Reference · Snp 4.4
Purity
Suddhaṭṭhakasutta
Working draft. Last revised April 23, 2026.
Identity
Segment range snp4.4:1.1–8.4. Sn 788–795 (8 verses). Received title Suddhaṭṭhakasutta — "the eight[-verses] about the pure." The third of the four titled -aṭṭhakasuttas (with 4.2, 4.3, 4.5). Where 4.3 conducted a general diṭṭhi-polemic, 4.4 narrows to a specific target: the doctrine that purity (suddhi) comes aññato ("from another") — through the seen, the heard, the precept-and-vow, or the sensed. The commentarial traditions, both Pāli and Chinese, identify this target with the Brahmanical darśan doctrine — the claim that the sight of a holy figure confers spiritual merit.
Text and form
Triṣṭubh throughout (Norman 2001 p. 333) — the third consecutive Triṣṭubh sutta. No narrative frame in the Pāli; no speaker-tags. But the sutta is the AV's first clearly dialogical composition: v. 1 opens with what is most naturally read as a quoted Brahmanical claim ("I see purity, supreme, disease-free; through what is seen a person attains purity"), followed by the AV's diagnostic gloss ("so thinking, knowing it as supreme, 'I am a seer-of-purity' — he falls back on knowledge"). Fronsdal and Lee mark lines 1a–b as quoted speech; Norman, Bodhi, Sujato, and Thanissaro render the verse as a single utterance without quotation marks. Both commentarial framings (Pj II and the Yizujing) make the dialogical structure explicit.
Architecture. Claim → analytic refutation → counter-thesis → diagnostic phenomenology → goal-portrait:
- v. 1 quotes the heretical position and diagnoses it as "falling back on knowledge";
- v. 2 supplies the sopadhīka-reductio: if the seen purified, one with attachments would already be pure — which is absurd; the view's own logic refutes itself;
- v. 3 states the brahmin's negative thesis — no purity from another, by any of the four claim-bases;
- vv. 4–5 give the phenomenology — religious shoppers taking up and letting go, like monkeys grabbing branches; going high and low among teachers;
- vv. 6–8 give the goal-portrait — visenibhūta ("unaffiliated"), vivaṭa ("open"), sīmātigo brāhmaṇo ("the brahmin gone beyond the boundary"), na rāgarāgī na virāgaratto ("neither in love with passion nor besotted by dispassion").
The composition is more sustainedly dialectical than 4.3. Where 4.3 argues positionally (the muni cannot be disputed with), 4.4 argues analytically (a specific claim, a specific refutation, a specific counter-thesis).
Content
Passāmi suddhaṁ paramaṁ arogaṁ, / Diṭṭhena saṁsuddhi narassa hoti: "I see purity, supreme, disease-free; through what is seen a person attains purity." So thinking, recognising it as supreme, "I am a seer-of-purity" — he falls back on knowledge. But if purity came through the seen, or if one abandoned suffering through knowledge, then one still with the bases of defilement would be purified by something other — his view itself proclaims him.
The brahmin does not say purity comes from another — not by belief, tradition, precept-and-vow, or rationale; unsmeared by merit and evil, a discarder-of-the-grasped, fashioning nothing new. Having abandoned the previous, dependent on the next, following turbulence, they do not cross the bond: they take up and let go, like a monkey grabbing a branch as it releases another. Undertaking vows on his own, a creature goes high and low, attached to perception. But the wise one, having understood the truth through the vedas, does not go high and low — vast in wisdom.
He, unaffiliated in regard to all things — whatever belief, tradition, or rationale — being a seer-of-it, going about openly: by what here in the world would one categorise him? They neither construct nor foreground; they do not say "this is the ultimate purity." Releasing the tied-up knot of grasping, they make no desire for anything in the world.
The brahmin gone beyond the boundary — for him there is nothing tightly grasped, on knowing or on seeing. Neither in love with passion nor besotted by dispassion; for him here there is nothing taken up as ultimate.
Key passages
v. 1 (Sn 788) — the quoted claim.
Passāmi suddhaṁ paramaṁ arogaṁ, / Diṭṭhena saṁsuddhi narassa hoti; / Evābhijānaṁ paramanti ñatvā, / Suddhānupassīti pacceti ñāṇaṁ.
"I see purity, the supreme, disease-free; through what is seen a person attains purity." Knowing this directly, recognising it as supreme, "I am a seer-of-purity" — he falls back on knowledge.
The verse's most striking feature is the functional equivalence of diṭṭha (past participle of √dṛś, "what is seen") and diṭṭhi ("view"). Norman (p. 334): "diṭṭha as a past participle in the sense of an action noun = diṭṭhi." Thanissaro (note 1): "an ancient Indian belief, dating back to the Vedas, was that the sight of certain things or beings was believed to purify... This belief survives today in the practice of darshan." Both commentarial traditions (Pj II Candābha, Yizujing Mojie) read the verse as a darśan-critique.
v. 3 (Sn 790) — the brahmin's thesis.
Na brāhmaṇo aññato suddhimāha, / Diṭṭhe sute sīlavate mute vā; / Puññe ca pāpe ca anūpalitto, / Attañjaho nayidha pakubbamāno.
The brahmin does not say purity comes from another — not by belief, tradition, precept-and-vow, or rationale. Unsmeared by merit and evil, a discarder-of-the-grasped, fashioning nothing new here.
Pāda b — Diṭṭhe sute sīlavate mute vā — is verbatim with Snp 4.5:2.2 and Snp 4.12:10.1: the AV's most stable three-sutta formulaic unit. The brāhmaṇa is the AV's re-appropriated goal-figure; the Niddesa (mnd4:17.3–4) etymologises via bāhita, "one who has expelled [seven things]" — sakkāyadiṭṭhi, vicikicchā, sīlabbataparāmāsa, rāga, dosa, moha, māna. The same re-appropriation recurs across 4.5, 4.12, 4.13 and the Pārāyaṇa.
v. 4 (Sn 791) — the monkey.
Purimaṁ pahāya aparaṁ sitāse, / Ejānugā te na taranti saṅgaṁ; / Te uggahāyanti nirassajanti, / Kapīva sākhaṁ pamuñcaṁ gahāyaṁ.
Having abandoned the previous, dependent on the next, following turbulence, they do not cross the bond. They take up and they let go, like a monkey grabbing a branch as it releases another.
The AV's most famous simile, and unique to this sutta in the Pāli canon. The Niddesa elaborates with the image of "the monkey roaming in the wilderness" grabbing and releasing tree-branches as the various ascetics grab and release view-positions. Te uggahāyanti nirassajanti uses the verbal form of the attā-niratta root-family (cf. 4.3:8.3 Attā nirattā etc.) — the monkey image concretises the formula. Sitāse preserves the Vedic archaic -āse plural (cf. 4.2's cutāse, avītataṇhāse).
v. 6 (Sn 793) — the unaffiliated sage.
Sa sabbadhammesu visenibhūto, / Yaṁ kiñci diṭṭhaṁ va sutaṁ mutaṁ vā; / Tameva dassiṁ vivaṭaṁ carantaṁ, / Kenīdha lokasmi vikappayeyya.
He, unaffiliated in regard to all things — whatever belief, tradition, or rationale — being a seer-of-it, going about openly: by what here in the world would one categorise him?
Both pādas a and b recur verbatim at Snp 4.13:20 — Sa sabbadhammesu visenibhūto / Yaṁ kiñci diṭṭhaṁ va sutaṁ mutaṁ vā. Two adjacent verbatim pādas shared across suttas is unusual; the second is the AV's longest verbatim 6-word cross-sutta pāda per data/analysis/av-ngrams/findings.md. 4.13's closing verse is a doublet-echo of this verse.
v. 8 (Sn 795) — the radical closing.
Sīmātigo brāhmaṇo tassa natthi, / Ñatvā va disvā va samuggahītaṁ; / Na rāgarāgī na virāgaratto, / Tassīdha natthi paramuggahītanti.
The brahmin gone beyond the boundary — for him there is nothing tightly grasped, on knowing or on seeing. Neither in love with passion nor besotted by dispassion; for him here there is nothing taken up as ultimate.
Pāda c — Na rāgarāgī na virāgaratto — recurs verbatim at Snp 4.13:18.4. Lee 2024 (p. 218 n. 19) highlights the doctrinal force: "the negation of even the virtue of 'dispassion' when it develops into an obsession. This is particularly intriguing considering the Nikāya/Āgama's portrayal of virāga (dispassion) as a stage immediately preceding the attainment of nirvāṇa." The AV explicitly transcends the standard virāga = nibbāna soteriology — not even "no-passion" can be held as ultimate.
Choice-points
v. 1, voice of the opening. Fronsdal (footnote 1 to ch. 4) and Lee (footnote 6 to Sn 788) mark lines 1a–b as the quoted claim of a Brahmanical asserter; Norman, Bodhi, Sujato, and Thanissaro render the verse as a single continuous utterance. Both the Pj II Candābha narrative and the Yizujing Mojie narrative make the dialogical reading explicit by giving the verse a specific interlocutor. The quoted-claim reading fits the verse's diagnostic gloss ("so thinking… he falls back on knowledge") and the commentarial framings; the single-utterance reading is more cautious but loses the sutta's dialogical shape.
v. 3, Diṭṭhe sute sīlavate mute vā. The same Lee-vs-consensus reading discussed at Snp 4.3. Lee 2024 (pp. 168, 209 nn. 10) reads the four-claim formula as ascetic-claim-bases — "beliefs, traditions, precepts-and-vows, rationales" — rather than as the standard sense-experience triplet. The argument: diṭṭha is functionally equivalent to diṭṭhi in the AV (Norman p. 334); muta and ñāṇa alternate in the equivalent formula at Snp 4.7 (Sn 839, 840 diṭṭha-suta-ñāṇa-sīlabbata); and the AV's polemic is against people defending their views and inherited teachings, not against cognitive perception as such. Five of six published translators keep the sense-experience reading; Lee stands alone. The choice matters: the claim-bases reading makes the three-sutta formula (4.4, 4.5, 4.12) a tight polemical unit, while the sense-experience reading disperses the polemic.
v. 6, visenibhūto. Two etymologies, sharply different.
- Pāli commentarial (Niddesa, Pj II): folk-etymology vi-senā-bhūta = "without [Māra's] army." Supported by the Niddesa's elaborate Padhāna-sutta (Snp 3.2) Māra-army intertext at
mnd4:47–52. - Buddhist Sanskrit + Jain (Norman p. 337): vi-śreṇī-bhūta = "without association / affiliation." Supported by a Jain parallel at Āyāraṅga 1.6.33 (vissaṇikajjau; Jacobi 1884 p. 58).
Norman, Bodhi, and Sujato adopt the association-reading; Fronsdal, Thanissaro, and Lee follow the Pāli folk-etymology. The philological case for Norman is stronger (cross-traditional Buddhist-Sanskrit + Jain attestation beats a Pāli folk-etymology), and the sense — "the goal-figure is unaffiliated with any religious party" — is dialectically apt for the diṭṭhi-debate context. The Niddesa's Mārasena intertext remains interesting independently as Niddesa-era exegesis, even if the underlying etymology is wrong.
v. 8, na virāgaratto. Translators uniformly preserve the "neither passion nor dispassion" formulation but none foreground its doctrinal implication. Lee (p. 218 n. 19) is the only commentator to flag it as an AV-internal move against standard virāga = nibbāna soteriology. Whether this reading carries through the rest of the AV is settled by the verbatim recurrence at Snp 4.13:18.4.
Vocabulary and commentary
Lexical profile. The diṭṭhi-debate lexicon of 4.3 continues and extends: suddhi, sopadhīka, attañjaho, ejānugā, saññasatto, visenibhūto, vivaṭa, accantasuddhi, ādānagantha, sīmātigo, paramuggahītan, bhūripañño. The sutta also makes systematic use of Brahmanical religious vocabulary re-appropriated for Buddhist ends: brāhmaṇa is the goal-figure (not the social class); vedā is the Buddhist wisdom-cluster (not the revealed texts); sīlavata is what the goal-figure does not depend on; suddhi (the Brahmanical goal) is what the goal-figure does not claim. The sutta is a sustained appropriation-and-displacement of the religious vocabulary of its cultural setting.
Three internal verbatim bonds are load-bearing: Diṭṭhe sute sīlavate mute vā (v. 3 = 4.5:2.2 = 4.12:10.1); Sa sabbadhammesu visenibhūto (v. 6a = 4.13:20.1); Yaṁ kiñci diṭṭhaṁ va sutaṁ mutaṁ vā (v. 6b = 4.13:20.2, the AV's longest cross-sutta verbatim pāda). Na rāgarāgī na virāgaratto (v. 8c = 4.13:18.4) adds a further verbatim link. Taken together, Snp 4.4 and Snp 4.13 have the densest formulaic overlap of any AV pair — 4.13's closing verse-cluster is a near-complete echo of 4.4's closing verse-cluster. The diṭṭhi-debate cluster's compositional unity is not evenly distributed: specific sutta-pairs are in close compositional conversation.
Mahāniddesa (Mnd 4). The Niddesa's most ethnographically valuable section of AV commentary falls here — the five-fold purity-sects taxonomy at mnd4:20–24, unfolding the diṭṭhe sute sīlavate mute vā of v. 3:
- Diṭṭhi-suddhika — purity by auspicious sights (chataka-bird, plantain-shoot, pregnant woman, child-on-shoulders, full pot, red fish, Sindh horse, horse-chariot, bull, brown cow).
- Suta-suddhika — purity by auspicious sounds ("growth," "grown," "full," "fortunate," "sorrowless").
- Sīla-suddhika — purity by precept, illustrated via the direct quotation of MN 78 Samaṇamuṇḍikasutta on Pūraṇa Kassapa's school's four-fold non-evil-action criterion.
- Vata-suddhika — purity by vow; 23 vow-types including the cow-vow (Govatika) and dog-vow (Kukkuravatika) known from MN 57.
- Muta-suddhika — purity by ritual sense-touch (morning touching of earth, green vegetation, cow-dung, tortoise, plough-share).
These are not Stratum-2 doctrinal taxonomies — they are descriptions of actual contemporary religious practice, with cross-references to the canonical accounts of the same sects. This is the Niddesa at its most sociologically sharp, genuinely illuminating what the verse opposes.
Other Niddesa moves: the full dependent-origination + four-noble-truth + abhiññeyya / pariññeyya / pahātabba / bhāvetabba / sacchikātabba tetrad is imported on the bare samecca dhammaṁ of v. 5 — among the most aggressive single-phrase doctrinal imports in the AV's commentarial corpus. The four magga-paths are mapped onto the bare sīmā ("boundary") of v. 8. The brāhmaṇa etymology at mnd4:17.3–4 and the vedā re-appropriation at mnd4:40.3 codify the Buddhist re-grounding of Brahmanical vocabulary the verse itself signals through context.
Cross-recensional witnesses
Pāli: full; 8 verses.
Chinese Yizujing YZJ-4 摩竭梵志經 ("Brahmin Moka Sūtra") at [T0198_p0177c20]–[T0198_p0178a17]: 8 parallel + 0 added verses (Lee 2024 Table 2). Like YZJ-2, a clean 1:1 verse-count parallel — the second of only two AV suttas in the Yizujing with zero added verses. The Chinese preserves the full Pāli verse-set without diluting it with frame-verses, keeping all narrative expansion in prose.
The frame narrative: Mojie (摩竭) is a brahmin who claims that anyone who saw him attained liberation, and anyone who sees his corpse after his death, and anyone who hears his name later. The brahmins parade his body through Sāvatthī after he dies. The bhikkhus see the procession on alms round, report to the Buddha, who delivers Snp 4.4 as the critique.
The Pj II Candābha story (Bodhi 2017 pp. 1053–1055). Past life: a Bārāṇasī merchant befriends a woodsman in the time of Kassapa Buddha and has him make a moon-disc of sandalwood paste for Kassapa Buddha's cetiya; the woodsman wishes: "may such rays appear on my chest in any rebirth." Reborn as a deva named Candābha ("Moon-light"), then as a brahmin in Sāvatthī with a disc of rays on his chest. Other brahmins worship him as Mahābrahmā and charge fees for people to see him, promising fame, wealth, and heavenly rebirth. Candābha reaches Sāvatthī, his light eclipsed by the Buddha's at 80 hatthas distance; he goes forth and attains arahantship. The bhikkhus debate whether people gained purity just by seeing his form; the Buddha responds with Snp 4.4.
Two independent commentarial traditions converge on the darśan-critique reading. Both the Mojie story (Chinese) and the Candābha story (Pāli) identify a charismatic Brahmanical figure whose visible/audible presence is claimed to be salvifically efficacious, and both have the Buddha delivering Snp 4.4 as the refutation. Lee 2024 (p. 204 n. 4) makes the convergence explicit: "It is plausible that at the time of their composition, these narratives may have been critiques of the Brahmanical darśana tradition, which posits that the revered vision of a deity or a holy individual confers spiritual merits to the beholder." A third witness: the Dhp commentary (5th c. CE, Buddhaghosa's school) uses the Candābha story to frame Dhp 413 — a different verse in the Brāhmaṇa-vagga, but drawing on the same narrative pool.
This is the third consecutive case of the commentarial-wiring pattern observed at Snp 4.3 (Sundarī / Ud 4.8) and Snp 4.5 (blind-men parable / Ud 6.4) — the Pāli commentary and the Chinese recension independently attach a vivid narrative occasion to what the Pāli canonically preserves as frameless verse. What is distinctive about 4.4 is that there is no competing canonical Udāna story against which to compare: the Mojie and Candābha narratives diverge in detail but agree on the darśan-critique reading without a third canonical variant. The convergence on this specific interpretive frame across recensional streams suggests it may approach the verse's actual original polemical target.
Sanskrit: not attested. Hoernle 1916 covers Snp 4.7–4.10.
Gāndhārī: not attested at the verse level. The visenibhūto choice-point turns on a Jain parallel (Āyāraṅga 1.6.33 vissaṇikajjau; Jacobi 1884 p. 58) — evidence that the renunciate-vocabulary of vi-śreṇī-bhūta ("unaffiliated") circulated across Buddhist and Jain traditions as a shared śramaṇic term.
Coverage note. Snp 4.4 is pure 2-recension at verse-level (Pāli + Chinese). The cross-recensional weight falls on the framing convergence — Pj II + Yizujing + Dhp commentary — rather than on verse-level variance.
Internal cross-references
Within the AV. 4.4 is in particularly tight compositional conversation with Snp 4.13 Mahāviyūha. Four verbatim bonds: Sa sabbadhammesu visenibhūto (v. 6a = 4.13:20.1); Yaṁ kiñci diṭṭhaṁ va sutaṁ mutaṁ vā (v. 6b = 4.13:20.2 — AV's longest cross-sutta pāda); Na rāgarāgī na virāgaratto (v. 8c = 4.13:18.4); and the shared brāhmaṇa re-appropriation. 4.13's closing verse is a doublet-echo of 4.4's closing verse.
Three-sutta bond: Diṭṭhe sute sīlavate mute vā (v. 3 = 4.5:2.2 = 4.12:10.1). The attā-niratta root-family surfaces at v. 4 in its verbal form Te uggahāyanti nirassajanti (cf. 4.3:8.3 Attā nirattā, 4.10:11.3, 4.14:5.4, 4.15:20.4). Anūpalitto (v. 3) shared with nopalitto at 4.2:7.2. Ejā (v. 4) recurs across the AV's polemic suttas.
Within the Khuddaka. The monkey simile (v. 4d) is unique to Snp 4.4 in the Pāli canon — no stock-formula parallel elsewhere. Bhūripañño (v. 5) links to Snp 5.19:13.3 (Pingiya's farewell, Pārāyaṇa) — cross-AV-PV goal-vocabulary. The Niddesa imports Snp 3.6 Sabhiyasutta for the brāhmaṇa etymology (Bāhitvā sabbapāpakāni… asito tādi pavuccate sa brahmā) and Snp 3.2 Padhānasutta verses for the visenibhūta (Mārasena) reading — two substantial cross-Snp imports in one commentary chapter. Dhp 83 (Na uccāvacaṁ paṇḍitā dassayanti) shares the uccāvacaṁ religious-tourism formula with v. 5; Dhp 39 shares the puñña/pāpa-unanointed arahant-thesis with v. 3.
Prose-nikāya uptake. No direct named-citation of Snp 4.4. The Niddesa's quotation of MN 78 Samaṇamuṇḍikasutta (for Sīla-suddhika) and its implicit reference to MN 57 Kukkuravatikasutta (for Vata-suddhika) ground the verse's opposition to contemporary religious sects in the prose-canon's own record of them. AN 6.30 uses uccāvacaṁ in the religious-tourism sense closely parallel to v. 5.
Reception and external attestation
Mahāniddesa: Mnd 4 covers all eight verses. See Vocabulary and commentary.
Paramatthajotikā II: the Candābha story (see Cross-recensional witnesses). The same narrative is used independently by the Dhp commentary to frame Dhp 413, suggesting the Candābha-cycle was a general commentarial resource for diṭṭhi- and brāhmaṇa-vocabulary verses.
Aśoka Bhabru edict: not identified.
Peṭakopadesa: no verse of Snp 4.4 is cited in Pe chapter 1's AV-extraction.
Reading
Snp 4.4 extends the diṭṭhi-polemic set-piece of 4.3 in two doctrinally-consequential directions.
First, it narrows the polemic to a specific target. The commentarial traditions — Pj II's Candābha story, the Yizujing's Mojie story, and the Dhp commentary's independent use of Candābha for Dhp 413 — converge on identifying the verse's opening claim as a Brahmanical darśan doctrine: the position that the sight or hearing of a holy figure is salvifically efficacious. What makes this case distinctive among AV commentarial wirings (compare 4.3 Sundarī / Ud 4.8; 4.5 blind-men / Ud 6.4) is that there is no competing canonical narrative to adjudicate against. The Pāli and Chinese commentaries independently chose a Brahmanical-figure-with-visible-presence as the story-type; neither is constrained by a prior Udāna. The convergence is therefore unusually load-bearing — it suggests the darśan-critique reading may approach the verse's actual original polemical target, and it accounts for the verse's conspicuously first-person opening Passāmi suddhaṁ paramaṁ arogaṁ, which reads more naturally as quotation of a real contemporary claim than as an abstract hypothetical.
Second, v. 8's Na rāgarāgī na virāgaratto is the AV's most radical no-position move. The prose Nikāyas / Āgamas routinely present virāga (dispassion) as a stage immediately preceding nibbāna; the standard virāga = nibbāna soteriology is foundational to later Buddhist doctrine. The AV here explicitly rules out attachment to the goal-state itself. Even "no-passion" can be held as a position; the goal-figure holds no position. The verbatim recurrence of the formula at Snp 4.13:18.4 confirms this is a recurring AV thesis, not a one-off formulation. When Chapter 3 and Chapter 4 return to the apophasis-vs-right-view-in-apophatic-register scholarly debate, this verse is a locus: the AV's no-position thesis extends even to the standard positive predicates of the canonical path, which makes the apophatic reading (Fronsdal, Gómez) harder to dismiss and the right-view-in-apophatic-register reading (Bodhi, Sujato) harder to secure.
The formulaic bonds between 4.4 and 4.13 are the densest AV-internal compositional link observed so far. Four verbatim parallels — three in a single verse-cluster (v. 6 pādas a and b + v. 8 pāda c) — make 4.13 Mahāviyūha's closing verses a near-echo of 4.4's closing verses. This is not random formulaic sharing across the diṭṭhi-debate cluster; it is specific suttas in close compositional conversation. Chapter 4 will need to account for the pattern, as will any stratigraphic reading of the collection.