Definition

Svabhāva — intrinsic nature (རང་བཞིན་), own-being, self-nature — is what Madhyamaka denies of all phenomena. In the Madhyamaka context the word means independent, self-sustaining, non-dependent existence: existence that does not borrow its nature from anything else. To deny svabhāva (niḥsvabhāva) is to assert emptiness (śūnyatā); the two are equivalent.

The term carries different senses in different Buddhist contexts. In Abhidharma, individual dharmas possess svabhāva, their own defining characteristic. The Mādhyamika then turns the Ābhidharmika’s own reductive logic outward: if composite entities lack svabhāva because they borrow their nature from their parts, then the parts themselves lack it too, since they in turn depend on further conditions.

Interpretations

Westerhoff’s threefold analysis. Westerhoff in westerhoff-madhyamaka-2009 distinguishes three senses of svabhāva drawn from Candrakīrti. (1) Essence-svabhāva is the essential property of an object, such as the heat of fire; it is accepted conventionally and is not the target of Madhyamaka negation. (2) Substance-svabhāva is independent, unconstructed, foundational existence (dravyasat), the primary target of every Madhyamaka argument. (3) Absolute svabhāva is the true nature of phenomena — emptiness itself — characterised as uncaused, unchanging, and non-dependent. Westerhoff argues these reduce to two: absolute svabhāva is essence-svabhāva applied universally, since emptiness is the property no object can lose without ceasing to be that object. The move dissolves an apparent contradiction, namely that emptiness seems to share the attributes of the very substance-svabhāva it negates. The “non-dependence” and “immutability” of emptiness are in fact weaker: non-dependence on specific objects, not on anything whatsoever. Tsongkhapa’s solution is also noted, adding “established from its own side” and “innate rather than acquired misconception” as further criteria that separate the object of negation from emptiness (pp. 40–46).

Cognitive dimension. Westerhoff stresses that svabhāva is not merely a theoretical-ontological concept but a cognitive default: the mind automatically superimposes independent existence onto phenomena. This is why Madhyamaka metaphysics, unlike its Western counterpart, is not a purely theoretical enterprise but one that requires meditative practice. His analogy is geometric. Proving theorems about four-dimensional space is one thing; developing a spatial intuition for the fourth dimension is quite another. In the same way, intellectually grasping the absence of svabhāva differs from actually ceasing to project it (pp. 13, 46–51).

The Williams-Burton argument. Burton in burton-emptiness-appraised-1999 argues that Nāgārjuna employs the Abhidharma notion of svabhāva, which means not merely “independent existence” but “unanalysable, more-than-conceptually-constructed existence” (dravyasat). In Abhidharma, dharmas possess svabhāva yet still dependently originate; the two are compatible. On Burton’s reading, then, Nāgārjuna’s denial of svabhāva means not just that entities dependently originate but that they have wholly conceptually constructed existence (prajñaptisat). Burton rejects the “terminological disagreement” reading, which holds that Nāgārjuna merely redefined svabhāva. He finds it implausible that a second-century Indian Buddhist would innovate with a key Abhidharma term “without notification.” His textual evidence includes RV I, 71 and 81 (dependence on parts), AS 6 and 44 (saṃvṛti/saṃvṛta = all entities), MMK XXIV, 18 (prajñaptir upādāya), and Nāgārjuna’s frequent use of kalpanā, vikalpa, and nāmamātra. The nihilistic consequence Burton draws is this: conceptual construction requires unconstructed material and an unconstructed constructor, and without these, nothing can exist. Westerhoff in westerhoff-nihilist-interpretation-2016 replies that dependence-structures can be circular or infinitely descending; they need not be hierarchically grounded in independent substances (pp. 356–357).

Eliminativism and non-foundationalism. Westerhoff argues that denying svabhāva commits the Mādhyamika to non-foundationalism: no ontological foundation exists. Combined with the Abhidharma heritage of eliminativism about the dependent, this yields what he calls a consistent nihilism, though not any of the forms of nihilism Madhyamaka rejects.

Siderits’s Abhidharma-rooted defence. Siderits in siderits-causation-emptiness-2004 (pp. 395–397) insists that svabhāva be rendered “intrinsic nature,” not “self-existence.” The latter, he argues, is a back-reading from Madhyamaka uses that never traces the term to its Abhidharma origin. The Abhidharmic core idea is that the properties of an ultimately real thing cannot be borrowed from other things but must be intrinsic to it. The chariot example shows how this works: the chariot’s capacity to carry is fully explicable through the properties of its parts, so the chariot is not ultimately real. Iterate the analysis, and the wheel turns out to be a whole made of parts as well; the regress ends only in things whose natures are genuinely their own. These are the dharmas, and only they are ultimately real. The svabhāva criterion of dharma-hood is what unifies the Abhidharma project across its differing dharma-lists (Sautrāntika 75, Theravāda 89, and so on). Siderits’s key point is that this criterion is no Madhyamaka rhetorical trick. It is a structural commitment of any non-conceptualist realism, something Vasubandhu (in his Yogācāra phase) and Berkeley share with Descartes and the Sautrāntikas, against subjective-idealist denials of mind-independent natures. The “self-existence” mistranslation feeds the suspicion that Madhyamaka equivocates; read correctly, the term ties Madhyamaka’s argument to a target the Abhidharmikas are themselves committed to. The reading converges with Westerhoff’s threefold scheme (essence-, substance-, and absolute-svabhāva), but Siderits reaches it independently, by the Abhidharma route.

Siderits on causation and emptiness. Siderits in siderits-causation-emptiness-2004 reconstructs the link between pratītyasamutpāda and niḥsvabhāva as a two-stage argument. Stage 1 (MMK 1): the causal relation itself is conceptually constructed. This is established by the three-times argument (MMK 1.5–7: production cannot occur when the effect already exists, does not yet exist, or in some impossible third time, since impartite ultimately-real entities cannot undergo extended processes of production), together with a Bradley regress against the asatkāryavādin’s appeal to a mediating causal force, or kriyā (MMK 1.4). Stage 2 (Principle P): “If a relational tie is conceptually constructed, then any property of one of its relata that involves essential reference to that tie must likewise be conceptually constructed.” The intrinsic nature of an effect essentially refers to the causal relation it originated in dependence on. So the intrinsic nature of anything causally dependent must itself be conceptually constructed, and therefore cannot be the intrinsic nature of an ultimately real entity. This is the missing premise that blocks Hayes 1994’s equivocation charge against MMK 15. Taken alone, MMK 15.1–2 does seem to slide between compounded₁ (made of parts) and compounded₂ (causally dependent in any sense), but MMK 1 supplies the wider argument through the conceptual construction of the causal relation. The 2007 textbook (siderits-buddhism-philosophy-2007 pp. 192–195, fn. 5) absorbs the argument; the 2004 article is its technical foundation.

Object of negation (dgag bya). In Tibetan Madhyamaka, identifying precisely which svabhāva one is negating is the crux of the whole interpretive enterprise. In tsongkhapa-illuminating-intent-1418 (Ch 9), Tsongkhapa gives six synonyms for the object of negation: “true existence,” “ultimate existence,” “absolute existence,” “existence by virtue of essential nature,” “existence through intrinsic characteristic,” and “intrinsic existence.” The Prāsaṅgika negates all six; the Svātantrika accepts the last three conventionally. Tsongkhapa holds that identifying the innate grasping at true existence — not merely the intellectual grasping bred by tenets — is “most essential.” Without this, he warns, “one might engage in eliminating the object of negation through reasoning, but this would not undermine whatsoever the clinging to true existence that persists since beginningless time.” Two senses of “ultimate” must be kept apart: (a) the “ultimate” of rational cognition, by which things are established; and (b) existence through a thing’s own objective mode of being, by which things are not established.

Gendün Chöpel faults the Geluk approach for making Madhyamaka too bland: the negation should apply directly to pots and pillars, not only to their “inherent existence.”

Tsongkhapa’s two senses of svabhāva (Ch 19). Expounding the sixteen emptinesses at MA 6.179–223, Tsongkhapa returns to MMK XV.2 (“intrinsic nature is unfabricated and not dependent on something else”) and distinguishes two senses of svabhāva. The first is something intrinsic in things, an essence by which they obtain existence and identity; this he categorically rejects. The second is the absence of intrinsic existence — emptiness — which “in an objective sense is the only true nature things have,” and which must be accepted, “for only through its knowledge can true release from grasping be attained.” Critics fix on exactly this second move: Gorampa, Mipham, the Ninth Karmapa, and Shakya Chokden all read it as a residual essentialism that the analysis has failed to deconstruct. Tsongkhapa’s defence is that svabhāva-as-emptiness is itself a non-implicative negation, and itself empty (the emptiness of emptiness): “even emptiness, which is the ultimate truth, is a conventional existent, not an ultimate existent.” So the concession does not relapse into pre-Madhyamaka realism; it supplies the very basis for the realisation that undoes innate grasping. (See tsongkhapa-illuminating-intent-1418, Ch 19; Jinpa’s reconstruction at the introduction’s fourth element draws the conceptual distinction between existence and intrinsic existence, the latter rejected even at the conventional level.)

Several interpreters press for a wider object of negation than Tsongkhapa allows, though they reach the demand by different routes.

Gorampa’s wider object of negation. In gorampa-distinguishing-views-1469, Gorampa argues that the Madhyamaka negation must apply to all four extremes of the catuṣkoṭi — existence, non-existence, both, neither — without qualification. The object of negation is therefore not simply “true existence” (bden grub), as Tsongkhapa holds, but the entire range of conceptual proliferation about the nature of phenomena. By confining the negation to “true existence” alone, Gorampa charges, Tsongkhapa renders three of the four koṭis pointless and reduces the Madhyamaka critique to what he calls a “scholastic epiphenomenon.”

The Ninth Karmapa’s wider object of negation (Kagyü). In karmapa-feast-fortunate-1578, the Ninth Karmapa argues that phenomena themselves must be refuted, not just their “true existence.” Tsongkhapa’s formula — “the vase is not empty of the vase; the vase is empty of true existence” — yields a “partial emptiness” (nyi tshe ba’i stong pa nyid) that misses what sentient beings actually grasp: they cling to “the vase,” not to “the true existence of the vase.” The Karmapa’s reasoning differs from Gorampa’s. Where Gorampa argues from the catuṣkoṭi, the Karmapa argues from the phenomenology of clinging: the wider object of negation is needed because it matches the confusion beings really have, a reading corroborated by Gendün Chöpel’s critique. Even so, this does not collapse into nihilism. At the stage of “no analysis,” worldly conventions — vases and pillars included — are accepted as they stand; the refutation applies only under analysis.

Atiśa’s “mere appearance” without a real basis. For Atiśa, both objects and cognitions are imputations of mere appearance (snang ba tsam), dependent designations without any real basis (avastuka). Unlike Śāntarakṣita and Kamalaśīla, who grant conventional reality a real basis in mental elements, Atiśa draws no hierarchical distinction between the ontological status of mind and objects, at either the conventional or the ultimate level. Both are unestablished dependent designations. The object of negation is a “conceived object” (zhen yul), produced by conceptualisation that imputes things as existing with their own-character (rang gi mtshan nyid). Atiśa separates objects negated by an antidote (gnyen po’i dgag bya), while one is practising the path, from objects negated by reasoning (rigs pa’i dgag bya), when one analyses inherent existence. All forms of reasoning are non-affirming negations (prasajyapratiṣedha) that dissolve themselves (see apple-jewels-middle-way-2018).

Kalupahana’s narrow historical target. In kalupahana-mmk-1986, Kalupahana argues that Nāgārjuna’s denial of svabhāva is aimed specifically at the Sarvāstivāda conception of substance: the theory that dharmas possess a “self-nature” persisting through all three periods of time. This is not a universal ontological negation but a targeted historical correction. The early Abhidharma theory of the four conditions (pratyaya) was sound; what went wrong was the Sarvāstivāda reading of those conditions in terms of underlying substance. And Nāgārjuna, on this account, rejects svabhāva on empirical grounds — “the self-nature of existents is not found [na vidyate]” (I.3) — rather than by dialectical reasoning. The definition of svabhāva as “unmade” (akṛtaka) and “non-contingent upon another” (nirapekṣaḥ paratra ca) at XV.2 is Nāgārjuna exposing an internal contradiction: a “caused substance” is a contradiction in terms. But this is not a universal critique of every notion of inherent nature; it targets the substantialist reading of the Abhidharma categories. The paired concept parabhāva (other-nature) stands for the Sautrāntika position, and both are rejected together.

Karunadasa: Theravāda sabhāva as the absence of substantial own-being. In karunadasa-theravada-abhidhamma-2010 (Ch 1, “The Definition of Dhamma as Own-Nature,” pp. 36–40; Introduction p. 9), Karunadasa documents that the Theravāda commentaries do define each dhamma as “own-nature” (sabhāva), and very probably took the usage from the Sarvāstivāda, where svabhāva served as a synonym for dharma before the Pāli commentaries were compiled. Yet the Theravāda exegetes hedge the term so heavily that it comes to mean “the very absence of sabhāva in any sense that implies a substantial mode of being.” Three moves do the work. (1) The agency-definition “dhammas bear their own-nature” (attano sabhāvaṃ dhārenti) is only kattu-sādhana, provisional and “not valid in an ultimate sense”; it is supplemented by the object-denotation “dhammas are borne by their conditions” (paccayehi dhārīyanti), which reverses the substantialising implication. (2) “sabhāva does not mean ‘own-sway’ (vasavattitā)”: a dhamma dependent on impermanent conditions “has no power to exercise any kind of sway.” (3) A sub-commentary states flatly that “the meaning of sabhāva is… the same as the meaning of ‘emptiness’” (sabhāvattho nāma suññattho), since sabhāva signifies “the mere fact of being a dhamma.” This bears directly on the negation debate. It cuts against the Burton/Siderits premise that holding the Abhidharma svabhāva-criterion commits one to dravyasat unconstructed existents. At least one Abhidharma tradition — the Theravāda, on its own commentarial reading — used sabhāva while expressly denying own-sway, denying any “behind-the-scenes substance,” and equating own-nature with emptiness. A caveat is in order: Karunadasa is an apologist defending the Theravāda against the reification charge. His is the commentarial-tradition reading, and his own treatment of Sarvāstivāda tritemporality, where svabhāva/dravya does persist across the three times, shows that the substantialising pressure was real within the wider Abhidharma.

Textual loci

  • MMK 15:1-11 — examination of svabhāva; it cannot arise from causes and conditions
  • MMK 24:8-10 — the Two Truths as necessary for understanding the denial of svabhāva
  • Vigrahavyāvartanī vv. 1–4, 21–29 (see westerhoff-vigrahavyavartani-2010) — Nāgārjuna’s defence against the self-referential objection: insubstantial (niḥsvabhāva) things can still fulfil functions; “the dependent existence of things is said to be emptiness, for what is dependently existent is lacking substance” (v. 22). This is the cleanest single-text refutation of the “empty = non-existent” misreading. Westerhoff renders svabhāva throughout as “substance” (an entity not depending on anything else). The rendering captures the substance-svabhāva of the threefold analysis, but it flattens the cognitive-projection resonance that the Tibetan rang bzhin preserves
  • Vigrahavyāvartanī vv. 11–12, 61–64 — Nāgārjuna on negation: negating svabhāva does not make an existent thing non-existent. It removes a mistaken superimposition (samāropa) of substance onto a world that lacks it. Negation as the ceasing of projection, not as the abolition of an entity
  • Bodhicaryāvatāra 9.28, 9.139 (commentary) (see kunzang-pelden-nectar-manjushri-2007) — the two senses of dngos med (non-entity / “absence of true existence”): “dngos med… is used in two different senses. On the one hand, it refers to what has no existence at all, even conventionally. On the other hand, it denotes things that are untrue in the sense of being like mirages. This is how Mādhyamikas reply to those who believe in true existence, who, through not understanding that things may very well appear without truly existing, think that the ‘absence of true existence’ means utter nothingness.” This is a traditional anticipation of the abhāva / niḥsvabhāva equivocation: the negandum is the substance-reading of phenomena, not their appearance. At BCA 9.139 the commentary adds that the “non-existence of the pot” reached merely by clearing away its existence-aspect “is a lesser, approximate form of emptiness,” because “things are deceptive or unreal in themselves, [so] their nonexistence is also unreal” — the object of negation is not swapped for a reified absence (cf. grasping-emptiness-is-itself-a-form-of-nihilism)