Madhyamaka
Madhyamaka begins with an argument that sounds almost mischievous: nothing that exists actually arises. That claim comes from the Indian monk and philosopher Nāgārjuna, who wrote around the year 150. His foundational text, the Mūlamadhyamakakārikā, translates as Root Verses on the Middle Way. Nāgārjuna's method was blunt. If cause and effect were truly separate things, he reasoned, fire could just as easily produce darkness. Barley could sprout from a handful of wheat. Since neither of those things happens, the whole notion of things arising from independent causes must be mistaken. What is left, once causation itself starts to look like an illusion? Nāgārjuna called his answer śūnyatā, emptiness. That single word would go on to shape Buddhist philosophy across Tibet and East Asia for centuries. But what exactly is empty, and empty of what? Does denying that things truly arise mean nothing exists at all? And how does a philosophy built on refuting nearly everything avoid collapsing into nihilism itself?
The Sanskrit word madhya simply means middle. Add the suffix -ma, a superlative marker, and the result is madhyama, meaning mid-most or medium. A further suffix, -ka, forms adjectives, so madhyamaka comes out as middling. Swap in -ika instead and the word becomes mādhyamika, meaning belonging to the mid-most, the term used for the school's adherents. In both words, the stress falls on the first syllable.
In a Buddhist context these terms point to the madhyamā pratipad, the middle path, a right view that steers clear of two opposite extremes. That same steadiness between extremes runs through nearly every argument Madhyamaka would later make.
Every Buddhist school claimed to walk some version of this middle path. But the name Madhyamaka came to mark one particular school, the one built around Nāgārjuna and the commentators who followed him.
Richard P. Hayes, a scholar cited in accounts of the tradition, reads svabhāva two ways: as identity, or as causal independence. Jan Westerhoff adds that the term carries both an ontological side and a cognitive one. On the ontological side, svabhāva means essence, the property that makes a thing what it is, and also substance. Candrakīrti, one of the tradition's central commentators, defines substance plainly: something that does not depend on anything else. It is this independent, substance-svabhāva that most Madhyamaka arguments set out to refute.
One recurring tool for dismantling that claim is the catuṣkoṭi, the four corners or tetralemma. A given proposition, in this scheme, may be true, false, both true and false, or neither. Denying svabhāva does not mean denying that things exist in any sense. Conventionally, in everyday speech, Madhyamaka accepts that one can still talk about things. Ultimately, though, those same things are said to be empty of any inherent existence.
Candrakīrti illustrates the mistake with an eye condition: vitreous floaters, which make a person see hairs drifting across their vision that are not really there. Just as understanding how that illusion works does not make it stop, he says, simply following Madhyamaka's logic does not end suffering on its own. Something further is required, a cognitive shift the tradition calls realization, brought about through practice rather than argument alone. Candrakīrti sums up the problem this way. To someone on the road of cyclic existence, pursuing an inverted view out of ignorance, a mistaken superimposition on the five aggregates appears real. It does not appear that way, though, to one close to seeing things as they are.
That shift is what the tradition's arguments are built to trigger, a series of specific refutations aimed at causation, motion, time, and the self.
The 'Diamond Splinters Argument' tests four ways something might arise: from itself, from something other, from both, or from no cause at all. Arising from itself is the first to fail. Something already existing would have to exist again to produce itself, which makes the whole idea meaningless. Taken to its absurd end, a thing would have to keep producing itself endlessly. The Sāṃkhya school held a version of this view, called satkāryavāda, arguing an effect already exists inside its cause, just unmanifest. Critics pointed out that cause and effect still differ in shape, color, and timing. Treating them as identical, critics warned, would collapse the whole universe into what one text calls a colourless, differenceless mass.
Arising from something other fares no better. The Vaiśeṣika school and the Buddhist Sarvāstivāda school both defended this kind of difference between cause and effect. But total otherness means total independence, and independent things could produce anything at all. There is also a timing problem: when the seed exists, the sprout does not, and the reverse holds too. The two, in that sense, never actually meet, 'just as two people who have already been born are not mutually dependent.' A third possibility, arising from both self and other, simply combines the flaws of the first two. A fourth possibility, arising from no cause, is refuted by pointing to the Cārvāka school, which held that the universe simply happened by itself. If that were true, a farmer's effort in planting seeds would count for nothing.
A second line of reasoning asks what kind of result actually emerges from a cause: an existent one, a nonexistent one, both, or neither. An existent result needs no further producing, since it already exists, so that option fails too. A nonexistent result is just as impossible, the tradition argues. No cause can conjure something out of what is utterly nonexistent, 'such as a rabbit's horns.' One text puts it starkly: even millions of causes joining forces could never make something intrinsically nonexistent pass into existence.
If none of the four possibilities holds, the arising people witness every day is no more real than arising in a dream.
'There is no ignorance nor any ending of ignorance,' the Heart Sūtra declares. It is a line from the Prajñāpāramitā literature that Madhyamaka enlists against the idea of true motion. Take any path a thing might travel. On the part already traversed, there is no motion, because motion is finished there. On the part not yet traversed, there is also no motion, because it has not started. Since 'there is no such place that one either has not already been or has yet to go,' no third option remains. The past has already happened and the future has not yet come, so motion cannot be found in either. The present dissolves the moment it is divided into smaller units. This same reasoning extends to a mover moving through saṃsāra, from one life to the next.
The past, present, and future rest on their own definitions in Madhyamaka's account of time. The past is marked by what has ceased, the present by what exists now, the future by what has not yet arisen. If the three times are independent of each other, the whole scheme falls apart. 'The present' only makes sense in relation to past and future. If instead the three depend on each other, the present would have to somehow exist in the past and future too. The conclusion drawn is that all three are 'imperceptible,' and that 'the three times are equality.'
Indian Nyāya philosophers relied on four epistemic instruments to ground knowledge: perception, inference, testimony, and likeness. But if all knowledge depends on these instruments, Madhyamaka asks, how is knowledge of the instruments themselves established? Instruments proven by other instruments lead to an infinite regress, with no final foundation. Instruments that supposedly prove themselves would have to exist independently of the very objects they are meant to reveal. That, the tradition argues, defeats the point of having them at all.
Madhyamaka calls one final argument the 'king of reasonings': the great dependent arising argument, said to contain all the others within it. Nāgārjuna opens the Mūlamadhyamakakārikā with a verse praising the Buddha for teaching that whatever is dependently arisen is 'unceasing, unborn, unannihilated, not permanent, not coming, not going.' That verse sets the stage for how Madhyamaka splits reality itself into two distinct kinds of truth.
Beginning with Nāgārjuna, Madhyamaka works with two levels of truth: conventional truth, saṃvṛti satya, and ultimate truth, paramārtha satya. Saṃvṛti literally means to cover, conceal, or obscure, marking conventional reality as a kind of ignorance. Chandrakirti also glosses saṃvṛti as customary, agreed-upon truth, comparable to a linguistic convention. He offers a third reading too: mutual dependence, parasparasaṃbhavana. Beings do perceive concrete objects in this conventional realm. Those objects function in dependence on causes, conditions, and concepts, so long as no one questions cause and effect too closely.
This conventional, limited truth covers everything, including the Buddha, his teachings, liberation itself, and even Nāgārjuna's own arguments. Nāgārjuna defends the two-truth framework against charges of nihilism directly: 'Without relying upon convention, the ultimate fruit is not taught. Without understanding the ultimate, nirvana is not attained.' According to Karl Brunnhölzl, the two truths are not two separate metaphysical realities at all. They describe what is experienced by two different kinds of beings, with different scopes of perception.
Śāntideva describes it as 'two kinds of world,' one belonging to yogins and one to common people. Buddhapālita explains why the ordinary world clings so hard to appearances: 'unskilled persons whose eye of intelligence is obscured by the darkness of delusion conceive of an essence of things and then generate attachment and hostility with regard to them.' According to Richard P. Hayes, the two truths can also map onto two goals in life: the highest goal of nirvana, and a lower goal he calls 'commercial good.'
Nirvana, in this framework, is reached only by passing through the conventional, not by skipping past it. That raises a harder question: what, exactly, is left once every convention is stripped away?
According to Paul Williams, Nāgārjuna does not treat emptiness as some kind of Absolute. It is simply the absence of true existence in ordinary, conventional things and events. Because the ultimate is itself empty, even the concept of emptiness counts only as conventional, not as the ultimate itself. This point is often called 'the emptiness of emptiness.' Candrakīrti describes the enlightened view this way: 'the noble ones who have accomplished what is to be accomplished do not see anything that is delusive or not delusive.' Nāgārjuna, in his Sixty Stanzas on Reasoning, puts it even more starkly: 'nirvana is the sole reality, is what the Victors have declared.'
Some writers, ancient and modern, have called Madhyamaka nihilistic, a reading scholar Roger Jackson notes has long been disputed. Nāgārjuna answers the charge directly: 'through explaining true reality as it is, the seeming saṃvṛti does not become disrupted.' Candrakīrti goes further in his Lucid Words. He warns that treating emptiness as mere nonexistence only 'increases the web of discursiveness' rather than pacifying it. Scholar David Kalupahana argues against reading emptiness as a transcendental absolute too, since that reading would strip Madhyamaka of its claim to be a middle way at all.
Madhyamaka names its two rejected extremes directly: essentialism, or sastavadava, the belief that things exist substantially and are worth craving, and nihilism, or ucchedavada, which can excuse people from responsibility for their own actions. Khenpo Ngawang Pelzang explains how each is avoided. Freedom from eternalism lies in the fact that causes do not persist into their results. Freedom from nihilism lies in the fact that results keep arising, unceasingly, from the causes that precede them.
Nāgārjuna insisted he held no position at all. In his Vigrahavyāvartanī, he writes: 'If I had any position, I thereby would be at fault. Since I have no position, I am not at fault at all.' In his Sixty Stanzas on Reasoning he warns that 'by taking any standpoint whatsoever, you will be snatched by the cunning snakes of the afflictions.' His student Āryadeva echoes the same progression in the Four Hundred Verses: first ending what is not meritorious, then ending belief in identity, and finally ending all views whatsoever.
Śāntideva and Bhāvaviveka both seem to defend emptiness as Madhyamaka's actual thesis, unlike Nāgārjuna's insistence on holding no position at all. Karl Brunnhölzl resolves the tension by noting that Madhyamaka claims a thesis only conventionally, for debate and teaching, never as something defended from its own point of view. Nāgārjuna's own warning captures the risk of missing that distinction: mishandling a view of emptiness, he writes, is 'just as when incorrectly seizing a snake.'
Jan Westerhoff calls Nāgārjuna 'one of the greatest thinkers in the history of Asian philosophy.' Scholars, he notes, agree on hardly any details about the man himself. His lifetime can only be narrowed to somewhere within the first three centuries CE. Joseph Walser proposes Amarāvatī, in the east Deccan, as the place where he lived. Scholars generally agree that a group of five texts, called the 'Yukti,' or analytical, corpus forms the core of his philosophy. Those five are the Mūlamadhyamakakārikā, the Sixty Stanzas on Reasoning, the Dispeller of Objections, the Treatise on Pulverization, and the Precious Garland. Even the attribution of these has been questioned, except for the Mūlamadhyamakakārikā, treated by default as his defining work.
Chapter 15.7 of the Mūlamadhyamakakārikā contains the only sutra Nāgārjuna cites by name: the Instructions to Kātyāyana. It is understood to be a Sanskrit version of the Pali Kaccānagotta Sutta, paralleled in the Chinese Saṃyuktāgama. That sutra records the Buddha telling Kātyāyana that the world usually relies on a duality of existence and nonexistence. But seeing the world's arising and cessation correctly, the sutra says, shows there is neither permanent existence nor true nonexistence in it. Joseph Walser also traces a second allusion, in the same chapter, to the Mahāhatthipadopama Sutta, part of an early collection of texts focused on avoiding extreme views.
Buddhapālita, active between 470 and 550, developed what later Tibetan writers called the prāsaṅgika approach. He refuted essentialism purely by exposing the absurd consequences of an opponent's own position. Bhāvaviveka, writing around the year 500, disagreed sharply. In his Prajñāpadīpa, he argued that mādhyamikas needed to build proper 'autonomous syllogisms,' svatantra, rather than simply attacking other people's arguments, a tactic Indian philosophical circles considered poor form. Candrakīrti, active between roughly 600 and 650, sided with Buddhapālita instead. He argued that a Madhyamaka thinker and an essentialist opponent do not even share the same basic understanding of what it means for something to exist, so building a shared syllogism between them was pointless from the start. Candrakīrti also turned his prasaṅga method against the Buddhist Yogācāra school, faulting its 'appearance only' doctrine for privileging consciousness over the very objects consciousness perceives.
By the eighth century, Śāntarakṣita had folded Yogācāra idealism into Madhyamaka's conventional level of truth, treating it as a preparatory stage before Madhyamaka's own ultimate truth took over. In his Madhyamakālaṃkāra, he tells students to rely on 'Mind Only,' cittamatra, first, in order to see that external entities do not exist. Tibetan philosophers and commentators later gave this style of argument, building positive proofs rather than only exposing contradictions, its own name: svātantrika, a label that still marks one side of the debate Nāgārjuna's root verses set in motion.
Up Next
Common questions
What is Madhyamaka in Buddhist philosophy?
Madhyamaka is a tradition of Buddhist philosophy and practice founded by the Indian monk and philosopher Nāgārjuna, who wrote around the year 150. Its foundational text is his Mūlamadhyamakakārikā, or Root Verses on the Middle Way. The term also refers more broadly to the ultimate nature of phenomena and to the non-conceptual realization of that reality experienced in meditation.
Who founded Madhyamaka Buddhist philosophy?
Nāgārjuna, an Indian Buddhist monk and philosopher who wrote around the year 150, founded Madhyamaka. Scholars agree on very few details of his life, though Joseph Walser has proposed that he lived in Amarāvatī, in the east Deccan.
What does the word Madhyamaka mean?
Madhyamaka comes from the Sanskrit word madhya, meaning middle, combined with suffixes that give it the sense of middling or mid-most. It refers to the madhyamā pratipad, the middle path between opposite philosophical extremes.
What does emptiness, or śūnyatā, mean in Madhyamaka?
In Madhyamaka, emptiness means that all phenomena, or dharmas, lack svabhāva, any independent essence, nature, or substance that would give them solid, independent existence. Madhyamaka also holds that emptiness itself is empty, meaning it has no existence of its own and does not point to any transcendental reality beyond phenomena.
Where has Madhyamaka philosophy been most influential?
Since the 4th century CE, Madhyamaka has been the dominant interpretation of Buddhist philosophy within Tibetan Buddhism, and it has also strongly influenced East Asian Buddhist thought. Its influence spread alongside the spread of Buddhism across Asia.
Why does Madhyamaka hold that nothing truly arises?
Madhyamaka holds that nothing truly arises because none of the four logical possibilities for arising, from itself, from something other, from both, or from no cause, can be shown to hold up. Nāgārjuna concluded that the arising people experience is a mere appearance, comparable to the arising experienced in a dream.
All sources
38 references cited across the entry
- 1The Theory of Two Truths in TibetSonam Thakchoe — The Metaphysics Research Lab, Center for the Study of Language and Information, Stanford University — Summer 2022
- 2MadhyamakaOxford University Press — 25 January 2017
- 3Maritime BuddhismOxford University Press — 20 December 2018
- 4Tibetan Epistemology and Philosophy of LanguagePascale Hugon — The Metaphysics Research Lab, Center for the Study of Language and Information, Stanford University — Spring 2020
- 6NāgārjunaJan Christoph Westerhoff — 10 February 2010
- 8The Two Truths of Buddhism and The Emptiness of EmptinessSusan Kahn — 11 September 2014
- 9BookThe Essentials of Buddhist PhilosophyJunjirō Takakusu — Motilal Banarsidass — 1998
- 10BookWays of Thinking of Eastern Peoples: India, China, Tibet, JapanHajime Nakamura — Motilal Banarsidass — 1991
- 11BookMādhyamika Śūnyatā, a Reappraisal: A Reappraisal of Mādhyamika Philosophical Enterprise with Special Reference to Nāgārjuna and ChandrakīrtiG. C. Nayak — Indian Council of Philosophical Research — 2001
- 12BookThe Fundamental Wisdom of the Middle Way: Nāgārjuna's MūlamadhyamakakārikāNāgārjuna — Oxford University Press — 1995
- 13AbhidharmaNoa Ronkin — 16 August 2010
- 16BookGreek Buddha: Pyrrho's Encounter with Early Buddhism in Central AsiaChristopher I. Beckwith — Princeton University Press — 2015
- 17MadhyamakaRichard Hayes — 6 November 2010
- 18JournalA Translation of the Madhyamakahrdayakarika with the Tarkajvala III. 137-146Chikafumi Watanabe — 1998
- 20TsongkhapaGareth Sparham — 18 July 2011
- 21Gorampa go rams paConstance Kassor — 2 May 2011
- 22The Theory of Two Truths in TibetSonam Thakchoe — 17 February 2011
- 31JournalThe nature of the Madhyamika trickC. W. Jr. Huntington — 2007
- 33'Sarva-shunya' is impossible!V. Subrahmanian — 24 May 2020
- 34Concept of Mukti and ShankaracharyaGeorge Cronk — 4 January 1999
- 35ThesisOn the relationship of Advaita Vedānta and Mādhyamika BuddhismEric T. Reynolds — University of British Columbia — 1975
- 36BookStudies in Buddhist PhilosophyMark Siderits — Oxford University Press — 2016
- 37Part 3.12 - Vitaṇḍā or Cavil16 May 2025
- 38BookNyāyasūtra