BUDDHISM—SCHOOLS: MADHYAMAKA

Madhyamaka is one of the two major schools of Mahāyāna Buddhist philosophy (the other being Yogācāra. It traces its origins to the work of the South Indian philosopher Nāgārjuna (c. 150 CE), who first gave systematic philosophical expression to insights articulated in the earliest Mahāyāna sūtras, the Prajñāpāramitā literature. Central to those texts was the claim that all things thought to be ultimately real are in fact "empty" or devoid of intrinsic nature. The Madhyamaka school arose out of efforts to defend this claim and explore its consequences. The Madhyamaka understanding of the concept of emptiness, and the dialectical strategies used to establish its validity, played central roles in the development of Mahāyāna thought in India and subsequently in Tibet and East Asia.

Emptiness as Lack of Intrinsic Nature

When the Mādhyamikas say that all things are empty (śūnya ), what they mean is that nothing bears an intrinsic nature (svabhāva ). To understand this claim, one must consider the concept of intrinsic nature as it was developed in the scholastic Abhidharma phase of Buddhist philosophy. It is a basic teaching of Buddhism that suffering is caused by one's ignorance of the truth of nonself: that one does not have a separately existing self and that what one thinks of as an enduring person just consists in a causal series of impermanent, impersonal physical and mental events. Philosophers of the Abhidharma schools sought to buttress this conclusion by arguing that all partite entities (wholes made up of parts) are conceptually constructed and thus not ultimately real. This would enable them to claim that the person is conceptually constructed out of the psychophysical elements making up a causal series and so is not itself objectively real.

The general argument is that a partite thing such as a chariot borrows all its properties from the properties of its parts: There is no fact about a chariot that cannot be explained strictly in terms of facts about its parts and their relations. This is taken to show that positing the chariot as an additional entity is superfluous, something one is inclined to do only because of facts about one's interests and cognitive limitations. Since this holds as well for the person, as a whole made up of the elements in a causal series, it follows that one's view of oneself as an enduring substance reflects a failure to distinguish between a mere useful fiction and what is ultimately or mind-independently real.

This reductionist line of thought in the Abhidharma rests on the assumption that there are entities that are ultimately real. To say that persons and other partite things are not ultimately real because they are conceptually constructed is to assume that there are those ultimately real things out of which partite things are constructed. Now conceptually constructed things were held to borrow their properties from the properties of their parts. So Abhidharma thinkers concluded that ultimately real things must have their natures intrinsically. Only that is ultimately real, they claimed, that "is found under analysis," that is conceptually irreducible. The Madhyamaka claim that all things are empty is meant to contest the Abhidharma view that there could be such things. Through a wide variety of arguments the Mādhyamikas seek to demonstrate the absurd consequences that would follow if it were held that there are entities with intrinsic nature.

The Argument from Causation

One such argument concerns the causal relation. It is a fact of one's experience that existing things are impermanent, and this would seem to hold for whatever is ultimately real. But it is also a fact of one's experience that things do not come into or go out of existence in an utterly random way. There seem instead to be patterns of regular succession. So an adequate account of the nature of reality seems to require that ultimately real things be said to arise and cease in accordance with causal laws. At this point the Mādhyamikas raise a simple question: Are cause and effect identical or distinct? Consider the first possibility. Certain Indian philosophers held that the effect is identical with the cause—that causation represents just the manifestation of what already exists in the cause in unmanifest form. But this view is readily dismissed. For it requires that there already exist something with the intrinsic nature of the effect before the effect is produced. And in that case one must wonder why one would set about trying to produce the effect. One might build a fire because one is cold and wants the heat of fire. But if the fire already existed in its cause, the fuel, then its heat should already be present there, so it would be pointless to build a fire.

If, on the contrary, cause and effect were distinct things, two difficulties would follow. First, if these are genuinely distinct things, some account must be given as to why things of the first sort regularly give rise to things of the second sort. Why should fuel give rise to fire and not, say, to cheese? The stock answer to this question is that fuel possesses the causal power to produce fire. But now it must be asked whether this causal power is a third thing that is distinct from both cause and effect or is rather identical with one or the other. If it is distinct from the cause, one may then ask why this sort of cause should be conjoined with just this sort of causal power. This quickly leads to an infinite regress: A second causal power will be required to account for the occurrence of the first, a third for that of the second, and so on. But if the causal power is identical with the cause, then no answer has been given to the original question, and likewise if the causal power is identical with the effect.

The second problem for the view that cause and effect are distinct things is that it is then unclear when the cause produces the effect. To call one thing the cause of another is to say that the first produces the second, so surely there must be some time when this productive activity takes place. There are three possibilities here: when the effect already exists, when the effect does not yet exist, and when the effect is coming into existence. The first is clearly ruled out, since production of something that already exists would be redundant. The second is likewise wrong, for something may be said to be productive only if there is some actually existing product. And with respect to things that are ultimately real, there could be no third time during which they are coming into existence. With respect to partite things like chariots it makes sense to speak of a process of assembly during which the entity is undergoing production. But this is possible only because the chariot is made of parts. Something impartite that bore its nature intrinsically could only be said to be either existent or nonexistent; a third intermediate time is ruled out for such a thing. The upshot of all this is that it appears impossible to account for the causal relations that should obtain among things with intrinsic natures.

The Argument from the Property-Bearer Relation

A second Madhyamaka argument for emptiness involves examining the relation between an ultimately real thing and its intrinsic nature. Either these are distinct or they are identical. If they are distinct, a number of difficulties follow. First, there is the problem of saying what the entity itself is like apart from its intrinsic nature. Since the notion of a pure propertyless substrate seems incoherent, this problem is likely to prove intractable. But there is also the difficulty that then the entity's acquiring its nature will depend on causes and conditions. Such dependence seems incompatible with calling its nature intrinsic; it then seems more appropriate to say that the entity borrows its nature from other things.

Suppose then that the entity and its intrinsic nature are identical—that one's distinction between the thing and its nature merely reflects the concepts one uses. In that case an occurrence of what one calls fire is really just the occurrence of heat (the property of being hot). But then the question arises how fires are to be individuated. Suppose there are two distinct fires of equal intensity. Each fire is just its heat, and the two heats are identical in nature. Ordinarily, one would say that the two occurrences of heat are distinct because each occurs in a distinct particular (the fire whose heat it is). But on the hypothesis under scrutiny there are no particulars over and above the property of heat; the occurrence of what one judges to be a particular fire just is the occurrence of heat. One might then suppose that each fire is individuated in terms of the discrete space that it occupies. But then the question arises what makes two spaces discrete? Suppose the intrinsic nature of a space is its nonresistance. Since one is now supposing that the existence of a space just is the occurrence of a certain nonresistance, it is not clear what will make two spaces distinct, unless it is their being occupied by distinct entities, such as two fires. But now one has come full circle. So it looks as if the hypothesis that entity and intrinsic nature are identical does not hold up to critical scrutiny either. It appears that no adequate account can be given of how something could have an intrinsic nature.

Madhyamaka as Nihilism

A host of similar arguments against things with intrinsic nature was developed by Mādhyamika philosophers such as Nāgārjuna, Āryadeva (170–270), Buddhapālita (c. 500), Bhāvaviveka (500–570), and Candrakīrti (c. 600–650). Nāgārjuna's targets were chiefly views held by Ābhidharmikas, but Āryadeva extended the scope of Mādhyamika dialectics to include the views of non-Buddhist Indian philosophers, a practice that becomes systematic in Bhāvaviveka's Tarkajvālā. Suppose that these arguments succeed in showing that nothing could bear an intrinsic nature. Suppose also that the Ābhidharmikas were correct in concluding that only something with an intrinsic nature could be ultimately real. What would then follow? What should one make of the Mādhyamika doctrine of emptiness? Modern scholars have put forward a wide variety of interpretations, but there is also some difference of opinion among classical Indian authors. One modern reading, that of Thomas E. Wood (1994) and David Burton (1999), that is also the common view of the Mādhyamikas' ancient Indian critics is that the doctrine of emptiness is tantamount to metaphysical nihilism, the thesis that reality is ultimately devoid of existing things. The stock characterization of the Mādhyamikas that one finds in the writings of their classical opponents is that the Mādhyamikas believe nothing whatever exists.

Of course, the thesis of metaphysical nihilism is virtually self-refuting: If nothing whatever existed, the thought could not occur that it might be true. Still, attributing this thesis to the Mādhyamikas might not seem unfair. If there is reason to believe that only things with intrinsic nature could be ultimately real, then demonstrating the incoherence of the concept of a thing with intrinsic nature seems equivalent to showing that ultimately nothing whatever is real. One major difficulty with this interpretation, however, is that it is explicitly argued against by the Mādhyamikas. Thus, Nāgārjuna points out that to understand the thesis that no ultimately real things exist, one must first understand what it would mean for there to be ultimately real things. But an ultimately real thing would have to be something with intrinsic nature. Since the Mādhyamikas claim there can be no such things, they would say one cannot understand the thesis that ultimately nothing exists. So perhaps they should not be interpreted as seeking to establish metaphysical nihilism.

Do MĀdhyamikas Affirm Contradictions?

Of the remaining interpretations of emptiness found in the modern scholarship, only some find support in the original sources. (Of course, the lack of such support need not detract from the philosophical significance of an interpretation.) For instance, Graham Priest and Jay L. Garfield (2002) claim Nāgārjuna as perhaps the first exponent of dialetheism, the view that there are true contradictions that arise at the limits of thought. As evidence, they cite his assertion that it cannot be ultimately true that all things are empty (Madhyamakakārikā chapter 22, verse 11). The notion of ultimate truth at work here derives from the Abhidharma distinction between two kinds of truth: conventional and ultimate. Only statements concerning ultimately real things can be said to be ultimately true; statements concerning such mere conceptual fictions as chariots and persons can only be conventionally true. For Abhidharma, then, the set of ultimately true statements would give the complete account of all those things with intrinsic natures; it would be a complete description of the ultimate nature of reality.

Now the Mādhyamikas claim to have shown that the only statement that can truly be made about those things that are thought to be ultimately real is that they are empty. But in the verse in question Nāgārjuna says that this cannot be ultimately true. Indeed, he says it is not ultimately true that all things are empty, or that they are nonempty, or both or neither. The reason for this is that emptiness is itself a mere conceptual fiction. So any statement about emptiness could at best be conventionally true. Priest and Garfield take Nāgārjuna to be thereby asserting both that the ultimate truth cannot be characterized and that it can be characterized (namely as being uncharacterizable). But this is not how Mādhyamika commentators have understood the verse. Instead, they assimilate it to the Buddha's treatment of the so-called indeterminate questions (avyākṛta ). When, for instance, the Buddha was asked whether the enlightened person survives death, does not survive death, both survives and does not survive, or neither survives nor does not survive death, the Buddha rejected all four possibilities. One can consistently do this, they explain, because all share an implicit presupposition—that there ultimately is such a thing as an enlightened person—and this presupposition should be rejected. By the same token, the commentators say, Nāgārjuna should be understood as rejecting the presupposition that there is such a thing as the ultimate truth. In that case he asserts neither that the ultimate truth is uncharacterizable nor that it can be characterized. He does not hold that a contradiction is true.

Madhyamaka as Skepticism

Other interpreters of the doctrine of emptiness, such as Thomas McEvilley (1982) and Bimal Krishna Matilal (1986), see it as a form of skepticism. This reading is suggested by the Mādhyamika response to objections coming from Indian epistemologists. The thrust of these objections is that since the Mādhyamikas hold all things to be empty, they must hold that all means of knowledge are empty. But in that case it cannot be known that all things are empty, so the Mādhyamika claim is a mere dogmatic assertion. Part of the Mādhyamika response involves calling into question the epistemologist's project of determining which are the means of knowledge. For instance, they argue that a given procedure can be known to be a means of knowledge—a reliable cause of veridical belief—only if one already possesses some means of knowing which beliefs are true. Thus, any attempt to determine which are the means of knowledge either is circular or else leads to an infinite regress.

An argument of this sort might be used to support the skeptic's claim that one can never know which, if any, of one's beliefs amount to knowledge. But this is not how the Mādhyamikas themselves see such arguments. For one thing, the skeptical conclusion requires the additional assumption that one can only know some statement p if one knows that one knows p—an assumption that neither the Mādhyamikas nor their opponents seem to have held. Second, nowhere do the Mādhyamikas appeal to the sorts of error possibilities that are the skeptic's stock in trade, such as perceptual illusions, hallucinations, dreams, and the like. Indeed, the Mādhyamikas do not deny that, conventionally, certain procedures can count as means of knowledge. What they deny is just that anything could ultimately be a means of knowledge, that anything could have the intrinsic character of reliably causing veridical beliefs as part of its mind-independent essential nature. The Mādhyamika epistemological stance seems to be that something can be a means of knowledge only through its relations to other things that are themselves equally empty of intrinsic nature. The resulting view may have its affinities with some forms of skepticism (particularly Pyrrhonian skepticism). But its chief concern is not to call into question the possibility of knowledge, but to deflate the pretensions of a certain sort of epistemological realism.

The MĀdhyamikas as Mystics or as Quietists?

Two interpretations of emptiness seem more firmly grounded in the self-understanding of the Mādhyamika tradition. The first sees emptiness as leading to a kind of mystical silence. Mādhyamika arguments are said to demonstrate that no set of concepts can ever adequately represent the world. This realization is said to then usher in a nondiscursive grasping of the nature of reality (perhaps through a kind of intuition that is cultivated in meditation). On this interpretation, emptiness serves to point to an ultimate reality that lies beyond the reach of philosophical rationality. The second of the two, by contrast, sees emptiness not as pointing to an ineffable ultimate, but as indicating that the very idea of an ultimate nature of reality is incoherent. The exercise of philosophical rationality leads not to the silence of the beyond, but back to the conventional. For Mādhyamika dialectic reveals the error in the notion of an ultimate truth that represents how things are independently of all facts about the cognizer. This shows that truth can only be transactional, a matter of what facilitates interactions among creatures like us. The notion of a truth that potentially outstrips all our conceptual resources is revealed to be no more than a useful fiction.

The "mystical silence" reading of emptiness has been championed by T. R. V. Murti (1955) and David Seyfort Ruegg (1977) among others. The second reading is commonly called a quietist interpretation, since it grows out of the attempt by Frederick J. Streng (1967) to read elements of the later Ludwig Josef Johann Wittgenstein into Nāgārjuna. But as developed by Tom J. F. Tillemans (2002), it has clear affinities with both antirealist and minimalist conceptions of truth. Both readings may be seen as seeking to explicate the claim that insight into emptiness results in a kind of nondual awareness. Mahāyāna Buddhist texts commonly claim that final liberation from suffering requires a kind of seeing that transcends all problematic dualities. On the mystical silence interpretation it is the dualism fostered by conceptualization that is to be overcome, for without concepts one cannot make such invidious distinctions as that between cognizing subject and object. On the quietist reading it is the dualism of ultimate and conventional truth that is erased through knowledge of emptiness. Presumably, this duality is problematic because the notion of ultimate truth as correspondence to mind-independent reality fosters a subtle form of belief in a self, namely that expressed through attachment to metaphysical theories.

Each reading is not without its own difficulties. For the mystical silence view, the chief problem is to explain how Madhyamaka then differs from other views that posit an ineffable ultimate, such as the absolute monisms of Advaita Vedānta and Parmenides of Elea. For the quietist there is the difficulty of explaining how there can be truth without there being such a thing as how the world is anyway. This problem is sometimes addressed by claiming that what emptiness really shows is just that no entity has a nature that is independent of its relations to other things. But to the extent that this addresses the problem of grounding truth in mind-independent reality, it contravenes the quietist claim to be showing a way out of the trap of metaphysical theories.

The SvĀtantrika-PrĀsaṄgika Distinction

Modern studies of the Mādhyamikas have profited enormously from contact with the Tibetan tradition, a tradition for which Mādhyamika thought continues to play a crucial role. But there are cases in which reliance on Tibetan doxographical categories has led to distortion of the Indian Mādhyamika sources. A case in point is the alleged distinction between two schools of Madhyamaka: Svātantrika and Prāsaṅgika. This distinction was invented by Tibetan doxographers, and it is a matter of some dispute to what extent it reflects substantive differences in the views of the Indian thinkers covered by the classification. It is in any event clear that Indian Mādhyamikas did not see themselves as falling into two camps to which these labels could be applied.

Those who accept the distinction identify a dispute between Bhāvaviveka and Candrakīrti as its point of origin. The dispute concerns the proper methodology for a Mādhyamika. The arguments of Bhāvaviveka's Mādhyamika predecessors were usually expressed in the reductio ad absurdum (prasaṅga ) style: The hypothesis to be refuted (e.g., that something with intrinsic nature could be an effect) is considered and then shown to lead to some absurd result (e.g., that its intrinsic nature is actually extrinsic). Employing the methods of the Buddhist logician Diṅnāga, Bhāvaviveka sought to convert such reductios into independent arguments (svatantra anumāna ). Thus, one would have:

It is not the case that ultimately an entity arises from distinct causes and conditions. Because of depending on them for its nature.

Whatever depends on other things for its nature is not ultimately real, like the chariot. Candrakīrti disagrees, claiming that the Mādhyamikas may only use reductios. But since the two types of argument turn out to be formally equivalent once the reductio has been fully spelled out, it may not be clear what the dispute is actually about.

The difference Candrakīrti sees between them is this: In giving a reductio one need not assert anything to be the case oneself; the proponent merely shows the opponent the inconsistency in his or her view, thereby impelling the opponent to withdraw assent from his or her thesis. In the case of an independent argument, on the contrary, both the proponent and opponent must agree about such things as the subject (in this case, an entity), the pervasion (that what is dependently originated is not ultimately real), and the example (the chariot). But the Mādhyamika proponent holds that entities can only exist conventionally, while the opponent thinks some entities are ultimately real, so the two sides do not agree about the subject. And likewise for the other elements of the argument that require consensus. From the perspective of the Mādhyamikas, the opponent is simply, hopelessly wrong about everything. So there can be no common framework for resolving their disagreement. Instead, the Mādhyamikas should just give their opponents the rope with which to hang themselves.

Syncretism in Madhyamaka

One may wonder if the opponent will be so obliging toward a proponent who seems to speak a different (and perhaps unintelligible) language. But there may be a deeper point here. Those Tibetan commentators such as Tsong-kha-pa (1357–1419) who align themselves with the Prāsaṅgika allege that Svātantrikas have not fully realized emptiness, since they continue to posit intrinsic natures, albeit at just the conventional level. And it is true that those Mādhyamikas who are identified as Svātantrikas exhibit a tendency toward syncretism, seeking to incorporate the views of overtly metaphysical Buddhist schools within an overall Mādhyamika framework. This tendency is especially clear in Śāntarakṣita (eighth century), who embraces Dharmakīrti's formulation of the Yogācāra school's subjective idealist ontology and epistemology. But it can already be seen in Bhāvaviveka, who champions the Sautrāntika school's realism about physical objects and its associated representationalist theory of sense perception. In neither case is the other school's view identified as anything more than the best way of representing conventional truth. As Mādhyamikas, Śāntarakṣita and Bhāvaviveka remain committed to the position that the only ultimate truth is that there is no ultimate truth (i.e., that all things, including emptiness, are empty). Still, they do take a position on the question whether external objects exist conventionally.

Candrakīrti does as well. He sides with Bhāvaviveka in rejecting the idealist view. But his reasons are different. Where Bhāvaviveka tries to answer Yogācāra arguments against the existence of physical objects, Candrakīrti simply dismisses the arguments. For him such arguments can only show that physical objects are ultimately empty—something a Mādhyamika already knows. But by the same token mental states (which the idealist thinks are real) are equally empty. So the availability of philosophical arguments against the conventional belief in external objects cannot show that they are not conventionally real. While Bhāvaviveka thinks the use of philosophical rationality can lead to improvements in one's conventional account of the world, Candrakīrti thinks it can only lead to the ultimate truth of emptiness. Conventional truth neither needs nor can it sustain either refinement or defense at the hands of philosophers. It is just simply that which is given through the everyday practices of ordinary people.

Given this difference in attitude, one can see why Svātantrikas might be described by their critics as positing conventionally real intrinsic natures. It is, after all, philosophical analysis that gives rise to the demand for things with intrinsic nature. So if philosophical rationality is allowed to play a role in shaping one's conventional worldview, the resulting theory will be committed to there being such things, the things at which analysis stops. And to Candrakīrti, Bhāvaviveka's demand that the Mādhyamikas give independent arguments and not mere reductios looks like a requirement that the Mādhyamikas construct a philosophically defensible version of the conventional truth. This will inevitably lead in the direction of syncretism, and with it the danger that the Mādhyamikas will become ensnared in metaphysical theories. The Prāsaṅgika side in this dispute is not without its dangers as well though. For on its account, conventional truth does not allow of progressive improvement, it can only be utterly overthrown. The result would seem to be a strong form of relativism about conventional truth. And an opponent could always use this to turn back the Prāsaṅgika's reductio arguments, in effect saying to the Mādhyamikas, "We simply disagree about whether there is an inconsistency in my position, and in such matters there is no right and wrong." What this dispute brings out, then, is a tension that seems inherent in the concept of truth, a tension that is also reflected in current debates between realists and antirealists.

Indian Madhyamaka came to an end in the late twelfth century, when all Buddhist philosophical activity ceased in India following the Turkish invasion. Madhyamaka has continued to play a prominent role in Tibetan Buddhism to this day. It also enjoyed some popularity among Chinese Buddhist philosophers, playing an important role in the development of the Huayan school. Perhaps a case might even be made for its having had a profound impact on Chan Buddhism. Chan formally eschews the study of precisely those sorts of doctrinal texts that form the core of Mādhyamika practice. But it does make extensive use of paradox in some of the methods it has devised for helping the adept attain enlightenment. Analysis of the structural features of those paradoxes and their uses might reveal more than merely superficial resemblances with the dialectical strategies of Madhyamaka.

See also Buddhism; Buddhism—Schools: Dge-lugs.

Bibliography

primary sources and translations

Āryadeva. "Catuḥśataka." In Āryadeva's Catuḥśataka, edited and translated by Karen Lang. Copenhagen: Akademisk Vorlag, 1986.

Bhāvaviveka. Madhyamakahṛdayakārikā Tarkajvālā. In Reason and Emptiness: A Study in Logic and Mysticism. Translated by Iida Shōtarō. Tokyo. Hokuseido Press, 1980.

Nāgārjuna. "Mūlamadhyamakakārikā." In Mūlamadhyamakakārikās (Mādhyamikasūtras) de Nāgārjuna avec la Prasannapadā, Commentaire de Candrakīrti, edited by Louis de La Vallée Poussin. Osnabrück, Germany: Biblio Verlag, 1970. Also in The Fundamental Wisdom of the Middle Way: Nāgārjuna Mūlamadhyamakakārikā. Translated by Jay L. Garfield. New York: Oxford University Press, 1995.

Nāgārjuna. "Vigrahavyāvartanī." In The Dialectical Method of Nāgārjuna: Vigrahavyāvartinī. Translated by Kamaleswar Bhattacharya and edited by E. H. Johnston and Arnold Kunst. Delhi: Motilal Banarsidass, 1978.

Śāntarakṣita. "Tattvasaṅgraha." In The Tattvasaṅgraha of Śāntarakṣita, with the Commentary of Kamalaśīla, edited by Embar Krishnamacharya. Baroda, India: Oriental Institute, 1984. Also in The Tattvasaṅgraha of Śāntarakṣita, with the Commentary of Kamalaśīla. Translated by Ganganatha Jha. Delhi: Motilal Banarsidass, 1986.

Śāntideva. "Bodhicāryāvatāra." In The Bodhicāryāvatāra of Ārya Śāntideva with the Commentary Pañjika of Śrī Prajñākaramati, edited by Dwārikādās Śāstrī. Varanasi, India: Bauddha Bharati, 2001. Also in The Bodhicaryāvatāra. Translated by Kate Crosby and Andrew Skilton. New York: Oxford University Press, 1993.

secondary literature

Burton, David. Emptiness Appraised: A Critical Study of Nāgārjuna's Philosophy. Richmond, U.K.: Curzon, 1999.

Dreyfus, Georges B. J., and Sara McClintock, eds. The Svātantrika-Prāsaṅgika Distinction: What Difference Does a Difference Make? Boston: Wisdom Publications, 2002.

Eckel, Malcolm David. Jñānagarbha's Commentary on the Distinction between the Two Truths: An Eighth Century Handbook of Madhyamaka Philosophy. Albany, NY: SUNY Press, 1987.

Hayes, Richard P. "Nāgārjuna's Appeal." Journal of Indian Philosophy 22 (1994): 299–378.

Huntington, C. W., Jr., with Geshé Namgyal Wangchen. The Emptiness of Emptiness: An Introduction to Early Indian Mādhyamaka. Honolulu: University of Hawaii Press, 1989.

Lopez, Donald S., Jr. A Study of Svātantrika. Ithaca, NY: Snow Lion, 1987.

Matilal, Bimal Krishna. Perception: An Essay on Classical Indian Theories of Knowledge. New York: Oxford University Press, 1986.

McEvilley, Thomas. "Pyrrhonism and Mādhyamika." Philosophy East and West 32 (1) (1982): 3–35.

Murti, T. R. V. The Central Philosophy of Buddhism: A Study of the Mādhyamika System. London: Allen and Unwin, 1955.

Oetke, Claus. "Remarks on the Interpretation of Nāgārjuna's Philosophy." Journal of Indian Philosophy 19 (3) (1991): 315–323.

Potter, Karl H., ed. Encyclopedia of Indian Philosophies, Vol. 8: Buddhist Philosophy from 100 to 350 A.D. Delhi: Motilal Banarsidass, 1999.

Potter, Karl H., ed. Encyclopedia of Indian Philosophies, Vol. 9: Buddhist Philosophy from 350 to 600 A.D. Delhi: Motilal Banarsidass, 2003.

Priest, Graham, and Jay L. Garfield. "Nāgārjuna and the Limits of Thought." In Beyond the Limits of Thought, edited by Graham Priest. New York: Oxford University Press, 2002.

Robinson, Richard H. Early Mādhyamika in India and China. Madison: University of Wisconsin Press, 1967.

Ruegg, David Seyfort. "The Uses of the Four Positions of the Catuṣkoṭi and the Problem of the Description of Reality in Mahāyāna Buddhism." Journal of Indian Philosophy 5 (1977): 1–71.

Ruegg, David Seyfort. The Literature of the Madhyamaka School of Philosophy in India. Wiesbaden, Germany: Harrassowitz, 1981.

Siderits, Mark. "The Madhyamaka Critique of Epistemology I." Journal of Indian Philosophy 8 (1980): 307–335.

Siderits, Mark. "Thinking on Empty: Madhyamaka Anti-realism and Canons of Rationality." In Rationality in Question: On Eastern and Western Views of Rationality, edited by Shlomo Biderman and Ben-Ami Scharfstein. Leiden, Netherlands: Brill, 1989

Streng, Frederick J. Emptiness: A Study in Religious Meaning. Nashville, T.N.: Abingdon Press, 1967.

Tillemans, Tom J. F. Materials for the Study of Āryadeva, Dharmapāla, and Candrakīrti: The Catuḥśataka of Āryadeva, chapters XII and XIII, with the commentaries of Dharmapāla and Candrakīrti. 2 vols. Vienna: Arbeitskreis für Tibetische und Buddhistische Studien Universität Wien, 1990.

Tillemans, Tom. "Metaphysics for Mādhyamikas." In The Svātantrika-Prāsaṅgika Distinction: What Difference Does a Difference Make? edited by Georges B. J. Dreyfus and Sara McClintock. Boston: Wisdom Publications, 2002.

Williams, Paul. "On the Interpretation of Madhyamaka Thought." Journal of Indian Philosophy 19 (1991): 191–218.

Wood, Thomas E. Nāgārjunian Disputations: A Philosophical Journey through an Indian Looking-Glass. Honolulu: University of Hawaii Press, 1994.

Mark Siderits (2005)