Causation and Emptiness in Early Madhyamaka

Bibliographic

Mark Siderits, “Causation and Emptiness in Early Madhyamaka,” Journal of Indian Philosophy 32:4 (August 2004), pp. 393–419. JSTOR 23496964.

Thesis

The Madhyamaka claim that whatever is causally dependent lacks intrinsic nature (svabhāva) has been read by Hayes (1994) as resting on an equivocation: Nāgārjuna allegedly slides from material-constitution dependence (where the inference goes through) to causal dependence in general (where it does not). Siderits’s contention is that there is a Madhyamaka argument for the broader claim once one reconstructs MMK 1 carefully. The connection between causal dependence and emptiness comes by way of the thesis that the causal relation itself is conceptually constructed. Once that is granted, anything originating in dependence on causes cannot be ultimately real. The argument thus has two stages: a Humean / three-times argument that causation is conceptually constructed (MMK 1.3–7), and a generalising principle (the article’s “Principle P”) that ties the conceptual construction of a relational tie to the conceptual construction of any property of its relata that essentially involves that tie.

Key claims

• Translation insistence: svabhāva is “intrinsic nature,” not “self-existence.” The “self-existence” rendering is the result of reading svabhāva through Madhyamaka uses alone without tracing it back to Abhidharma. 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 mistranslation underlies the suspicion that Madhyamaka equivocates (pp. 395–396)
• The svabhāva criterion of dharma-hood unifies the Abhidharma project. Across the Sautrāntika 75-dharma list, Theravāda 89, and the rest, the criterion that only that which has svabhāva is a dharma is the shared assumption (pp. 394–395). Madhyamaka does not introduce the criterion; it accepts it and pushes it through to a conclusion the Abhidharmika resists
• Defence of the svabhāva criterion as a realist commitment. Siderits argues the criterion should recommend itself to anyone who is a realist in the sense of holding that there is such a thing as how things are independently of the concepts we happen to employ. Vasubandhu (in his Yogācāra phase) and Berkeley both count as realists in this sense — mind-dependence of existence is a different question from mind-dependence of nature (pp. 396–397). Hence the criterion is not a Madhyamaka rhetorical trick but a structural commitment of any non-conceptualist realism
• Hayes’s equivocation charge identified precisely. Hayes 1994 reads Nāgārjuna at MMK 4.7 as illegitimately generalising from material-constitution causation (rūpa from the four elements; the inference goes through via reductive supervenience) to non-material causation (consciousness from sense-object contact; the inference is not licensed because the relation is not material constitution). Tillemans 2001 responds with later Madhyamaka materials; Siderits offers a complementary defence drawn from MMK 1 itself (pp. 398–401)
• The argument from the three times (MMK 1.5–7) reconstructed. If a set of conditions is the hetu-condition of an effect, there must be some time at which they produce the effect. Not when the effect already exists (production is then superfluous); not when the effect does not yet exist (the conditions then have not yet produced anything); not in a third time, since impartite ultimately-real entities cannot undergo extended production processes. Burton’s reply (cause and effect simultaneous) is rebutted by analysis of apparent reciprocal-causation cases: balance cups, threads-and-cloth (pp. 405–407)
• Bradley regress against causal force / kriyā (MMK 1.4). The asatkāryavādin’s natural move — invoke a kriyā to mediate between distinct conditions and effect — generates an infinite regress. The mediating causal force itself either exists among the conditions (back to satkāryavāda) or arises from them, in which case a further causal force is needed to explain its arising, and so on. Siderits notes that taking kriyā as a relation rather than a thing escapes the regress, but only at the cost of conceding that a new question (the three-times problem) now opens (pp. 401–404). Buddhapālita and Candrakīrti read MMK 1.4 as itself a three-times argument; Siderits explores the regress reading as a complementary possibility supported by the Akutobhayā (n. 25)
• Conclusion of MMK 1: the causal relation is conceptually constructed, not eliminated. Siderits is explicit (n. 35) that he reads MMK 1 as showing only that the causal relation is conceptually constructed, not that the relata are. This is a narrower conclusion than Hayes (1994) attributes to Nāgārjuna and aligns with Bhāviveka’s Hume-anticipating gloss at Prajñāpradīpa on MMK 1.3 (Pandeya 26): “parabhāva of the conditions is found due to intentness of the mind on the desire for what is productive of the arising of bhāva” (p. 408)
• Principle P (the article’s central new contribution). “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, on the Madhyamaka view, essentially involves reference to the causal relation in dependence on which the effect originated. The K2 / Sherlock Holmes counter-example (relational tie being sacred to is conceptually constructed but K2 itself is not) is met by the essential-reference qualifier. Worked example: the property of being a chariot essentially involves reference to having been manufactured, which essentially involves reference to conceptual resources (intentions, plans, recognition of goal-achievement); so the property of being a chariot is a conceptual construction (pp. 410–413)
• The combined argument blocks Hayes’s equivocation charge. MMK 15.1–2 in isolation does seem to slide between compounded₁ (made of parts) and compounded₂ (causally dependent in any sense). But MMK 1 supplies the missing step via the three-times argument and Principle P, so the slide is not an equivocation but an articulated two-stage argument (pp. 412–413)

Methodology

Rational-reconstructive analytic philosophy of Buddhism, with full primary-text engagement. Siderits works the Sanskrit of MMK and the four early commentaries (Akutobhayā, Buddhapālitavṛtti, Prajñāpradīpa, Prasannapadā) at the same level of authority as his analytic interlocutors (Hayes, Tillemans, Garfield, Taber). Engages the Abhidharma (Vasubandhu’s Abhidharmakośabhāṣya) as a standing technical context for the Madhyamaka argument. Frames the dispute throughout in the satkāryavāda / asatkāryavāda problematic that organises Indian causation debates (Sāṃkhya, Nyāya, Buddhist), so that the Madhyamaka move is shown as intervening in a live Indian debate, not as a stand-alone provocation. Hume’s bundle theory, Berkeley’s idealism, and Lewis-style possible-worlds semantics for nomological necessity appear as analytic foils. Notably absent: any engagement with Two Truths as pedagogical method, neyārtha / nītārtha, the Three Turnings, or tathāgatagarbha. The 2004 article works one level below the framework register that the 2007 textbook (siderits-buddhism-philosophy-2007 Ch 9) occasionally reaches for.

Notable quote

“It will be my contention that this situation has come about in part because of a lack of clarity concerning the notion of svabhāva and its roots in Abhidharma.” (p. 393)

Connections

• siderits-buddhism-philosophy-2007 — direct successor; the 2007 textbook (Ch 9) re-deploys the equivocation rebuttal in absorbed form (pp. 192–195, fn. 5 names Burton 1999 and Hayes 1994). The 2004 article is the technical foundation; 2007 is the systematic statement
• siderits-reality-altruism-2000 — earlier Williams review; the 2000 piece is framework-engaging at the graded-teaching level (BCA 8 / 9 + MMK 18:8); the 2004 article is not framework-engaging in that sense. The progression 2000 → 2004 → 2007 is not monotonic on framework-engagement
• burton-emptiness-appraised-1999 — principal target of the equivocation rebuttal; Burton’s regress-from-universal-prajñaptimātra is the broader form of the same charge Hayes makes more locally on MMK 1 / 15
• westerhoff-vigrahavyavartani-2010 and westerhoff-madhyamaka-2009 — Westerhoff’s threefold svabhāva analysis (essence-svabhāva / substance-svabhāva / absolute svabhāva) is structurally close to Siderits’s careful translation insistence here; both reject “self-existence” as a misleading rendering. The convergence is independent — Westerhoff via Candrakīrti’s Madhyamakāvatāra, Siderits via Abhidharma roots
• ames-prajnapradipa-ch2-1995 — Ames’s translation of Prajñāpradīpa Ch 2 covers Bhāviveka on motion; Siderits engages Bhāviveka on Ch 1 (causation), so the two pieces together cover the two foundational chapters
• coghlan-buddhapalita-2021 — Coghlan’s full Buddhapālitavṛtti translation makes Siderits’s primary-text references to BP at MMK 1.3 and 1.4 directly checkable
• Hayes 1994 (“Nāgārjuna’s Appeal,” Journal of Indian Philosophy) — principal target; not yet in the wiki
• Tillemans 2001 (“Trying to be Fair to Mādhyamika Buddhism,” Calgary) — complementary defence against Hayes; not yet in the wiki
• Garfield 1994 / 1995 — FWMW commentary on MMK; Siderits agrees with Garfield (1994) on the importance of Nāgārjuna beginning with causation but disagrees with Garfield’s positive-causal-account reading of MMK 1 (n. 18)
• Taber 1998 (“On Nāgārjuna’s So-called Fallacies”) — sympathetic earlier defence of Nāgārjuna; Siderits criticises Taber’s “principle of coexisting counterparts” reading of MMK 1.3cd as a “singularly bad argument” (p. 401)