Riding the Yoked Necks of the Lions of the Middle Way and Epistemology

Sat., Aug. 23rd, 09:00-15:30

Conventional Truth and Epistemology for Jñānagarbha − The Relationship between Conventional Truth and the Means of Valid Cognition

Akahane, Ritsu (Institute for the Cultural and Intellectual History of Asia, Vienna, AUT)

Jñānagarbha’s (ca. 700) Satyadvayavibhaṅgavṛtti, which deals with the Two Truths Theory, the most important concept for the Madhyamaka school, had a strong impact on later Indian Mādhyamikas such as Śāntarakṣita and Kamalaśīla, who assumed a key role in transmitting Indian Buddhism into Tibet. It is well known that Jñānagarbha’s exegesis on “conventional truth” is influenced by the epistemology of Dharmakīrti (ca. 600−660). In the final analysis, Jñānagarbha defines conventional truth as “something just like appearance” (yathābhāsa/yathādarśana), which is the object of direct perception (pratyakṣa) and produces effective action (arthakriyā). At the same time, Jñānagarbha claims that one should not examine conventional truth through inference (anumāna) or reasoning (nyāya/yukti).

Nevertheless, Jñānagarbha uses inference in his text, and thus he seems to examine conventional things by means of inference. How should we come to terms with this contradiction? Or rather, by which means of valid cognition (pramāṇa) can we grasp “conventional truth”? Concerning such questions, Jñānagarbha claims as follows: “My explanation of the conventional truth is not a contradiction. If the two means of valid cognition are contradictions then inference is not correct. Therefore the conventional truth is that which can be grasped through direct perception, but need not be examined through inference. Inference should be used in order to merely deny ‘real things,’ which opponents insist on.”

My paper will investigate passages of Dharmakīrti’s Pramāṇavārttikasvavṛtti (PVSV: comments on verse 20ab) and Pramāṇaviniścaya (PVin: second chapter) on which Jñānagarbha bases his discussion of this topic. These passages, which have not so far been brought to the fore in modern research, focus on the relationship between the two means of valid cognition. This investigation will give us a clue to the process of integration of Madhyamaka thought and Buddhist logic in the early eighth century. Moreover, my paper intends to explore the question of whether we should describe Jñānagarbha as a Svātantrika when we examine Jñānagarbha’s particular usage of inference. Finally, I would like to present my views on the question of Jñānagarbha’s affiliation to a specific Buddhist school.


A Legacy of “Pure” Madhyamaka: Atiśa and Early Kadampa’s on the Relations/Non-Relations between the Middle Way and Epistemology

Apple, James (University of Calgary, Calgary, CAN)

This paper examines the tensions between Buddhist epistemology and Madhyamaka-oriented views of reality as expressed through Atiśa’s (982–1054 C.E.) “Entry to the Two Realities” (Satyadvayāvatāra) and recently identified early Kadampa (bka’ gdams pa) commentaries on the two realities (satyadvaya). An issue for a number of Indian Buddhist thinkers was the question of the role that the means of valid knowledge (pramāṇa) had in Buddhist soteriology. This paper argues that Atiśa and his Kadampa commentators upheld a traditional pre-sixth century Buddhist separation of ‘reasoning’ (yukti), an ‘internal’ Buddhist form of critical analysis, from hetu-vidyā, the ‘external’ epistemological devices used to defend Buddhist Dharma and defeat non-Buddhist opponents. The tension between Buddhist epistemology and Madhyamaka for these authors was mitigated through employing epistemological devices to refute opponents but not for realizing ultimate reality. Atiśa and his followers emphasize that direct realization of ultimate reality comes from non-conceptual meditative cultivation and not through means of valid knowledge (pramāṇa, tshad ma). As these authors never directly cite or reference either Dignāga or Dharmakīrti regarding reasoning procedures, the paper concludes that the processes of yukti followed by these commentaries represents a “pure” Madhyamakan lineage tradition derived from the Prajñāpāramitā and the works of Nāgārjuna to realize emptiness.


Self-Awareness and the Integration of Pramāṇa and Madhyamaka

Duckworth, Douglas (Temple University, Philadelphia, PA, USA)

Buddhist theories of mind pivot around two distinct interpretative strands: (1) an epistemological tradition in which the mind, or the mental, is the foundation for valid knowledge and (2) a tradition of deconstruction, in which there is no privileged vantage point for truth claims. The contested status of these two strands is evident in the debates surrounding the relationship between epistemology (pramāṇa) and madhyamaka that extend from India to Tibet. This paper discusses this relationship between these two theoretical approaches and their implications for a Buddhist theory of mind, with reference in particular to the works of Śākya Chokden (shākya mchog ldan, 1428-1507) and Tsongkhapa (tsong kha pa blo bzang grags pa, 1357-1419).

Śākya Chokden and Tsongkhapa both claim to integrate madhyamaka and pramāṇa, but the ways they do so are quite distinct. Their theories reveal the distinctive foundations of pramāṇa and madhyamaka upon which their systems are built. Following Dharmakīrti, Śākya Chokden represents the Buddhist tradition of epistemology, and affirms self-awareness as the only valid cognition. In contrast, Tsongkhapa follows Candrakīrti in denying self-awareness. He does not partake in phenomenology; rather, his pursuit is one of critical ontology.

Tsongkhapa attempts to integrate pramāṇa and madhyamaka by introducing a third alternative to the dichotomy of valid and invalid – an unreflective stance. Śākya Chokden tries to integrate pramāṇa and madhyamaka by delineating the respective claims of the traditions into different registers of truth, and by inserting a third category of truth, ultimate self-awareness, beyond the duality of subject and object. The role that this third category of ultimate self-awareness plays for Śākya Chokden is paralleled in Tsongkhapa's third category, where something is neither correct nor incorrect. In both cases, this category of truth becomes an incontrovertible “foundation.” While Tsongkhapa evokes an unreflective stance of commonsense to avoid unwanted foundational assumptions of epistemology, he reinserts the conventional as foundational in the same breath.

In this paper, I will show how a third category functions in both Śākya Chokden's and Tsongkhapa's integration of pramāṇa and madhyamaka as not only a foundation, but the end of their epistemology. For Tsongkhapa, commonsense is the starting point and end of his philosophy, as it is the nature of the conventional (and by implication, ultimate) truth. For Śākya Chokden, self-awareness is the cornerstone of his philosophy: it is the starting point of epistemology in that it is the transcendental condition for experience. It is the end, too, because for him ultimate self-awareness (a.k.a. gnosis) is the only thing that is real in the end.


Dignāga, Bhāviveka and Xuanzang “On the Restriction of the Thesis” (*pratijñāviśeṣaṇa)

HE, Huanhuan (Institute of Philosophy, Chinese Academy of Social Sciences, Beijing, CHN)

It is well known that the terms tattvatas and paramārthatas as epistemological and ontological limiters or restrictive clauses are one of the remarkable characteristics of Bhāviveka’s (ca. 490-570) thought. Much work has been published on these terms from the perspective of the two-truth/two-reality theory of the Madhyamaka traditions. In Bhāviveka’s Jewel in the Hand (Da sheng zhangzhen lun, *Mahāyāna-hastaratna-śāstra, hereafter Zhangzhenlun ), the expression “in reality / in truth” (*tattvatas) – Chinese zhenxing, can reflect both tattva and tattvatas – is used to restrict the thesis (*pratijñā) of an argument, jianbie lizong (*pratijñāviśeṣaṇa).

What is of further interest is that in a famous debate held in the city Kānyakubja in 642 CE under the auspices of King Śīlāditya, Xuanzang (600/2-664) not only used the term “in reality / in truth” (zhengu, *tattvatas – the Chinese affix gu in zhengu corresponds to the ablative case-ending –) to restrict the thesis or proposition. He also used the term “agreed upon” (jicheng, *prasiddha) to restrict the dharmin, as well as “I accept” (zixu, *svamata) to restrict the argument’s reason (hetu) or justifier.

In this paper, I will begin my discussion with an analysis of the way in which Xuanzang and his main disciples understood and made use of the technical term zhengu (*tattvatas) to restrict the range of the content of the thesis or proposition. I then argue that the inference Xuanzang seems to have used in the course of a debate in which he was engaged while he was in India appears to have been deeply influenced by Bhāviveka. I will not discuss here the ways in which Sthiramati, Dharmapāla and Candrakīrti criticized this notion, but I will note its possible relationship with Dignāga’s (ca. 480-540) theory of the logical argument.


Reframing Dharmakīrti. On Some Aspects of Phya pa’s Blending of Epistemology and Madhyamaka

Hugon, Pascale (Austrian Academy of Sciences, Vienna, AUT)

While the Madhyamaka works of the Tibetan scholar Phya pa Chos kyi seng ge (1109–1169) display a pervasive integration of the tools from his epistemological system, his epistemological works conversely encompass the discussion of typically Madhyamaka philosophy-related topics such as the proof of emptiness. The combination of the two fields is not superficially a matter of terminology, argumentative technique or shared topics of discussion, but extends to a more significant blending. As I will discuss in my paper, Phya pa is attempting to craft an epistemological system liable to range over the usual respective frames of reference of Indian epistemology (pramāṇavāda) and Madhyamaka, which notably allows valid means of cognition to be applied to the realization of the ultimate. This endeavor leads him to carry out substantial modifications of Dharmakīrti’s system, a significant one undertaken in the very definition of valid cognition. While Phya pa’s predecessor rNgog Blo ldan shes rab (1059–1109) was reportedly reading Dharmakīrti as a Madhyamaka, Phya pa appears to be deliberately adapting Dharmakīrti’s system to his own philosophical project.

Building upon a previous discussion of Phya pa’s definition of valid cognition and the circumscription of its object (P. Hugon, “Phya pa Chos kyi seng geʼs Views on Perception,” in Religion and Logic in Buddhist Philosophical Analysis, Vienna, 2011, pp. 159‒176), my paper will examine two further issues: (i) Phya pa’s reframing of Dharmakīrti’s distinction between ultimate and conventional reality as two subtypes of conventional reality to which a redefined ultimate reality is appended as well as the valid cognitions associated with both, and (ii) his discussion of the various philosophical positions relative to the nature and mode of cognition of objects. Regarding the latter, I will examine in particular whether the direct realist position supported by Phya pa in this context can be considered Phya pa’s final position in the overall framework of his system (hence a position within Madhyamaka) or whether it should, like Dharmakīrti’s adoption of representationalism that is forsaken at some point for idealism, be apprehended as an intermediate step in an ascending scale of analysis culminating in Madhyamaka.


All Dharmakīrtis are not Created Equal: A Tibetan Madhyamaka/Pramāṇa Debate

Kassor, Constance (Smith College, Northampton, USA)

Gorampa Sonam Senge (go rams pa bsod nams seng ge, 1429-89) is renowned as one of the greatest thinkers in the Sakya tradition of Tibetan Buddhism, as well as one of the fiercest philosophical opponents of Tsongkhapa (tsong kha blo bzang grags pa, 1357-1419), founder of what later came to be called the Gelug school. In spite of their philosophical differences, both Gorampa and Tsongkhapa rely heavily on the Prāsaṅgika-Madhyamaka interpretation of Nāgārjuna put forth by Candrakīrti, and they are both influenced by Dharmakīrti’s Pramāṇa tradition. They disagree, however, with respect to the ways in which Dharmakīrti’s epistemological concerns can be said to align with Candrakīrti’s Madhyamaka project.

This disagreement concerning epistemology is informed by each thinker’s primary epistemological influences; Tsongkhapa understands Dharmakīrti through the lens of Chapa Chokyi Senge (phya pa chos kyi seng ge, 1109-69), and Gorampa relies on the interpretation of Sakya Paṇḍita (sa skya paNDi ta kun dga’ rgyal mtshan, 1182-1251). As a result of these different influences, Gorampa and Tsongkhapa come to radically different conclusions about the role of reason on the path to enlightenment, as well as about the nature of enlightenment itself. For example, Tsongkhapa contends that reason extends on the Buddhist path all the way to buddhahood. A fully enlightened Buddha can – and in fact must – engage in rational analysis. Gorampa, on the other hand, argues that reason is essential on the Buddhist path, but that eventually, reason is abandoned in favor of a state that is “free from conceptual proliferations” (spros bral). As such, a Buddha is incapable of engaging in rational analysis.

This paper addresses one way in which Gorampa utilizes both Madhyamaka and Pramāṇa arguments to argue against the philosophical views of Tsongkhapa. Specifically, I focus on the debate in Tibetan Buddhist thought regarding the distinctions between Svātantrika-Madhyamaka and Prāsaṅgika-Madhyamaka, showing the ways in which Gorampa’s interpretation of Pramāṇa in his encyclopedic text Synopsis of Madhyamaka (dbu ma’i spyi don) influences his criticism of Tsongkhapa’s “eight difficult points” (dka’ gnad brgyad) that distinguish the Prāsaṅgika view from that of the Svātantrika.

Gorampa argues that while Svātantrika and Prāsaṅgika differ with respect to methods of rational analysis, these two traditions do not differ with respect to their views of the ultimate. Although the final result of the Buddhist path involves a state that is free from conceptual proliferations, Gorampa argues that analytical reasoning is nevertheless essential in order for a practitioner to arrive at this state. In other words, the methods of reasoning employed by the Svātantrikas and Prāsaṅgikas are different, but both claim that these methods lead to spros bral.

Tsongkhapa, on the other hand, argues that Prāsaṅgikas are different than (and therefore superior to) Svātantrikas – not only with respect to methods of rational analysis, but also with respect to views concerning the ultimate. This is because Tsongkhapa’s interpretation of epistemology lends itself to a view which insists that a realization of the ultimate truth involves the cultivation of a certain concept of emptiness, rather than the complete cessation of concepts.

The key to understanding these different interpretations of Svātantrika and Prāsaṅgika lies in understanding the ways in which Gorampa and Tsongkhapa attempt to unite Madhyamaka and Pramāṇa. Tsongkhapa’s attempt to unify Candrakīrti and Chapa results in a view that emphasizes conceptual thought, while Gorampa’s synthesis of Candrakīrti and Sakya Paṇḍita results in a view that negates conceptual thought in its entirety.


And Ne’re the Twain Shall Meet: Candrakīrti’s Critique of the Logical–Epistemological School

MacDonald, Anne (Austrian Academy of Sciences, Vienna, AUT)

In the Prasannapadā and other of his works Candrakīrti uncompromisingly critiques fundamental theories of the logical–epistemological school, deeming them unfit for use by Mādhyamikas, and in many cases, inapplicable even within the context of the logical–epistemological school itself. It is especially with regard to Dignāga’s (and subsequently Dharmakīrti’s) presentation of pratyakṣa that Candrakīrti expresses disapproval, in the Catuḥśatakaṭīkā brazenly asserting that the epistemologist, like a young child, requires basic training in the means of valid cognition. Of greater consequence to the position of his interlocutor is Candrakīrti’s claim that the epistemologist’s understanding of perception contravenes teachings of the Buddha.

The paper aims to outline Candrakīrti’s main objections to the views of the logical–epistemological fold and to elaborate on the reasons behind them, and, in doing so, to provide a historical–philosophical backdrop for other papers on the panel which deal with later discussions related to either the melding of Madhyamaka and logical–epistemological theory or the deliberate segregation of the former from the latter.


Sensitive Cognitive Episodes, Emptiness, and the Goal of Knowledge

Stoltz, Jonathan (University of St. Thomas, St. Paul, USA)

This essay examines one point of intersection between Nāgārjuna’s account of emptiness and Dharmakīrti’s account of knowledge. More specifically, I discuss one way in which “emptiness of intrinsic existence” is related to the question of whether a given cognitive episode is an instance of pramāṇa. Though there are potentially many connections between these two items, my presentation will focus on just a single issue – the thesis that, in contemporary epistemology, is called the “sensitivity” of beliefs. A belief (i.e., an instance where S believes that p) is said to be sensitive if and only if “Had p been false, S would not have believed that p.” In the Buddhist tradition the thesis of sensitivity is frequently captured in one particular use of the notion avinābhāva (med na mi ’byung). The claim is that, in order for one to have knowledge, it must be the case that if the object of cognition had not existed, the cognition would not have occurred ([don] med na [blo] mi ’byung). In my talk I will discuss the connection that this application of avinābhāva has to both pramāṇa theory and to the emptiness of intrinsic existence.

My discussion will focus on the role that avinābhāva plays in Chaba Chokyi Senge’s (phywa pa chos kyi seng ge) account of knowledge (pramāṇa, tshad ma). I will explain how this thesis is connected to the contemporary doctrine of sensitivity in the analytic tradition of epistemology, and will explain why sensitivity is sometimes thought to be a necessary condition for knowledge. Finally, I will discuss how the thesis of sensitivity is connected to the doctrine of emptiness espoused by Nāgārjuna and his followers. I will argue that this use of avinābhāva connects up with criticisms that Nāgārjuna raised against the possibility of pramāṇa in his Vigrahavyāvartanī.


Inference for All Mādhyamikas: Pa tshab’s Alternative to Svatantra

Vose, Kevin (College of William and Mary, Williamsburg, USA)

We know well that the dominant strain of Madhyamaka in Tibet rejects the use of svatantra inference to establish or realize the central teaching of emptiness. Whether or not svatantra inference can be equated with the form of inference adduced by Dignāga and Dharmakīrti has occasioned significant debate both among Tibetan Buddhists and contemporary scholars. Among the issues at stake in this debate is the possibility that—having rejected svatantra inference—Mādhyamikas might employ some other kind of inference. The well-rehearsed dGe lugs pa answer, that svatantra need not mean Dharmakīrtian inference, allows this tradition to yoke together Candrakīrti’s interpretation of Madhyamaka with Dignāga-Dharmakīrti epistemology. This route, however, was not taken by Tibet’s first Prāsaṅgikas, who saw Candrakīrti’s rejection of svatantra to be aimed squarely at the Buddhist epistemological tradition.

This paper examines the writings of Pa tshab Nyi ma grags (c. 1055-1145) to explore how this pivotal figure understood a Mādhyamika’s use of inference. Pa tshab translated Candrakīrti’s Prasannapadā and Madhyamakāvatārabhāṣya into Tibetan and wrote commentaries on the Prasannapadā and Nāgārjuna’s Mūlamadhyamakakārikās. In these commentaries, he offers a thorough examination of the problems with svatantra, with reference to passages in Dharmakīrti’s Pramāṇavārttika and Pramāṇaviniścaya that are infrequently brought to bear on this issue. After briefly surveying that analysis, this paper focuses on Pa tshab’s discussion of a kind of inference that can be useful in inducing realization of emptiness. Pa tshab’s preferred form of logic only negates an opponent’s false conception, making no attempt to establish emptiness. Yet, in doing so, the logical form gives rise to the opponent’s knowing emptiness. We can see, then, that Pa tshab offers what may be the earliest attempt (beyond Candrakīrti’s own tender of asmad-anumāna) to offer an alternative to svatantra inference. Further, Pa tshab’s presentation of a logical form intended to induce knowledge of the final nature of things suggests that considering his work to be a form of skepticism may miss important features of his thought.


IABS 2014 | Universitätsring 1  | 1010 Wien