
In addition to criticizing van Eemeren's (and Grootendorst's) reading of Popper which led them to reject "final justification", Siegel and Biro (2008) argued that their account of "reasonableness" was inadequate. Their imagined counterexample is a group of white racists coming to agreement, through a critical discussion, that one of them should not vote for a black candidate. The critical discussion entirely accords with their intersubjectivelyagreed-upon pragma-dialectical rules. Still, while the conclusion comes about through conventionally-valid (agreed-upon) rules and problem-valid (non-fallacious) rules, Siegel and Biro concluded that the resultant belief: "I should not vote for a black candidate" was not in fact justified by the racist beliefs that the interlocutors hold (p. 194). They diagnosed van Eemeren's hesitancy with "final justification" as, at least in part, a worry that no resolution could ever be entirely immune to further investigation, and thus that any stopping point is arbitrary. What is needed, they argued, is the possibility of a non-arbitrary stopping point in a critical discussion which provides positive justification for a standpoint (p. 200).
Here is where Nyāya discussion, with its conception of truth-seeking debate, provides a view of justification which is consistent with fallibilism: a person could erroneously believe they have settled a disagreement on the merits when they have not drawn a rational conclusion. At the same time, Naiyāyikas are adamant that truth is the long game for discussion. As Jayanta put it, considering the usefulness of suppositional reasoning (note 9 below), "On the other hand, without a limit, its result remaining unaccomplished, what person would undertake reasoning?" The result is not mere agreement, but rather the ascertainment of truth. Thus, for Nyāya philosophers, truth-seeking discussion persuades interlocutors of the truth, by resolving uncertainty. The uncertainty is about a fact of the matter, and the goal is to acquire new beliefs (or defend previously-held ones) which are grounded in reasons and which are true.
This goal is especially important in the context of differences of opinion over soteriological matters, which is the focus of Jayanta's play, theĀgamad . ambara, or Much Ado About Religion.

Disagreements about Religion, Settled on the Merits?
In addition to his works of Nyāya philosophy, Jayanta Bhat . t . a wrote a four-act play which is set in Kashmir during the reign of a historical ruler, KingŚaṅkaravarman, who lived 883-902 CE, and to whom he was an adviser. 37 Jayanta's play mostly centers on a young Mīmām . saka, Saṅkars . an . a, who roams around Kashmir, engaging opponents in debate: a Buddhist (Act I), a Jain (Act II), and a Cārvāka (Act III).
The main issue in the play is how a kingdom with a plurality of different religious groups ought to draw boundaries between what is acceptable and not. While the play begins with a Mīmām . saka, it is a Naiyāyika, Dhairyarāśi, who has the final say against the backdrop of growing controversy about the tantric "black-blankets" (nīlāmbaras) and whether the Bhāgavatā scriptures should be considered authoritative. He argues in a lengthy monologue, quoting Jayanta, that "The many means taught by various scriptural approaches converge in the single summum bonum (śreyasa) as the currents of the Ganges meet in the ocean" (Dezső 2005, p. 229). Further, he argues that as long as a scripture does not go beyond certain agreed-upon moral standards-as in the case of the black-blanketseven Buddhist, Jain, orŚaivite Bhagavatās are traceable to the Vedas in some sense. 38 The play's heavily philosophical content and insistence on dealing with current events may mean its artistic value is limited to a very narrow audience. A.K. Warder (Warder 1988, p. 310) said, it is "not one of the most entertaining pieces produced by the Indian theatre", and the "last act is little but a long lecture". Yet, setting aside questions of its aesthetic merit, it is, as he said, a "historical document", and one which "shows the practice of debating" (pp. 310, 311). While the debates are surely stylized, their content parallels known texts, so that the dialogues in the play can be treated as at least somewhat representative of the kind of argumentative discourse that might have been found in 9th and 10th century Kashmir.
In his introduction to the play, Csaba Dezső has already shown how it illustrates Jayanta's understanding of Nyāya as a defender of Vedic orthodoxy, focusing on the final monologue. The next part of the paper will show how Jayanta's play illustrates his understanding of the relationship between argumentative strategies and truth-seeking in debate. Further, I apply pragma-dialectic analysis to the play, showing how well the debates fit the four-stage framework of that approach, and how we can appreciate the various speech acts deployed in debate through this analysis.

Noisy Blather in Debate
First, a remark about the title of the play:Āgamad . ambara. The term d . ambara can mean "noise, bombast, confused mass" (Macdonell 2001, p. 104), and it is what Dezső translated as "much ado" in his title Much Ado About Religion, playing on Shakespeare's Much Ado About Nothing. Jayanta used this term in the play and his NM in a way that suggests the "much ado" in the play may be restricted to portions of the debates, and not the entire play's "ado" or "controversy" over religious texts (āgama). In his comments on NS 1.2.1, he distinguished the discussion's conclusion from the conclusions of other kinds of debate, such as disputation (jalpa), saying: . . . discussion indeed results in settling one of the standpoints, although it does not terminate in flawed criticisms which are composed of blather (d . ambaraviracita) out of contempt for another person, like disputation. Thus in this way, lacking the use of intentional fallacies, it is said to be discourse without self-centeredness. 39 Earlier, in his opening remarks, he used the same term, strengthened by the prefixāto mean "a lot of" or "intense" blather. Here again, it refers to language used in a debate, specifically the wrangling and disputation sorts of debate. In his initial gloss of these two terms from NS 1.1.1, he says: But with regard to wrangling and disputation: on some occasions, these are able to be employed by someone who is dispassionate, too, in order to protect the knowledge of the truth situated in the hearts of those honest people who are greatly afraid of the boisterous blather (ād . ambara) of fraudulent criticism constructed by wicked reasoners, by greatly consoling them. 40 Thus "blather" (d . ambara) characterizes wrangling and disputation, since it is a great deal of empty noise with no logical impact. Yet Jayanta allowed that wrangling and disputation, in which fallacious moves as well as forcing points of defeat could be used in order to acquire victory, was, in some special circumstances, acceptable for people whose aims were truth-seeking and who were morally upright. This raises the question as to whether the debates in Much Ado are discussion or not, since we know in advance that our protagonist is at least a defender of the Vedas (even if he is, by Jayanta's lights, a confused Mīmām . saka).
In the play itself, this question is implicitly raised by the charge to the protagonist Saṅkars . an . a as he is about to debate his first opponent, the Buddhist monk Dharmottara. Several onlookers agree to play the role of an adjudicating audience only if the participants agree to several conditions, one of which is to avoid kathād . ambara, or, as Dezső translated it, "a noisy mass of bad disputation". 41 That they are to avoid disputation in the technical sense (as jalpa) is implicit in the attachment of kathā (discourse or debate) to d . ambara ("blather" or "noise"), since as we have seen, the kinds of debate which are characterized by noisy blather are disputation and wrangling. However, it might be asked whether the protagonist succeeds, or if this is one of those special occasions where he must stoop below the paradigmatic level of discussion in order to defend the Vedas. With this allusion to sub-par forms of debate in mind, I turn now to Act One.

Act One: Argumentation Stages of the Buddhists Defeated
The debate between Dharmottara and Saṅkars . an . a in Act One of Jayanta's play unfolds in the sequence identified in pragma-dialectic analysis. Unlike discourse in which the stages are not sequential and must be reconstructed, here, the progression is clear. 42 This act begins with Saṅkars . an . a eavesdropping on the Buddhists at their monastery, finding fault with their drinking alcohol and eating meat, despite their supposedly ascetic vows. He approaches a monk named Dharmottara-which is the name of a follower of Dharmakīrti, a famous Buddhist predating Jayanta-and introduces himself and his companion to the Buddhist monk, who is teaching a young disciple. The two trade insults, and when the Buddhist disciple restates what he has been taught (Table 3, confrontation), Saṅkars . an . a insults the idea that one could attain salvation (śreya) by rejecting the idea of the self. In response, Dharmottara accuses Saṅkars . an . a of having a morally and cognitively "muddied" mind, since he kills animals to achieve his salvation (p. 63). Saṅkars . an . a notices that some Vedic-professing people have come to watch his dispute, and engages them in a role as "arbiters" (prāśnikas).

Stage
Example Dialogue

Confrontation
Buddhist Disciple: "Why, the master has taught me the Four Noble Truths: Suffering, the Cause of Suffering, the Cessation of Suffering, and the Way". Saṅkars . an . a: "That's the teaching of the supremely compassionate one? And this is not gibberish (pralāpa) 43 where the realization of having no Self is celebrated as the path leading to salvation?" (p. 63) The arbiters agree to observe the debate, but only on certain conditions: If your talk is correct, moderate, and springs from the settled view (siddhānta); if you avoid blathering discourse (kathād . ambara) full of misinterpretations (chala), futile rejoinders (jāti), and points of defeat, if there is no envy at all in your heart, no roughness in your words, no frowning on your face; if this is a discussion (vāda) among virtuous persons, then we are always ready to serve as assessors (p. 65, adapted).
Essentially, the arbiters restrict the debaters to engage in discussion, and not stoop to the other forms of debate, wrangling and disputation. Unless the two interlocutors agree on these terms, they will not agree as to what behavior is out of bounds. 44 In addition to the requirements to keep to already-mentioned norms about speech and rationality, the arbiters introduce requirements about attitude and decorum. The debaters should lack envy (īrs . yā), words that are abusive (parus . a), and faces that have furrowed brows (bhrūvibheda) (Dezső 2005, p. 64). While the latter two are performative, the first is about the reason a person engages in debate. Envy is one of the moral flaws that, according to the NS, motivates acting, and is a kind of aversion (dves . a). Envy motivates obstructive behavior, which may be why it is singled out in this context-an envious debater may try to keep the other from coming to know the truth, even though ostensibly, the truth is what both participants desire. 45 This restriction goes well beyond anything pragma-dialectic approaches require, though, as some have argued, requirements about the character of interlocutors are consistent with the model. 46 Once the two agree to the requirements, the arbiters ask who has stated the first standpoint (Table 4). This is a neutral way to characterize the positions involved in a debate: one begins and the other responds. However, Saṅkars . an . a responds by characterizing the Buddhist's view as the prima-facie view (purvapaks . a), a term reserved for views which initially seem correct but will later be shown faulty. He restates what the Buddhist saysa requirement in Nyāya debate known as restatement (anubhās . ana)-and Dharmottara grudgingly accepts it as accurate. 47 This requirement does not neatly fall into the speech act categories within pragma-dialectics, based on Searle's taxonomy: assertives, directives, commissives, declarations, and expressives. It is a kind of meta-discourse, or a quotation of previous speech, which may be direct quotation or restating in other words (paraphrase, indirect quotation), or some combination (mixed). 48 Table 4. Opening in Act One.

Stage Example Dialogue
Opening Saṅkars . an . a: "Here am I, and here is the monk, you are prudent arbiters: what an incomparable opportunity to consider the strong and weak points of the argument!" (p. 65) [ . . . ] Arbiters: "So which one of you has put forward his thesis first (prathamapaks . avādī)?" Saṅkars . an . a: "The monk has indeed set forth the prima-facie view (purvapaks . a-) while teaching the disciple". (p. 65) The argumentation stage of the act is complex and would require an independent treatment to do it justice, as it alludes to a range of Buddhist philosophical ideas. 49 Below I analyze one portion, shown in Table 5. One major claim is presented as a counterfactualthe relationship between the momentariness of everything and the Buddhist path to nirvān . a. The momentariness of everything is supposed to be a reason, in the five-fold structure of inference, for the existence of the Buddhist path. In other words, there is an invariable relationship between momentariness and the path: if everything is momentary, then the Buddhist path is established, and if the Buddhist path is established, everything is momentary. Momentariness itself is then argued for, as well as the claim that consciousness alone, and no external objects, exist. Table 5. Argumentation in Act One.

Stage Example Dialogue
Argumentation Saṅkars . an . a: Even if the aforementioned path to liberation were to exist when the existence of momentariness is established, still, things, being examined carefully, do not experience momentariness. Dharmottara: Why? Saṅkars . an . a: Just because it lacks a reason. Dharmottara: But hasn't the reason been stated? "Because it exists". Saṅkars . an . a: The concomitance with the probandum, i.e., "momentariness", of the proof of "existence", which you Buddhists teach cannot be ascertained along a straight path, as can be in the case of smoke and fire, since no example appears to support it, and therefore it is useless. Dharmottara: If it's this way, what follows? Even understanding the pervasion relationship by exclusion is also understanding the pervasion relationship. Further, being excluded from permanent things because it lacks sequential or simultaneous efficacy, what exists "places its foot" in momentary entities, because another place is impossible.
The opening part of the argument, which focuses on the relationship between momentariness and the Buddhist path, illustrates how debate proceeds by a range of speech acts beyond assertives. Questions and indirect argumentation are strategic responses. At each point in the debate, the interlocutors have a range of choices, and they respond in ways that are constrained by the agreed-upon rules and their stated goal of evaluating the truth of the claims (not primarily rhetorical victory). Their goal is persuasive and also rational. Simultaneously, because the debate is agonistic, the moves they make are ones which they judge will resolve the debate favorably for them.
In the first set of exchanges, Saṅkars . an . a criticizes Dharmottara's argumentation by showing that the reasoning has fallacies. His reply begins with an approach explicitly described at NS 1.1.31, in which one accepts another person's (or tradition's) views as settled provisionally, for the purpose of investigating them: When, for the sake of careful examination of the details of a view, there is the suppositional acceptance of something that has not yet been examined closely, this is a position accepted on the basis of supposition. (Dasti and Phillips 2017, p. 51) Following this provisional acceptance, one then goes on to "make known the excellence of one's own thesis and to condemn the thesis of another" (p. 52).
This is precisely what Saṅkars . an . a does. One aspect of the Buddhist standpoint, as he has characterized it, is that " . . . the path to attain [the cessation of suffering] is obtained from establishing the momentariness [of everything]". 50 Saṅkars . an . a argues counterfactually, allowing that this path could exist on the basis of establishing momentariness, but then denying that momentariness exists.
Even if the aforementioned path to liberation were to exist when the existence of momentariness is established, still, things, being examined carefully, do not experience momentariness. 51 Since the relationship between momentariness and the truth of the Buddhist path is supposed to be invariable-in other terms, necessary and sufficient-demonstrating that momentariness is false undermines the truth of the Buddhist path.
Dharmottara's response is to elicit more argumentation, simply asking, "why?" (kutah . ). It is a directive-a speech act that "involve(s) an attempt to make the addressee do what is expressed in the propositions they relate to" (p. 39). At this stage in the debate, he has little room to maneuver, since Saṅkars . an . a has chosen to begin by temporarily accepting his starting point, and then merely denying one claim, without any supporting reason for that denial.
The subsequent reply and counter-reply exchange is curt: Saṅkars . an . a: Just because it lacks a reason.
Dharmottara: But hasn't the reason been stated? "Because it exists". 52 Here, in response to Saṅkars . an . a's assertive, which charges Dharmottara's inference with lacking a reason, the Buddhist replies with a discourse particle that can be read rhetorically: nanu. He could have responded with a straightforward assertion that denies the charge, like, "No. The reason has already been stated" (na. purvokto hetuh . ). The choice of a rhetorical question could be a strategic choice, as such questions state things so that an affirmative reply is tacitly expected against what has already occurred in the debate. 53 We notice that there are choices available to debaters beyond mere assertion which are consistent with Nyāya rational requirements. Assertions or rhetorical questions both carry the same content, though with different illocutionary force.
In reply, Saṅkars . an . a first implicitly charges Dharmottara with a fallacy in the use of momentariness as a reason, which is that it lacks any supporting example (dr . s . t .ā nta). 54 This fallacy, known as deviation (savyabicāra), is particularly a problem for universal claims, since it is a failure to present a supporting example which is outside the scope of the target under investigation. Ordinarily, when arguing for a thesis that asserts some property exists in a target location-paradigmatically, fieriness characterizes a distant mountain-an agent puts forward a reason ("The mountain is smoky") and also a supporting positive example ("campfires are place where smoke and fire are found together") that supports the invariable connection between smoke and fire ("where there is smoke, there is fire"). If one put forward the mountain itself as support for this rule, it would be question-begging. However, when the target is everything that exists, then no independent positive example can be given. 55 In speech act terms, however, Saṅkars . an . a does not charge Dharmottara with a fallacy openly, through an assertive which lays claim to a particular fallacy by name. Rather, he simply states that the connection between the thesis and reason cannot be understood "along a straight path" (naśakyam r . junā mārgena) because it lacks an example. 56 Keeping in mind the strategic purposes of his replies, his choice of language may be to subtly poke at the Buddhist "path" (mārga) as not being "straight" in the sense of honest or moral, another common connotation of r . ju. He is not only concerned with refuting Dharmottara's arguments, but with swaying the arbiters, and by slyly suggesting that Buddhists are immoral, he appeals to their shared Vedic commitments. Since the act is bookended-as I will show-with criticisms about the honesty of the Buddhist religious path, this reading is in line with Saṅkars . an . a's stated concerns, which are not just that the Buddhists are mistaken, but that they are charlatans.
Dharmottara's response is to ask, "If it's this way, what follows?" (yady evam . tatah . kim). While not an assertive speech act, the question implies one, a rejection of the purported fallacy. He is essentially asserting that the structure of his reasoning is acceptable, even while agreeing with Saṅkars . an . a's characterization of it. This is clear in his following sentence, which clarifies why the structure should be allowed: "Even understanding the pervasion relationship by exclusion (vyatireka) is also understanding the pervasion relationship". 57 Both concomitance (anvaya) and exclusion (vyatireka) in examples are legitimate means of understanding the pervasion relationship. In addition to stating that the campfire is an instance where smoke and fire are found (a positive example of concomitance), one could put forward a negative example of exclusion, of a lake, where neither are found. After defending his original formulation, Dharmottara gives an additional argument to support his claim of momentariness.
Further, being excluded from permanent things because it lacks sequential or simultaneous efficacy, what exists "places its foot" in momentary entities, because another place is impossible. 58 Without getting too deep into the technical details of this dispute, the crucial move Dharmottara makes is to bolster his claim that the negative formulation is necessary because there are absolutely no instances of something that exists being non-momentary. The idea, roughly, is that all effects must either occur after their cause (sequentially) or at the same time. However, something which exists in a non-momentary or permanent condition (nityatva), cannot bring about effects in a sequence, since this would be to produce a change in the supposedly permanent object (before it brings about the effect E, it is not effective of E, but afterwards, it is). Neither can it bring about the effect simultaneously, because that would be to remove cause and effect of all intelligibility (if watering a plant is the cause of its growth, then the plant cannot grow at the very same immediate instant as the water touches the plant-this would essentially make its growth uncaused). 59 In terms of Nyāya debate, Dharmottara is giving support (sādhana) for his position, which is one of the requirements in discussion seen above ("supporting and criticizing is by epistemic instruments and suppositional reasoning"). Eventually, Saṅkars . an . a explicitly presents some competing positive claims, such as that permanent things are efficacious, along with explicit denials of what Dharmottara has claimed, such as that momentary things are not efficacious (Dezső 2005, p. 73). However, he does not argue in an extended manner for a competing counter-standpoint not-p, put explicitly in contrast to Dharmottara's p. (In fact, Dharmottara argues for a conjunction of claims, p and q and etc.) In pragma-dialectic terms, this is a multiple non-mixed critical discussion. This means that in Nyāya terms, it is a case of wrangling (vitan . d .ā ).
Finally, in the closing stage, Saṅkars . an . a asks the arbiters to decide which position is better (Table 6). This is preceded by three different stage directions that indicate the monk sits silently, rather than replying to his interlocutor. The arbiters have already stated, in a declarative speech act, that Saṅkars . an . a's reasoning has convinced them, and that, with it, "the position of momentariness has been refuted". 60 In perhaps a sly reference to his earlier implicit criticism of Dharmottara, they characterize Saṅkars . an . a's speech as nītimārga. Dezső translates this as "course of argumentation" (p. 81) but it perhaps has the additional connotation of an "upright (nīti) path (mārga)". They urge him to move on to another topic, in a directive speech act, telling him in an imperative form (abhidhatsva) to "speak" on the topic of consciousness (p. 78). Table 6. Closing in Act One.

Stage Example Dialogue
Closing Stage directions: The monk draws on the ground in silence, with eyes downcast. Saṅkars . an . a: Honorable arbiters, tell us which one of the two positions is superior? Arbiters: Why are you asking us? Your position is supported by the monk himself by his entering into silence. 61 These speech acts, declaratives from the arbiters about being partially convinced and directives urging debaters to speak or to stop speaking, were not explicitly discussed in the Nyāyasūtra. However, Jayanta presented them as an ordinary part of a debate. In the play, they are informed by the intersubjectively available procedural rules set out in the opening-what pragma-dialectics would consider "conventionally valid" rules. The arbiters rarely intervene in the debate, only in the closing when requested by Saṅkars . an . a to fulfill their agreed-upon institutional role, and one earlier time, which I discuss below.
Once Saṅkars . an . a has set out his argumentation on consciousness, Dharmottara once again fails to respond-an action suggesting one of the twenty-two points of defeat known as "lack of an idea" or apratibhā, in which the interlocutor is not able to give a reply (Todeschini 2010, p. 64). However, rather than appealing to a point of defeat as judgment against Dharmottara, the arbiters interpret his silence as a kind of speech act. Through entering into a state of silence, which is presented as an action, rather than a failure to speak, the monk implies he agrees. At least, this is the arbiters' interpretation.
With this, the debate concludes, and Saṅkars . an . a urges the monk to take the results of their discourse seriously: he should either abandon Buddhism, as it has been established that its path to salvation is no such thing, or he should openly accept that he is engaged in "snobbery and masses of blather (kaurukucīkūrcad . ambara-)". 62 At the close of the act, we again see the titular Sanskrit term d . ambara (blather). Here it is reserved for the monk's teachings, emphasizing that they have finally been shown to be flawed.
From the Nyāya perspective previously discussed, at the conclusion of a debate, once a standpoint has been supported and opposing views against it successfully refuted, then it is justified. On the assumption that the inferences involved were legitimate, and not merely apparent inferences, the standpoint is true, and the debate has enabled both participants to become aware of its truth.
Unlike Nyāya debate theory, pragma-dialectics, as discussed earlier, does not focus on truth explicitly, but rather acceptability. Yet in rule eight (above), we also see that only when appropriate argument schemes are correctly used can the standpoint be considered "concisively defended". Similarly, rule nine governs the reasonable epistemic attitudes a person may have after a critical discussion concludes: if a defense has been conclusive, the standpoint cannot continue to be doubted, and if a defense has been inconclusive, that standpoint cannot be maintained. Thus, in pragma-dialectic terms as well, Dharmottara, with his inconclusive defense, ought to abandon his standpoint.

Conclusions
While Saṅkars . an . a seems to be an ardent defender of the Vedas and thus would be a candidate for a truth-seeking debate practitioner, from what I have argued, he is more a vaitan . d . ika, a "wrangler", and perhaps even motivated more by a concern with victory than defense of the truth. Before his opening confrontation with the Buddhist Dharmottara, the Mimamsaka introduces himself by stating that his goal is to "humiliate the enemies of the Veda", and if he is not able to fulfill this goal, his previous studies will be all for naught (Dezső 2005, p. 53). Here, we might think Saṅkars . an . a is less concerned with the truth than he is in showing himself to be intelligent, in contrast to the Buddhists with their bad reasoning (kutarka). Further supporting this interpretation, when he encounters Dharmottara, as we have seen, his confrontation is critical, but it does not carry an explicit counter-standpoint. In pragma-dialectical terms, the argumentation is "non-mixed", or one party claiming p and the other simply doubting or criticizing p.
Additionally, he violates the norms he has agreed to in the opening stage, presenting arguments while sneering, and calling his opponent an "fool" (mud . ha). 63 The latter violates the requirement of harsh words given earlier, and while a smile may be the opposite of furrowed brows, it suggests an envious heart, gleeful at his opponent's inabilities. After this exchange, Dharmottara's watching disciple and Saṅkars . an . a's young brahmin friend nearly come to blows, and the arbiters, along with Dharmottara and Saṅkars . an . a must stop the fight.
Further, as Saṅkars . an . a picks up his argumentative thread after this near fistfight, the arbiters ask him to stop talking, saying he has done "enough with his verbosity" (kr . tam . vistaren . a). Here he may have fallen afoul of the point of defeats above known as "excess", in which a person adds additional examples or a reason statement. At the very least, from a pragmatic standpoint, he has gone beyond cooperative engagement with Dharmottara, which requires him to ensure his contributions are all and only those which resolve the standpoint on the merits. By swamping his opponent with excess reasons, he is being uncooperative.
From a standpoint of the virtues of discussants, he is concerned now with his performance, not with the persuasiveness of his argument for Dharmottara. Vācaspati Miśra makes this point about "excess" as a point of defeat. After all, we might think that more reasons are better, and if these are genuine reasons, they contribute to resolution of the issue under discussion. Certainly, they do not obstruct the resolution in the way that logical fallacies might. However, Vācaspati says: But the (point of defeat) "excessive", even if there is no directly obstructing understanding of the truth, even still, as a result of its use, the other one who hears, whose mind is cluttered, is not able to understand the truth. Therefore, even "the excessive" is a cause of an obstruction to understanding the truth. 64 In Jayanta's presentation, then, although Dharmottara is defeated in debate by a Mī mām . saka, there are doubts about the extent to which his debate was a genuine discussiontruth-seeking and rationally persuasive-and to what extent it was a kind of wranglingfocusing on brow-beating his opponent into silence with an excess of argumentation and sly language use. However, whether Saṅkars . an . a manages to engage in discussion by Nyāya's lights, the play sets up the norms for its possibility in the opening charge, norms which are more extensively discussed in Jayanta's other works. The play shows us the range of strategies which interlocutors can deploy in the course of a debate, which opens up choices at different stages for responses. These strategies include more than the formal structure of the inferences being deployed, but also the pragmatics of the speech acts in which inferences, suppositional reasoning, and so on, are communicated. Consequently, I have shown two main principles guiding Nyāya debate theory: 1.
Truth-seeking discussion (debate) persuades interlocutors of the truth by resolving uncertainty; 2.
Discussion's persuasive strategies are context-sensitive, thereby subject to pragmatic analysis, but also rationally constrained, thereby subject to epistemic norms.
For early Nyāya debate theorists, especially Jayanta, the paradigmatic kind of debate, discussion, resolves differences of opinions on the merits, but in a way which is truthseeking. They are attentive to the pragmatic features of debate (whether discussion or other kinds) in their theorizing, too, showing that they do not think rhetorical persuasiveness interferes with epistemic aims. Jayanta's play illustrates that not only external performance but virtuous dispositions are important for achieving these high standards, and also shows that the boundary between discussion and wrangling is difficult to maintain.
There are two other debates in the play, and a final monologue by a Nyāya philosopher, Dhairyarāśi, who sets out the boundaries of religious toleration in the kingdom-these depend on what is accepted by people. According to Arnold (forthcoming), whether this acceptance comes from great people or a great many people (the compound can be read in two ways), acceptability need not entail rational acceptability (pp. 12-13). Arnold's solution is to appeal to Bhat . t . a Mīmām . sā epistemology, distinguishing between subjectively-held justification and objectively-grounded truth. Whether Mīmām . sā (or Nyāya) epistemology (and which varieties therein) are successful models for addressing problems of religious toleration is beyond this paper's scope. From a literary and philosophical standpoint, there are remaining questions as to why Jayanta does not present a Naiyāyika as a victorious discussant but rather as an almost Hobbesian defender of politically expedient boundaries, set out in a monologue in the final act. Further analysis of the structure of the other acts of the play-and Act One in its entirety-using both pragma-dialectics and Jayanta's understanding of debate may prove fruitful in answering these questions.
What this paper has tried to show is that Nyāya debate is concerned with both rational acceptability and persuasion, at the same time as it is aware that argumentation occurs in particular contexts between individuals who have competing goals and yet agree to cooperate in a limited manner. We have seen that concern with justification and truth are found together in Nyāya in debate theory, which allows for agonistic confrontation within rational constraints. Pragma-dialectic analysis may help us further appreciate the pragmatic underpinnings of their concern with the performative elements of Nyāya debate, but without leaving behind its ever-present epistemic aims. Conversely, as pragma-dialectics focuses on acceptability and, arguably, a deficient sense of "justification", Nyāya approaches can bolster its stated concern with rationality. Thus, while engaging in dispassionate discourse may be conceptually compatible with a limited kind of agonism between virtuous persons, when constrained by a mutual truth-seeking project, Jayanta's play shows the difficulty of maintaining such an adversarial position in actual argumentative contexts.
Funding: This research received no external funding.
Acknowledgments: Thanks to Matthew Dasti for helpful conversations about Vātsyāyana's Nyāyabhās . ya, Brendan Gillon for reading a draft and suggesting revisions, the participants of the First International Workshop on Oriental Logic who gave feedback on a draft of this paper, especially Purushottama Billimoria, Amita Chatterjee, Ali Sadegh Daghighi, and Anil Mudra, and two anonymous reviewers for helpful suggestions which improved this paper.

Conflicts of Interest:
The author declares no conflict of interest.

Abbreviations
NBh Thakur (1997) NMŚukla (1936 in NVT NVT Thakur (1996) Notes 1 For discussion of dates, see (Hacker 1951) and (Oberhammer 1962). I omit those Naiyāyikas who follow Bhasarvajña's Nyāyasāra. Beginning with Udayana, we might say, as Thakur (1997, p. xii) does, that a new period in Nyāya philosophy began. Given the importance of the earlier Vātsyāyana (c. 450-500 CE) and , and the fact that Jayanta Bhat . t . a's work presupposes them, their views will also be included. However, within the space of this article, I cannot explore the full ambit of all three philosophers and their full corpora cannot be explored within the space of this article. Further, the question of how these thinkers developed their positions in relation to the originary sūtra text is also beyond its scope. 2 Translation adapted from (Dasti and Phillips 2017, p. 9). Unless otherwise noted, translations are by Malcolm Keating. Footnotes have been referred to in (G. Jha 1984) for some portions of Vācaspati's text and to (Bhattacharyya 1978) for Jayanta's. 3 As is well-worn territory by now, jñāna is sometimes translatable as "knowledge", a success term; although other times, it simply means an agent's contentful experience, whether or not it is veridical. The latter is the use in NS 1.1.2, and so "false knowledge" would be inapt for mithyajñāna. See discussion in (Ganeri 2018 Todeschini (2010, p. 50, n. 3) criticizes Matilal's attribution of intentions to the sūtra text's author(s) and compiler(s) as "historically problematic". 5 Nicholson does not discuss Jayanta Bhat . t . a, so I do not know if he would characterize him in this same manner as other commentators. However, since he draws on the same source material and largely aligns himself with the other commentators, especially Vācaspati Miśra on the importance of "dispassionate debate", it is plausible that he would fall within the purview of Nicholson's hypothesis. 6 He mentions discussion in explaining suppositional reasoning (tarka) and implies it in relationship to certainty (nirn . aya). 7 vimr .ś ya paks . apratipaks .ā bhyām arthāvadhāran . am . nirn . ayah . vādah . (Thakur 1996, p. 265, line 11). 8 That certainty cannot be merely psychological is suggested by Vātsyāyana's rejection of a prima facie view at NS 1.1.41 (G. Jha 1984, vol. 1, pp. 459-60). The suggestion is that both proponent and opponent are already certain of their positions, otherwise they would not engage in debate, and so it makes no sense to say that certainty is ascertained by discussion. Further, there is no need for both lines of reasoning (support and criticism), since it is only the one which supports the finally established position that is relevant. Vātsyāyana's reply clarifies that discussion is not necessary for certainty (it can be achieved other ways) and that certainty is the end result of not only supporting p but rejecting not-p. The idea seems to be that while both participants are initially certain, once they enter into discussion, their certainty should rationally be undermined because they have an opposing view. This is why Uddyotakara says that doubt comes about by controversy, and from the statement of the opponent's view (NV 1.1.23). Thanks to anonymous reviewers for suggesting that I develop the idea of certainty further. 9 . . . anyatarādhikaran . a-nirn . ayam antaren . a na paryavasyati nyāyoparamakāran . atvena tasya pravarttako nirn . aya itarathā niravasānam anāsāditaphalam . ko nāma nyāyamārabheta. (NMāhnika 1, ad NS 1.1.1, vol. 1, p. 10, lines 4-5). " . . . Suppositional reasoning (tarka) does not stop without certainty (nirn . aya) about the authority of one of the two positions, since certainty's function is causing the completion of reasoning (nyāya). On the other hand, without a limit, its result remaining unaccomplished, what person would undertake reasoning?" and vāde tu vicāryamān . e nyāyah . sam .ś ayacchedanena adhyavasitāvabodham adhyavasitābhyanujñātam . ca vidadhat tattvapariśuddhimād dhātīti vītarāgaih .ś is . yasabrahmacāribhih . saha vādah . prayoktavyah . . (NMāhnika 1, ad NS 1.1.3, vol. 1, p. 10, lines 4-5). "Now, with regard to (the category of) discussion (vāda), which is deliberation: reasoning (nyāya), by removing doubt, bestows a settled understanding and a settled agreement from considering the completely correct truth. Thus one ought to engage in discussion with dispassionate students and peers". NB: I correctŚukla tattvapariśuddhimādādhātīti to tattvapariśuddhimād dhātīti in light of SARIT's transliteration and the context (SARIT 2014). 10 An anonymous reviewer raises an important related question, too large to address in this context: what is the relationship between Nyāya epistemology, debate, and what is today called "naturalized epistemology" (NE)? "Naturalized epistemology" is an umbrella term, but typically NE holds to a rejection of the a priori/aposteriori distinction, some close relation between natural properties and epistemic properties (whether reduction or supervenience), and the incorporating of psychology, and other empirical sciences into the study of knowledge. If Nyāya epistemology is broadly externalist, and explains epistemic virtues in terms of causal processes, then is it a kind of NE? Amita Chatterjee (2017) argues that Nyāya thought is a kind of moderate NE. If it is, where does that leave rational norms? Are they explained only in terms of such processes, which might be unavailable to a reasoner's awareness? Work by externalists such as Goldman (1994Goldman ( , 2004 focuses on giving explanations for argumentation's rational norms in such a framework. Goldman argues that arguments (as a series of propositions) lack epistemic properties, but rather episodes of thought bear them, and that inferential justification is a matter of "how the target belief is arrived at (or at least causally sustained)" and not merely logical relationships among abstract propositions (Prets 2003, p. 57). If, and how, this approach coheres with Nyāya epistemology, and its implications for argumentation is an area worth further inquiry. 11 pramān . atarkasādhanopālambhah . siddhāntāviruddhah . pañcāvayavopapannah . paks . apratipaks . aparigraho vādah . (Thakur 1996, p. 270, lines 3-4 To an interlocutor inquiring when certainty arises in a discussion, Uddyotakara explains that there is no fixed point, but it depends on whether the supports and criticisms are successful at each stage of the exchange (G. Jha 1984, p.1, line 467 " . . . Even in discussion, there is certainty which is devoid of reflection, when two people who are both certain engage in discussion, because in that case, there is no reference to something which is doubted. 'How then is this conversation (between certain interlocutors) an interruption of doubt, the result of discussion?' First, even for both of those who are certain, while the discussion is going on, doubt suddenly appears because of the strength belonging to the two arguments being deployed. Since discussion terminates in certainty about a topic, it is said that its result is the interruption of doubt". 15 These five components are: thesis (pratijñā), reason (hetu), illustration (udāharan . a), application (upanayana) and conclusion (nigamana). The stock example is an inference to the existence of fire on a mountain: There is fire on the mountain (pratijñā), because there is smoke on the mountain (hetu), and where there is smoke there is fire, just as in the kitchen and unlike the lake (udāharan . a), and this case is like the kitchen and unlike the lake (udāharan . a), and therefore there is fire on the mountain (nigamana). 16 "And by this ['supporting and criticizing is by ways of knowing and suppositional reasoning'], the bringing about of insufficiency and excess is understood, as in 'There is insuffuciency (nyūna) by what is missing even one of the component parts', (NS 5.2.12); 'There is excess by what has an excess reason or example' (NS 5.2.13)". anena ca hīnam anyatamenāpy avayavena nyūnam. hetūdāharan .ā dhikam adhikam iti nyūnādhikayor udbhāvanam anujñayate, NMāhnika 11, ad NS 1.2. 1, vol. 2, p. 150, lines 26-27. 17 For more detail, see (Phillips 2017 "This is inapt: 'By the mention of 'which does not contradict settled claims (siddhānta)', there is 'Inconsistency (apasiddhānta), the continuation of discourse without qualification, once having accepted the settled claims' (NS 5.2.23), thus in debate the points of defeat called 'in opposition to what is settled' are to be pointed out". siddhāntāviruddhagrahan . ena siddhāntam abhyupetya aniyamāt kathāprasaṅgo 'pasiddhānta ity apasiddhāntākhyanigrahasthānāni vāde udbhāvyanta iti tad anupapannam . . NM,āhnika 11, ad NS 1.2.1, vol. 2, p. 151, lines 1-8. 22 bhrāntyā tu katham . cit prayuktānām avaśyam udbhāvanam anudbhāvane vastupariśuddher abhāvāt. NM,āhnika 11, ad NS 1.2.1, vol. 2, p. 151, lines 1-8. 23 "Remembering and understanding either the support or the criticism-on these points a speaker does not acquire victory, but rather, things such as abandoning a thesis, are the whole cause for genuine victory over another-this is accepted by these two words here (siddhāntāviruddha)".āgatyānusmr . tya ca sādhanam upālambham . vā tatra vadan na parājīyate, pratijñāhānyādi tu samyakparājayakāran . am akhilamābhyām . padābhyām abhyanujātam atreti. NM,āhnika 11, ad NS 1.2.1, vol. 2, p. 151, lines 9-13. See discussion also in Solomon (1976, pp. 106-7 (Prets 2001) for discussion of the history of jātis in early Nyāya. He argues that they are "no way futile or unsound, but question the validity of proof in the early dialectic tradition" (p. 553). 27 For an overview of the history and major components of the pragma-dialectical approach, see (van Eemeren 2015). 28 Definitions based on discussion of these concepts especially on page 30 of van Eemeren (2018). 29 The list below is taken from Chapter 4, section 3 of (van Eemeren 2018). The following characterization of pragma-dialectics depends on this work. 30 One starting point is the categorization of fallacies in van Eemeren (2018, pp. 66-67). This could be coupled with analyses such as Todeschini (2010 tam asāvabhyupagacchaty eva na tatra sādanam upadiśati parapaks . am evāks . ipannāste. NM II, 151.9-10. "With regard to that thing to which the [destructive debater] agrees, there is no statement of support, and the counter-standpoint itself is implied". 34 This changes after Udayana, as both Solomon (1976) and Nicholson (2010) Solomon (1976, p. 117) notes that in his Nyāyapariśuddhi, the 14th century Veṅkat . anātha refers to, but does not endorse, the distinction between a wrangler aiming at victory and one who argues dispassionately against a position without having one of their own. Veṅkat . anātha rejects this distinction because he thinks that someone who is genuinely dispassionate (vītarāga) would not be "satisfied" with mere criticism and would instead want to reach "conviction regarding the true nature of the thing" under discussion (p. 117). While not going so far as to endorse a necessary connection between the structure of wrangling and the entrance of fallacious reasoning, we may be able to see a further philosophical motivation for the general Nyāya tendency to preserve this connection, in pragma-dialectic terms. The Unexpressed Premise Rule states that participants should not "disown responsibility for their own unexpressed premises". Since a wrangler criticizing p is implicitly committed to not-p, we might wonder about their commitment to participation in the norms of discussion, in which one sets out beliefs openly for investigation. Someone who seeks to conceal their commitment to not-p by not only failing to state it, but failing to argue for it, has demonstrated a non-cooperative attitude. Thus we might think they are more likely to resist cooperation in other ways, such as availing themselves of tricky fallacies. Thanks to an anonymous referee for pressing on the point of what kind of connection there is between implicit positions and irrationality. 37 The following discussion of the play and Jayanta's life draws on material in ) and a longer introduction, (Dezső 2004). 38 For how this argument coheres with those in the NM, see (Dezső 2004, pp. xxi-xxiii). 39 . . . eva cānytarapaks . anirn . ayāvasāna eva bhavati vādah . , na jalpavadalīkadūs . an . ad . ambaraviracitaparaparibhavaparyavasāno 'pīti tathā buddhipūrvamābhāsānām aprayoga iti nirmatsarakathātvam asyoktam. (NMāhnika 11,vol. 2,p. 151,. The larger context of the quote is about how to understand the reference to suppositional reasoning and epistemic instruments in the sūtra.   (Dezső 2005, pp. 64-65.) Note that in the edition, Sanskrit is always printed on even page numbers and the English translation on odd. 42 Translations and page numbers are  unless otherwise noted. 43 Dezső translates pralāpa as "raving", but I use "gibberish" as it connotes nonsense and lacks the emotive connotations of "raving". 44 In his play, Jayanta omits controversy over what constitute points of defeat and in what contexts, a controversy which we find in Buddhist texts especially, such as the Vādanyāya of Dharmakīrti. Thus we might imagine that, in reality, a Buddhist might not be in agreement about the rules of debate-both their problem and conventional validity in pragma-dialectic terms. 45 For further discussion, see (Das 2020). He translates Uddyotakara's definition of envy (p. 817): "Envy is the desire to thwart the adherence (abhiniveśa) of others to a common object; the desire to thwart the adherence of others to that which is common, i.e., unacquired by anyone, is envy". In contrast, I would take Uddyotakara as talking about the possession of (abhiniveśa), rather than "adherence to", a common object. In the context of discussion (vāda), that "object" is the truth, and, consistent with Jayanta's idea that ignorance is the ground of both envy and desire, we might hypothesize that the mistaken conception of truth as something one can possess motivates such envy. See (V. N. Jha 2018, p. 39) for Jayanta's remarks.