Apoha
On this page
The Buddhist theory, originated by Dignāga and reworked by Dharmakīrti, that words and concepts pick out things not through real universals but by excluding what is other: 'cow' means, in effect, 'not non-cow'.
Also: anyāpoha, apoha theory, exclusion theory of meaning
The argument at a glance
A guide to the entry below, quoting its own words. Skip to the full entry and its sources.
- Question
- How can a general word such as 'cow' or 'existent' apply to many things, and be learned and used, if the only real things are unique particulars? Any answer must say what such a word denotes: the individuals, a universal, the relation between them, or the individual as possessing the universal. Hideyo Ogawa's study of the fifth chapter of Dignāga's Pramāṇasamuccaya reports that Dignāga considers and rejects exactly these four candidates (individuals, generic properties, relations, and individuals qualified by a generic property) before stating his own theory.
- Reasons , COT reconstruction: the numbering is ours, not the philosopher's
-
(COT's reconstruction, following Hayes and Ogawa; Dignāga does not number the steps)
- If a general word has a real object, that object is the particulars, a universal, the relation between universal and particular, or the particular as qualified by the universal.
- It is not the particulars: they are endless, so no word could express them all, and a word that expressed only some would deviate from the rest of its domain (V.2ab).
- It is not a universal or the inherence relation, since a general word is used co-referentially with words for individuals and is not heard apart from them (V.2cd), and no nominal denotes a relation as such (V.3).
- Dignāga also rejects the remaining option, the particular as qualified by the universal (Ogawa lists it among the four views refuted).
- Yet general words do convey something and guide action.
- Therefore: what they convey is not a real entity but a conceptual construction, the exclusion of what is other (anyāpoha). (the entry continues below)
- Intended conclusion
- General words and the concepts they express are about exclusions constructed by thought, not about real universals, so that language and conceptual thought can be explained without admitting universals among the things that are real.
Objections and replies
- Objections , Raised by other thinkers and critics, as reported
-
The best-known objection, circularity, is set out with Kumārila's text, Uddyotakara's earlier charge and the Buddhist replies in the separate structured entry Kumārila's circularity objection to apoha, which this entry links rather than repeats. A second line of objection is modern. Tuske reports Bob Hale's argument that, where the term is a name for a particular, 'x is not non-n' reduces to 'x is n', so that double negation collapses back into the particular; Tuske adds his own example of exclusion failing to pick out only the intended things, and gives it as his own view that these objections "are devastating for the top-down version of the apoha theory."
- Replies
-
Defenders reshaped the theory rather than abandon it. Tuske reports Tillemans's distinction between a 'top-down' version, associated with Dignāga and centred on the logic of negation, and a 'bottom-up' version, associated with Dharmakīrti and built on causal chains from particulars through perception and thought to language; Hale's objections bear most directly on the first. Tuske also reports a standard reply to the circularity charge, which distinguishes negation bound to a term, where the law of excluded middle fails and double negation need not return the original term, from sentential negation, where it does. A later Buddhist defence is Ratnakīrti's Apohasiddhi (Proof of Exclusion). McAllister's entry describes exclusion in his thought as "the Buddhist substitute for universals", explains that every awareness has a positive element (vidhi), its manifest form, and a negative element (pratiṣedha), its difference from other forms, and presents the Apohasiddhi as showing how concept-driven activity is possible without real similarities among things: apparent universals "only let us group particular things in such a way that our expectations are met." Śāntarakṣita's distinction between kinds of exclusion is set out in the linked circularity entry; Tuske notes that Hale's chapter counters a reply by Mark Siderits to this objection.
Coordinate 1.1.2Coordinate 6.10Coordinate 6.10.1DignāgaDharmakīrtiKumārila BhaṭṭaRatnakīrtiSources and primary textsRead the full entry
What it means
Kind of record. Apoha is a theory of meaning, with the arguments that founded and defended it, rather than a single argument; Dignāga's and Dharmakīrti's versions and their critics are set out below.
The question. How can a general word such as 'cow' or 'existent' apply to many things, and be learned and used, if the only real things are unique particulars? Any answer must say what such a word denotes: the individuals, a universal, the relation between them, or the individual as possessing the universal. Hideyo Ogawa's study of the fifth chapter of Dignāga's Pramāṇasamuccaya reports that Dignāga considers and rejects exactly these four candidates (individuals, generic properties, relations, and individuals qualified by a generic property) before stating his own theory.
Apoha ('exclusion') is the theory of concepts and word meaning developed by the Buddhist epistemologists. It allows them to explain general terms without admitting real universals. Tom Tillemans's Stanford Encyclopedia entry on Dharmakīrti explains that the school is nominalist. Only particulars (svalakṣaṇa), which have causal powers, are real. Universals (sāmānyalakṣaṇa) are conceptual fictions that exist only conventionally, needed for thought and language but not part of the furniture of the world.
Provenance. The theory and its founding argument originate with Dignāga, in the fifth chapter of the Pramāṇasamuccaya and his own commentary (vṛtti) on it; the abstract on the PhilPapers record of Pind's annotated translation (2015) describes it as the first comprehensive presentation of Dignāga's philosophy of language and presents the apoha thesis as a crucial complement to his theory of inference. Dharmakīrti reformulated the theory. Tillemans names Śāntarakṣita among the Indo-Tibetan thinkers who discussed the problem intensely, and Patrick McAllister's Stanford Encyclopedia entry places Ratnakīrti, whose teacher in philosophical matters was Jñānaśrīmitra, in the first half of the eleventh century. 'Buddhist nominalism' is a modern label, used for example in the title of the 2011 volume edited by Mark Siderits, Tom Tillemans and Arindam Chakrabarti.
Dignāga introduced apoha to explain how concepts can be about particulars. In the SEP's rendering, Dignāga wrote that "A word talks about entities only as they are qualified by the negation of other things." Tillemans describes these exclusions of what is other (anyāpoha) as a kind of double negation: 'blue' applies to what is non-non-blue. Dharmakīrti reworked the theory in terms of causal relations, which Tillemans calls his major contribution to it. Particular blue things cause perceptual images. These lead to the judgement 'this is blue', in which an apoha appears as a generic representation, and words relate to things through these causal chains. Dharmakīrti argued that things can be grouped together without a common real feature. Different plants that all reduce fever, for example, are grouped only by the negative fact that they differ from things that lack that power.
The text's own reasoning. Dignāga argues by elimination. Against the view that a word for a kind denotes the particulars, Pramāṇasamuccaya V.2ab states, in the Sanskrit as quoted by Richard Hayes, "na jātiśabdo bhedānām ānantyavyabhicāratah vācakaḥ": a kind-word does not express the particulars, because of their infinity (ānantya) and because of deviation (vyabhicāra). Hayes explains that ānantya rules out a word's expressing the totality of particulars to which it applies, and vyabhicāra rules out its picking out only part of its domain. Against universals and relations, Ogawa reports that at V.2cd a general term such as 'existent' (sat) does not denote a generic property or the inherence relation, because it is not 'heard apart' from words for individuals but is co-referential with them, as in 'a substance is existent' (sad dravyam); at V.3 Dignāga denies that a nominal denotes a relation through its own property. The abstract of Hayes's chapter summarises the upshot of Dignāga's nominalism: words that name complexes and words that name absolute simples alike "can in fact be naming only concepts", not things of the world as they might be independent of our experience.
Structure. (COT's reconstruction, following Hayes and Ogawa; Dignāga does not number the steps) (1) If a general word has a real object, that object is the particulars, a universal, the relation between universal and particular, or the particular as qualified by the universal. (2) It is not the particulars: they are endless, so no word could express them all, and a word that expressed only some would deviate from the rest of its domain (V.2ab). (3) It is not a universal or the inherence relation, since a general word is used co-referentially with words for individuals and is not heard apart from them (V.2cd), and no nominal denotes a relation as such (V.3). (4) Dignāga also rejects the remaining option, the particular as qualified by the universal (Ogawa lists it among the four views refuted). (5) Yet general words do convey something and guide action. Therefore (6) what they convey is not a real entity but a conceptual construction, the exclusion of what is other (anyāpoha). Joerg Tuske's review of the 2011 volume reports that Pind and Prabal Kumar Sen discuss the relationship between this top-down version of the theory and Dignāga's theory of inference.
Intended conclusion. General words and the concepts they express are about exclusions constructed by thought, not about real universals, so that language and conceptual thought can be explained without admitting universals among the things that are real.
Apoha drew strong objections. Kumārila Bhaṭṭa of the Mīmāṃsā school and the Naiyāyika Uddyotakara charged it with circularity: to understand 'non-non-blue' one must already understand 'blue'. Tillemans reports Dharmakīrti's tu quoque reply and also a modern argument that the apoha theorist seems to run counter to the compositionality of language. The debate between apoha theorists and the realist schools Nyāya, Vaiśeṣika and Mīmāṃsā is one of the most sophisticated pre-modern discussions of universals, concepts and meaning.
Objections. The best-known objection, circularity, is set out with Kumārila's text, Uddyotakara's earlier charge and the Buddhist replies in the separate structured entry Kumārila's circularity objection to apoha, which this entry links rather than repeats. A second line of objection is modern. Tuske reports Bob Hale's argument that, where the term is a name for a particular, 'x is not non-n' reduces to 'x is n', so that double negation collapses back into the particular; Tuske adds his own example of exclusion failing to pick out only the intended things, and gives it as his own view that these objections "are devastating for the top-down version of the apoha theory."
Replies. Defenders reshaped the theory rather than abandon it. Tuske reports Tillemans's distinction between a 'top-down' version, associated with Dignāga and centred on the logic of negation, and a 'bottom-up' version, associated with Dharmakīrti and built on causal chains from particulars through perception and thought to language; Hale's objections bear most directly on the first. Tuske also reports a standard reply to the circularity charge, which distinguishes negation bound to a term, where the law of excluded middle fails and double negation need not return the original term, from sentential negation, where it does. A later Buddhist defence is Ratnakīrti's Apohasiddhi (Proof of Exclusion). McAllister's entry describes exclusion in his thought as "the Buddhist substitute for universals", explains that every awareness has a positive element (vidhi), its manifest form, and a negative element (pratiṣedha), its difference from other forms, and presents the Apohasiddhi as showing how concept-driven activity is possible without real similarities among things: apparent universals "only let us group particular things in such a way that our expectations are met." Śāntarakṣita's distinction between kinds of exclusion is set out in the linked circularity entry; Tuske notes that Hale's chapter counters a reply by Mark Siderits to this objection.
Key thinkers and texts. Dignāga, Pramāṇasamuccaya V with his vṛtti (annotated translation by Pind, 2015); Dharmakīrti, Pramāṇavārttika; Kumārila Bhaṭṭa, Ślokavārttika, Apohavāda chapter (critical edition and translation by John Taber and Kei Kataoka, 2021); Uddyotakara, Nyāyavārttika; Śāntarakṣita, Tattvasaṅgraha, with Kamalaśīla's commentary; Ratnakīrti, Apohasiddhi; and, for modern assessment, Mark Siderits, Tom Tillemans and Arindam Chakrabarti (eds), Apoha: Buddhist Nominalism and Human Cognition (Columbia University Press, 2011).
Current status. Apoha is now discussed both as history and as a live option in the theory of concepts. Tuske quotes the volume's introduction: "The apoha theory is first and foremost an approach to the problem of universals". He reports that the top-down version has received more scholarly attention, and notes Jonardon Ganeri's 'feature-placing' account of concept formation in connection with the bottom-up version. Whether the theory can meet the charges of circularity and over-generation remains disputed; there is no consensus.
Further connections. Because the argument concerns what general terms refer to, it also bears on Coordinate 6.10.1 (Reference). The Mīmāṃsā theory of sentence meaning that accompanies the realist alternative is set out in Abhihitānvaya and anvitābhidhāna (theories of sentence meaning), and the grammarians' account of the meaning-bearer in Sphoṭa. Compare the early Chinese dispute over whether a white horse is a horse in the White Horse Dialogue; the comparison concerns a shared problem about general names, not influence.
On this site it belongs to Coordinates 1.1.2 (Universals and Particulars) and 6.10 (Philosophy of Language).
See also the glossary entries Nominalism, Universals, Vaiśeṣika, Mīmāṃsā, Sphoṭa and Semantics.
The questions it belongs to
- 1.1.2 Universals and ParticularsA tomato and a fire engine are both red. Is there something they share?
- 6.10 Philosophy of LanguageHow do noises come to be about things?
- 6.10.1 ReferenceWhat makes your word for a thing refer to that thing?
Thinkers
Traditions
To read
- Kṣaṇabhaṅgasiddhi (Proof of Momentary Cessation) — Ratnakīrti
- Pramāṇasamuccaya (Compendium of Valid Knowledge) — Dignāga
- Pramāṇavārttika (Commentary on Valid Knowledge) — Dharmakīrti
Sources and further reading
- Tom Tillemans, 'Dharmakīrti', Stanford Encyclopedia of Philosophy (substantive revision 7 April 2026)
- Stephen Phillips and Anand Vaidya, 'Epistemology in Classical Indian Philosophy', Stanford Encyclopedia of Philosophy (substantive revision 13 March 2024)
- Richard P. Hayes, 'Diṅnāga's nominalism', in Dignaga on the Interpretation of Signs, Studies of Classical India 9 (Dordrecht: Kluwer Academic Publishers, 1988), pp. 173–219, DOI 10.1007/978-94-009-2899-2_5 (publisher's chapter page, abstract and notes)
- Hideyo Ogawa, 'Dignāga on the View of a Generic Term as Denoting a Relation', in Birgit Kellner et al. (eds), Reverberations of Dharmakīrti's Philosophy: Proceedings of the Fifth International Dharmakīrti Conference, Heidelberg, August 26 to 30, 2014 (Vienna, 2020), pp. 313–330 (Austrian Academy of Sciences Press, austriaca.at PDF)
- PhilPapers record for Dignāga, Pramāṇasamuccayavṛtti chapter 5 (on apoha), translated and annotated by Pind (Wien: Verlag der Österreichischen Akademie der Wissenschaften, 2015); the record lists Dignāga as author and does not show the translator's given name
- Joerg Tuske, review of Mark Siderits, Tom Tillemans and Arindam Chakrabarti (eds), Apoha: Buddhist Nominalism and Human Cognition (Columbia University Press, 2011), Notre Dame Philosophical Reviews 2012.03.33
- Patrick McAllister, 'Ratnakīrti', Stanford Encyclopedia of Philosophy (first published 26 August 2026)
Cite this page
Responsibility: Coordinates of Thought Editors. Each statement was drafted from the sources listed above and checked against them in a separate pass before publishing; this page is not peer-reviewed by outside specialists. Source-checked 20 November 2026. How to cite this site · How we write and check.
APA (7th edition)
Coordinates of Thought Editors. (2026, November 20). Apoha (Arguments). Coordinates of Thought. Retrieved [date accessed], from https://coordinatesofthought.com/arguments/apoha/
MLA (9th edition)
“Apoha (Arguments).” Coordinates of Thought, 20 Nov. 2026, coordinatesofthought.com/arguments/apoha/. Accessed [date accessed].
Chicago (17th edition)
Coordinates of Thought Editors. “Apoha (Arguments).” Coordinates of Thought. Source-checked November 20, 2026. https://coordinatesofthought.com/arguments/apoha/.
BibTeX
@misc{cot-glossary-apoha-2026,
author = {{Coordinates of Thought Editors}},
title = {Apoha (Arguments)},
howpublished = {Coordinates of Thought},
year = {2026},
month = nov,
url = {https://coordinatesofthought.com/arguments/apoha/},
note = {Source-checked 20 November 2026}
}
RIS
TY - ELEC AU - Coordinates of Thought Editors TI - Apoha (Arguments) T2 - Coordinates of Thought PB - Coordinates of Thought PY - 2026 DA - 2026/11/20 UR - https://coordinatesofthought.com/arguments/apoha/ LA - en ER -
For reference managers (Zotero, Mendeley, EndNote): BibTeX file · RIS file