Madhyamaka Wiki

Kalupahana vs. Buddhapālita and VV

14 min read

The claim

Kalupahana reads the Mūlamadhyamakakārikā as “a superb commentary on the Buddha’s own Kaccāyanagotta-sutta” (kalupahana-mmk-1986 p. 5). That thesis is exegetically vulnerable on two independent grounds, each anchored in primary texts. First, the earliest surviving commentary, Buddhapālita’s, reads the MMK as a Mahāyāna treatise: its colophon describes the MMK as “Great Vehicle Abhidharma” (TCE p. 387, attributed by Coghlan to Buddhapālita), and the commentary sets out the two truths at 24:8–10 (coghlan-buddhapalita-2021). Second, Nāgārjuna’s own Vigrahavyāvartanī (westerhoff-vigrahavyavartani-2010) engages Nyāya-style realist epistemology in a way the Kaccāyanagotta-sutta does not authorise. Both witnesses are prima facie incompatible with the Kaccāyanagotta-only reduction. The contradiction is therefore not confined to modern academic interpretation; it is a primary-text counter-witness against the deflationary line that runs through Kalupahana, Burton, and Sprung.

Evidence for

The Buddhapālita colophon witness. The colophon after MMK 27 in the Buddhapālitavṛtti says that the MMK “presents Great Vehicle Abhidharma, perfectly elucidates ultimate reality, and clarifies the system of the Perfection of Wisdom”, and that the commentary “teaches the definitive meaning of the Great Vehicle” (coghlan-buddhapalita-2021, TCE pp. 387–388). Coghlan (Introduction) attributes the description to Buddhapālita; the colophon speaks of “Venerable Master Buddhapālita” in the third person and names the translators, so it may not be his. It places the MMK within the Perfection of Wisdom literature that Kalupahana treats as a later imposition. Coghlan goes further and reads the MMK as staging debates between Upper Abhidharma (second and third turnings) and Lower Abhidharma (first turning), but he presents that as a Tibetan classification (“In Tibet, Abhidharma was classified in two”), and the commentary itself does not invoke the turnings. What the commentary does say in its own voice is enough for this page: the Dharma the buddhas teach rests on the two truths, which “must be set out” (TCE pp. 348–349). That cannot be squared with the Kaccāyanagotta reduction. Kalupahana, writing in 1986, did not have Coghlan’s translation; he treats Buddhapālita as a Tibetan-tradition reconstruction rather than as a recoverable Indian voice. Coghlan’s edition recovers exactly that voice.

The Buddhapālita structural witness. BP’s structural taxonomy of MMK (eight negations as the spine, with the chapters as several ways into the subject whose order is not fixed, TCE p. 36; the Upper/Lower Abhidharma dialogue as organising frame is Coghlan’s reading) is incompatible with Kalupahana’s four-part Kaccāyanagotta taxonomy (I–II causality; III–XV dharma-nairātmya; XVI–XXI pudgala-nairātmya; XXVI–XXVII conclusion). The two structural readings cannot both be right. The earliest extant commentator by a known author reads the text differently from Kalupahana — and reads it differently in a Mahāyāna-framework-presupposing direction.

The VP-Nyāya witness. Per westerhoff-vaidalyaprakarana-2018, the Vaidalyaprakaraṇa is a chapter-length, sūtra-by-sūtra critique of the sixteen Naiyāyika categories — pramāṇa, prameya, saṃśaya, prayojana, dṛṣṭānta, siddhānta, avayava, and so on. The text presupposes detailed knowledge of the Nyāyasūtra (the only reason the VP has no Tibetan commentary, per Westerhoff, is that the Nyāyasūtra itself was never translated into Tibetan). This is more engagement with non-Buddhist Indian philosophy than the VV — and at the level of an entire treatise structured around the opponent’s own taxonomy. Kalupahana’s claim that “no sophisticated Mahāyāna sūtras were available” to Nāgārjuna and that he therefore drew on the Nikāyas/Āgamas leaves the VP completely unexplained: the Kaccāyanagotta-sutta gives Nāgārjuna no resource for engaging the sixteen Naiyāyika categories. The VP is also third-attested across the Indian tradition: cited by Bhāviveka (Madhyamakaratnapradīpa), Candrakīrti (Madhyamakaśāstrastuti), and read by Tsong Khapa as the probans-refuting partner of MMK in the MMK–VP–VV triad. The VP is the strongest single piece of primary-text evidence that Nāgārjuna’s interlocutor extends well beyond Pāli śrāvaka opponents.

The VV-Nyāya witness. Per westerhoff-vigrahavyavartani-2010, the Vigrahavyāvartanī (especially vv. 5–6 and vv. 30–51) engages Nyāya theories of pramāṇa and Nyāya realist semantics directly and at length. The Kaccāyanagotta-sutta gives Nāgārjuna no resource for this engagement: the sutta is a brief Pāli discourse about avoiding “is” and “is not” extremes, not a treatise on epistemology. If Kalupahana is right that “no sophisticated Mahāyāna sūtras were available” to Nāgārjuna (kalupahana-mmk-1986 Preface pp. viii–ix), and he turned to the Nikāyas and Āgamas, then the Vigrahavyāvartanī’s sustained engagement with non-Buddhist Indian philosophy is unaccounted for. The VV’s existence and contents are prima facie evidence that Nāgārjuna’s interlocutors were broader than Kalupahana’s reduction allows.

The VV self-anchoring witness. Vigrahavyāvartanī v. 28 quotes Mūlamadhyamakakārikā 24:10 verbatim — “Not having had recourse to the conventional, the absolute is not taught” — as Nāgārjuna’s own answer to the Madhyamaka dilemma. This is Nāgārjuna self-quoting MMK 24’s Two Truths chapter in a different work to defend the very thesis of universal emptiness. Kalupahana’s flat reading of saṃvṛti and paramārtha as “equal-standing” pragmatic concepts cannot accommodate Nāgārjuna’s own use of the distinction as a defence-mechanism against self-referential objection. See vv-29-three-readings.

The Candrakīrti and Mabja MMK 15:7 witness (primary). Candrakīrti makes the point first, on the same verse: “This sutra is taught in all the Buddhist schools” (sprung-lucid-exposition-1979, p. 159), quoting the Samādhirāja there too. Per mabja-ornament-of-reason (Ch 15 commentary, “Refutation by Means of Scripture”), the Instructions to Kātyāyana is “a scripture that is accepted by all Buddhist schools.” Mabja’s is an early Tibetan primary-text articulation of the appeal-to-commonly-accepted-authority reading of MMK 15:7. The function of the citation, on Mabja’s twelfth-century Tibetan reading, is to demonstrate that the Transcendent Conqueror has also refuted the views of both existence and nonexistence in a discourse that is canonical across śrāvaka schools — a śrāvaka-acceptable scriptural anchor for a Mahāyāna treatise’s svabhāva critique. The reading that MMK is as a whole a commentary on the Kātyāyana-avavāda is structurally precluded by this framing. Mabja supplies, from the pre-Tsongkhapa Tibetan side, the same reading Westerhoff 2018 and Walser 2005 reconstruct externally on social-historical grounds.

The commentarial-consensus witness on MMK 15.7. Every extant commentary on MMK — beginning with Buddhapālita and continuing through Bhāviveka, Candrakīrti, the twelfth-century pre-sectarian Tibetan voice of Mabja, and the Tibetan tradition more broadly — reads Nāgārjuna’s invocation of the Kātyāyanāvavāda (Kaccāyanagotta-sutta) at MMK 15.7 as a citation of scriptural authority: a demonstration that the Madhyamaka critique of svabhāva is already vouched for in the Nikāya/Āgama corpus that the śrāvaka schools themselves accept. None reads the citation as evidence that MMK as a whole is a commentary on the sutta. The commentarial unanimity here is not framework-imposed Mahāyāna distortion; it is the structurally obvious reading of a single-verse appeal to authority within a 448-verse treatise. Kalupahana’s reduction inverts the function of the citation — treating an appeal-to-authority gesture as the hermeneutic key to the entire work — and no commentator, Indian or Tibetan, sees the text that way. This is a separate witness from the colophon: even bracketing its “Great Vehicle Abhidharma” description, the narrow commentarial reading of MMK 15.7 itself is unanimous against Kalupahana.

The Garfield MMK 15:7 witness (modern, framework-respecting). Per garfield-fundamental-wisdom-1995 (Ch XV), Garfield reads the very citation Kalupahana makes load-bearing — Nāgārjuna’s invocation of the Discourse to Kātyāyana at MMK 15:7 — in the opposite framework-direction. For Garfield the sutta is “one of the fundamental suttas of the Pali canon for Mahāyāna philosophy”: a Mahāyāna middle-way anchor whose lesson is that reification springs from “the failure to note impermanence” and nihilism from “the failure to note the empirical reality of arising phenomena,” the middle path being conventional existence. Here a modern, analytically-trained reader — not a tradition-internal commentator Kalupahana can discount as Mahāyāna back-projection — reaches the same narrow appeal-to-authority reading as Mabja and the Indian commentators, and treats the verse as a Mahāyāna anchor rather than as the hermeneutic key to the whole text. Two readers, one verse, opposite framework-verdicts. The disagreement therefore does not divide framework-respecting tradition from framework-neutral modernity; it runs within modern academic reading. This is the Garfield textual-range diagnostic, grounded on his full MMK commentary (garfield-fundamental-wisdom-1995).

The Walser strategic-citation witness (primary). Per walser-nagarjuna-2005 (Introduction pp. 9–15 and chapters not in the extracted PDF but summarised throughout), the absence of Mahāyāna sūtra citations in MMK is strategic minority-school positioning: Nāgārjuna writes as a Mahāyāna-minority voice within a non-Mahāyāna host monastery (most plausibly Mahāsāṅghika in the Lower Krishna Valley) whose cooperation he needs to secure the copying and preservation of his text. Restricting himself to authorities the host monastery already accepts is the institutional precondition for the śāstra falling within the monastery’s prior legal commitment to reproducing canonical material. Walser additionally argues that MMK is in conversation with three abhidharma collections at once — Sarvāstivāda attacked, Mahāsāṅghika allied with, Pudgalavāda/Saṃmitīya rehabilitated — and that the Ratnāvalī shows the opposite alliance pattern (direct refutation of Pudgalavāda, engagement with the kṣaṇavāda of the Pūrvaśaila / Aparaśaila). The shift in alliance pattern between MMK and Ratnāvalī is the strongest internal evidence Walser musters: Nāgārjuna’s targets and allies track the schools whose cooperation his immediate setting required. The Walser thesis cuts in a complicated way against Kalupahana. On the one hand it strongly supports the conclusion that the Tripiṭaka-only citation pattern in MMK is not evidence that MMK is Kaccāyanagotta-anchored Early Buddhism — it is evidence of strategic restriction. On the other hand it reframes Burton’s prajñaptisat-equation reading of MMK 24:18 as a strategic alliance with the Prajñaptivādins, with one narrow residual tension on MMK 24:18 prajñaptir upādāya (see framework-absence-yields-nihilism). See walser-nagarjuna-2005 the wiki author’s critical notes for the tension. Independent confirmation in westerhoff-golden-age-madhyamaka-2018 pp. 105–107, which adopts the strategic-citation thesis explicitly, names Warder 1973 and Kalupahana 1991 (pp. 5–8) on the standalone-MMK side, and explicitly names Ruegg, Lindtner, and Bronkhorst (2009 p. 136) as “remain unconvinced.” Walser additionally notes that the namaskāra of MMK is itself derived from a passage in the Perfection of Wisdom in 25,000 verses (per Conze 1975 n. 11), a fact Kalupahana does not engage.

The Śūnyatāsaptati witness, read by its root. Per komito-seventy-stanzas-1987 and the Śūnyatāsaptati text page, the root of v. 44 says that the Buddha speaks of “existence”, “non-existence” and both through his intention, and that this is hard to understand. In Komito’s translation the words that make this intention the two truths, and the glosses “conventional existence” / “non-inherent existence”, are interpolations from the commentaries, set in roman (pp. 156, 182). The opponents’ names at vv. 15 (Vaibhāṣika) and 18 (Hīnayānist) and the horns-of-a-rabbit image are likewise interpolated; the root’s objections — no distinctions, no causal production, no arising or cessation if things lack intrinsic nature — are standard Abhidharma-side objections, but the text does not name who raises them. What the Śūnyatāsaptati contributes to this page, then, is a further Yukti-corpus text whose concerns (arising, enduring and disintegrating; action and agent; the Buddha’s intentional speech) go beyond anything the Kaccāyanagotta-sutta supplies. It is not a two-truths declaration in Nāgārjuna’s own words, and a hermeneutic of intended meaning has a Pāli counterpart in the nītattha / neyyattha pair, so on its own it does not tell against a Tripiṭaka-only reading.

The VV authenticity witness. Westerhoff’s defence of VV’s authenticity against Tola & Dragonetti 1998 (westerhoff-vigrahavyavartani-2010 Introduction) is independently sound and mainstream. Kalupahana cannot escape the contradiction by demoting VV to the apocrypha.

The Cūḷahatthipadopama symmetry argument (minor supplement). Per tenpa-personal-notes-2025 under “My critique of Kalupahana”: the move “MMK’s only named citation is the Kātyāyana-avavāda, therefore MMK is a commentary on that sutta” is structurally parallel to a putative argument that the entire Mahāyāna pāramitā literature is a commentary on the Cūḷahatthipadopama Sutta (MN 27), which can be read as implicitly walking the practitioner through the six perfections in sequence — letting go of the worldly life (dāna), discipline (śīla), restraint from covetousness and displeasure (kṣānti), bliss in virtue / joyful effort (vīrya), samādhi (dhyāna), and the mind extended toward wisdom (prajñā). The pāramitās are also active in the Theravāda tradition as the pāramīs. This is a minor supplement — its purpose is to de-value the only-named-citation move by symmetry, not to replace the principal rebuttals (Candrakīrti’s “taught in all the Buddhist schools” and Mabja’s “accepted by all Buddhist schools” gloss, Garfield’s textual-range diagnostic, the BP colophon’s “Great Vehicle Abhidharma” description, the VV/VP Nyāya engagement, the ŚS v. 44 hermeneutic of intention). The argument lands because the Pāli canon does contain analogues of the principal Mahāyāna doctrinal items (the pāramī tradition, the avyākata on Vacchagotta), so reduction-by-single-citation is methodologically symmetric across the two traditions.

The Tsongkhapa six-treatise unified-system witness. tsongkhapa-ocean-of-reasoning-1408 Preliminary supplies the systematic Tibetan-tradition reading of Nāgārjuna’s six treatises as a single coordinated project with determinate divisions of labour: MMK refutes opponents’ theses on essence of persons and phenomena; Vaidalyaprakaraṇa refutes the Nyāya probans (the sixteen categories including pramāṇa) and is “the probans-refuting partner of MMK”; Vigrahavyāvartanī defends action-and-agent within essencelessness and is “a supplement to the first chapter of Mūlamadhyamakakārikā”; Śūnyatāsaptati unpacks “conventional existence” as merely nominally designated and is “a supplement to [chapter seven]”; Yuktiṣaṣṭikā and Ratnāvalī articulate the soteriological function of the middle path. “Among all these treatises the Mūlamadhyamakakārikā is the supreme, as it convincingly and extensively establishes the profound meaning with diverse arguments.” This is the Tibetan-tradition anchor for the six-treatise reading that Westerhoff reaches independently from Oxford-analytic Madhyamaka scholarship in 2010–18. Cross-period cross-tradition convergence: a 1407–08 Geluk-systematic reading and a 2010–18 Oxford-analytic reading independently arrive at the same division of labour. Kalupahana’s MMK-as-Kaccāyanagotta-commentary reading cannot accommodate the VP, the VV, the ŚS, or this systematic Tibetan account of how the six texts hang together. The witness therefore comprises four strata: the sixth-century Buddhapālita commentary’s Mahāyāna reading (its two-truths passages, and the colophon’s “Great Vehicle Abhidharma”), Nāgārjuna’s own VV/VP/ŚS, Tsongkhapa’s 1407–08 six-treatise system, and Westerhoff’s 2010–18 endorsement.

Evidence against

  • The MMK 15.7 citation of the Kātyāyanāvavāda is genuine and load-bearing. Nāgārjuna does invoke the sutta. The question is one of use vs reduction: does invoking a sutta as one anchor amount to writing a 448-verse commentary on it? Kalupahana could reply that the colophon’s Mahāyāna-Abhidharma description (whose authorship is open in any case) is itself a third-century retroactive Mahāyāna framing, and that the original second-century-CE Nāgārjuna was closer to Kalupahana’s reading than to Buddhapālita’s.
  • Buddhapālita is not Nāgārjuna. The earliest extant commentator by a known author (the anonymous Akutobhayā appears to be earlier; Ames 1995, p. 296) is still a commentator. Kalupahana’s whole strategy is to bracket the commentarial tradition; pointing to the earliest commentator reading MMK as Mahāyāna-Abhidharma is exactly the kind of evidence Kalupahana programmatically discounts. The argument therefore convinces only readers who accept some commentarial weight in the first place.
  • The VV is not the MMK. Even granting VV’s authenticity, the engagement with Nyāya could in principle be a later development of Nāgārjuna’s thought, with the MMK still primarily anchored in the Kaccāyanagotta-sutta. This reply is weak (the texts are widely held to be roughly contemporary and conceptually continuous, with VV v. 28 quoting MMK 24:10 verbatim) but not formally impossible.
  • Kalupahana himself reads the no-thesis line differently. Per kalupahana-mmk-1986 pp. 92–93, Kalupahana glosses VV v. 29’s pratijñā as “commitment” rather than “proposition,” neutralising the verse. Under this gloss the VV is less anti-realist than Westerhoff makes it. The reply collapses if Westerhoff’s translation is correct, but Kalupahana would resist Westerhoff on principle.
  • No Madhyamaka school holds the Kaccāyanagotta-only reading either. This is both a strength and a weakness of the rebuttal: the Indian and Tibetan commentarial tradition is unanimous against Kalupahana, which is evidence; but it is also the kind of unanimity Kalupahana programmatically reads as the very Mahāyāna distortion he is trying to undo. The argument therefore lands more decisively on framework-friendly readers than on framework-suspicious ones.

Linked pages

Graph View

Backlinks