Practice · SīlaVolume FourPart III · Three Vinaya SchoolsPaper 10 of 30

Three Vinaya Lineages in Conversation

中文主題：三系律藏之對話

Author: Shi Huijing 釋慧鏡

Date: Spring 2026

Version: 1.0

ID: SILA-P10

Abstract

Cross-tradition Vinaya comparison is often presumed to be a twentieth-century Western scholarly imposition; this assumption does not survive the textual evidence. Yijing's late seventh-century critique of the multi-Vinaya situation in Tang China — written after twenty-five years of direct observation at Nālandā — together with his contemporary Daoxuan's six-recension cataloguing in the Sifenlü Shanfan Buque Xingshichao (T1804), already establishes a robust indigenous foundation for comparative Vinaya scholarship. This paper takes the placement of the eight garudhammas (八敬法) as the primary comparative axis, the eight-category prātimokṣa sequence as the stable-core baseline, and five Vinaya recensions — Mahīśāsaka, Mahāsāṅghika, Sarvāstivāda, Dharmaguptaka, and Mūlasarvāstivāda — as the cross-comparison sample. Three findings emerge: the eight-category prātimokṣa sequence is stable across all five recensions; the placement of the garudhammas is consolidated in four of the five, with Mūlasarvāstivāda alone exhibiting a doubled register (narrative form in T1451 Kṣudrakavastu juans 29–30, dispersed-rule form in T1443 Bhikṣuṇī-vinaya juan 19); and Mahāsāṅghika carries a ninth rubric (dharma-anudharma) beyond the shared octave. The pairing of a stable core with a variable periphery constitutes a redactor's-eye finding internal to the comparative Vinaya tradition itself — neither an external ranking of any single recension nor a contemporary normative critique.

I. Introduction

In the late seventh century, after twenty-five years of direct observation at Nālandā, the Chinese pilgrim Yijing returned to Tang China and, in his Nanhai jigui neifa zhuan (《南海寄歸內法傳》, T2125), summarised the state of Vinaya practice in his home country with two pointed phrases: that on Chinese soil "the various Vinaya-holding schools were pulling on each other" and that the situation had become "tangled and disorderly in its commentaries and excerpts" (神州持律諸部互牽 / 章鈔繁雜).¹ This was not an externally imposed comparative perspective. It was the response of an East Asian Vinaya master with direct experience of how the Indian Vinaya was actually being lived, to the lived reality of his home country, where several Vinaya traditions — Dharmaguptaka, Sarvāstivāda, Mahīśāsaka, Mūlasarvāstivāda — were co-circulating in monastic practice with porous and sometimes confused boundaries.

His contemporary Daoxuan, working from a distinctively Chinese vantage point, performed the same comparative work in a different register. In fascicle 1 of his Sifenlü Shanfan Buque Xingshichao (《四分律刪繁補闕行事鈔》, T1804), Daoxuan enumerated "the six authentic recensions" — Saṃghika, Dharmaguptaka, Sarvāstivāda, Mahīśāsaka, Kāśyapīya, and Vātsīputrīya — and offered a typology of "the six kinds of Vinaya-holders in the world", together with a self-disclosed comparative practice ("the texts list four readings; the Sarvāstivāda Daśa-bhāṇavāra is in seal-print agreement in meaning").² Yijing's critique and Daoxuan's classification approach the same fact from opposite vantage points — one Indian, one Chinese — but what they jointly address is the brute fact of cross-tradition coexistence. By the seventh and eighth centuries in East Asia, comparative Vinaya scholarship was already an indigenous concern of working Vinaya masters. It was not first imported by Western Buddhist studies in the twentieth century.

The present paper takes the three-lineage frame as its principal stage: by "three lineages" we mean the Pāli Vinaya of the Theravāda, the Dharmaguptaka-vinaya (T1428) on which the East Asian Vinaya-school depends, and the Mūlasarvāstivāda Vinaya group translated by Yijing. To ensure that the comparison sample is representative, we extend the witness layer to a five-recension array by adding the Mahīśāsaka Vinaya (T1421), the Mahāsāṅghika Vinaya (T1425), and the Sarvāstivāda Vinaya (T1435), with the first three lineages forming the principal axis and the latter three forming the cross-witness axis. The previous paper (P09) read the Pāṭimokkha within a single tradition as a self-legislating casuistic case law.³ The present paper extends that reading along its natural axis: from one tradition to many, from intra-tradition architecture to synchronic cross-tradition comparison. Two questions follow. First, does the self-legislating categorial logic of the prātimokṣa hold across the three lineages? Second, if what is stable across the three lies in one stratum and what diverges lies in another, what does that distribution itself disclose about the deeper architecture of Vinaya redaction?

The thesis of the paper is the following. The comparison of the three Vinaya-lineages takes no single recension as benchmark. The textual core — the eight-category sequence of the prātimokṣa — is stable across all three lineages and across all five recensions. The textual periphery — where the eight garudhammas (the eight weighty rules) are placed within the Vinaya — diverges, with Mūlasarvāstivāda alone exhibiting a doubled register: a narrative form in the Kṣudrakavastu (T1451) and a dispersed-rule form in the Bhikṣuṇī-vinaya (T1443). This pairing of a stable core with a variable periphery is a redactor's-eye finding internal to the comparative Vinaya tradition itself. The register of this paper is descriptive-historical, not normative-prescriptive. We name no living saṅgha; we take no position on the contemporary bhikkhunī-ordination debates; we rank no recension above another.

II. The Three Textual Axes Placed Side by Side

2.1 The Theravāda Pāli axis

The most direct attestation of the eight-category Pāṭimokkha sequence in the Pāli Vinaya appears in the First Council scene of Cullavagga XI, where Upāli is recited as having recited the Pāṭimokkha in this order: pārājika, saṅghādisesa, aniyata, nissaggiya-pācittiya, pācittiya, pāṭidesanīya, sekhiya, and adhikaraṇa-samatha — eight categories in total.⁴ In the same chapter, Mahākassapa, when ruling on the question of the "lesser and minor training-rules" (khuddānukhuddaka), instituted what would become the transmission discipline of the whole tradition. The Pāli formula — apaññattaṁ nappaññapeyya paññattaṁ na samucchindeyya yathāpaññattesu sikkhāpadesu samādāya vatteyya, "what has not been laid down shall not be laid down; what has been laid down shall not be cut off; one should accept and live by the training-rules as they have been laid down"⁵ — is a foundational textual proposition shared across the lineages: the content of the Vinaya may have its bounds set by the Buddha's institution, but its transmission must not be expanded or pruned by later hands at will.

Pācittiya 72 of the Bhikkhu-pāṭimokkha, the Vilekhanasikkhāpada, supplies the Pāṭimokkha's self-referential signature: the rule, within the text of the Pāṭimokkha itself, refers to the event of the Pāṭimokkha being recited (pātimokkhe uddissamāne).⁶ At the grammatical surface this is a recursive feature of the text; at the structural level it attests the Vinaya's self-legislating categorial logic — that the institution within which the rule operates is itself protected, by name, inside the rule.

The eight garudhammas (aṭṭha garudhammā) in the Pāli Vinaya are placed in their own chapter, Cullavagga X, the Bhikkhunī-khandhaka. The eight rules are framed by the ordination narrative of Mahāprajāpatī Gotamī and laid out as a continuous block.⁷

Across the Pāli stratum there is one further continuity-of-doctrine link worth noting. The Aṅguttara-nikāya situates the prātimokṣa-restraint within a developmental sequence — AN 4.245 Sikkhānisaṃsa-sutta on the benefits of training, and AN 11.1 Kimatthiya-sutta on the eleven-link chain that places sīla at the head of the path to wisdom. This developmental axis is shared canonical ground across all three Vinaya-lineages: whatever the recension-level divergences, the sikkhā axis itself, and the sīla → paññā chain that builds upon it, is convergent doctrinal territory.

2.2 The Dharmaguptaka Sifenlü axis

The Dharmaguptaka-vinaya (T1428) is the textual axis on which the East Asian Vinaya-school has depended for more than a thousand years. Its opening fascicle announces the "ten purposes" (dasa atthavase paṭicca — 十句義) for which the precepts were laid down, and the eight-category, seven-aggregate (pañca-khanda / satta-piṇḍa) architecture organises the rest of the text.⁸ The eight-category sequence of the prātimokṣa appears in fascicle 54 in the First Council scene of Upāli's recitation — pārājika, saṅghādisesa, aniyata, naiḥsargika-pāyantika, pāyantika, pratideśanīya, śaikṣa, adhikaraṇa-śamatha — in one-to-one correspondence with Cullavagga XI. The Second Council scene of the seven hundred follows in the same fascicle (lines 291–560), where Mahākāśyapa, in the parallel role to the Pāli redaction, similarly handles the question of the "lesser and minor" rules across the eight categories.⁹ The Dharmaguptaka–Pāli correspondence at the prātimokṣa sequence level admits direct, point-by-point cross-mapping.

The eight garudhammas in the Dharmaguptaka recension are consolidated in fascicle 48, the Bhikṣuṇī-khandhaka. The eight rules sit together in a single block, each closed by the uniform formula "this rule is to be respected, revered, and praised; it is not to be transgressed for as long as life endures"; the block is framed by the marriage-garland upasampadā simile ("if a woman can keep these, this itself is upasampadā") and sealed by the five-hundred-year saddhamma-duration prediction.¹⁰ Both the consolidation and the closure are strong textual signals and admit direct cross-mapping with Cullavagga X.

The Dharmaguptaka-bhikṣu-prātimokṣa (T1429) carries a parallel attestation of the eightfold sequence in the form of a recurring framing-formula: "fortnight by fortnight to be recited, drawn from the Bhikṣu-prātimokṣa" (半月半月說、戒經中來). The formula occurs nine times across the Prātimokṣa text, covering all eight categories of the eightfold division — the precise functional equivalent of the Pāli "venerable ones, these four pārājika rules come up for recitation in the Pāṭimokkha" (imāni kho panāyasmanto cattāro pārājikā dhammā uddesaṃ āgacchanti).¹¹ The nine framing-occurrences themselves are an independent witness to the stability of the eight-category sequence at the Prātimokṣa level.

One terminological note is warranted before proceeding. The Pāli term pācittiya and its Sanskrit cognate pāyantika receive three different Chinese transcriptions across the recensions: Dharmaguptaka renders 波逸提, Mahāsāṅghika renders 波夜提, Mūlasarvāstivāda renders 波逸底迦, while Sarvāstivāda again renders 波逸提. These are three transcriptions of the same root, not three categorial entities. We note this once and do not repeat the gloss.

2.3 The Mūlasarvāstivāda doubled register

The Mūlasarvāstivāda Vinaya in Chinese is the work of a single translation project — Yijing's, between 700 and 712 in Luoyang and Chang'an — and survives in a cluster of texts: the Bhikṣu-vinaya (T1442), the Bhikṣuṇī-vinaya (T1443), the Saṅghabhedavastu (T1450), the Kṣudrakavastu (T1451), and several others. Within this single tradition, the eight garudhammas are placed in a manner that differs from the other four recensions in a structurally distinctive way. There is no consolidated block, and there is no single textual register. Instead the same eight rules are distributed across two registers within the same tradition: a narrative form and a dispersed-rule form.

The narrative form occurs at the close of fascicle 29 and the opening of fascicle 30 of the Kṣudrakavastu (T1451). Here we find Mahāprajāpatī's three requests for ordination; the Buddha's three hesitations expressed in three similes (a household with few men and many women, a crop struck by hail, a sugarcane stalk with red joints); Ānanda's intercession; the eight garudhammas presented one by one, each addressed to "Ānanda!"; and Mahāprajāpatī's post-acceptance ordination posture, the receiving of the rules upon her head as a garland. The whole passage reads as a jātaka-like narrative, not as a regulatory code.¹² One correction is required for the present series' internal record: an earlier project note placed the garudhamma narrative at fascicle 40 of T1451; in fact fascicle 40 belongs to the Second Council material (the territory of the next paper, P11). The garudhamma narrative is at fascicles 29–30. The opening of T1451 fascicle 1, moreover, names itself as "Kṣudraka-vastu" — organised by uddāna-summary topical clusters rather than by numbered śikṣāpada — and this self-naming is the narrative register's own framing.¹³

The dispersed-rule form appears in fascicle 19 of the Mūlasarvāstivāda-bhikṣuṇī-vinaya (T1443). The same eight garudhammas are no longer continuous in this register; they are distributed across the numbered śikṣāpada table as individual pāyantika rules. Śikṣāpada 126, the "rule on requesting instruction", corresponds to Pāli garudhamma 3 (the fortnightly request for instruction); śikṣāpada 127, the "rule on holding the uposatha where there are no bhikṣus", corresponds to the requirement that uposatha be performed in the presence of bhikṣus; śikṣāpada 128, the "rule on entering the rains-retreat where there are no bhikṣus", corresponds to Pāli garudhamma 2 (rains-residence in dependence on the bhikṣu-residence); śikṣāpada 129, the "rule on not performing the pravāraṇā three observances at the joint assembly", corresponds to Pāli garudhamma 4 (the threefold pravāraṇā before both communities). Each of the four closes with the pāyantika category-marker and is embedded in the continuous pāyantika numbering — fascicles 15 through 20 cover śikṣāpada 38 onward into the 180s.¹⁴ In format, these four rules sit fully on a level with their neighbours: no consolidation, no narrative frame, no closure formula distinguishing them from the surrounding pāyantika material. The opening Vinaya-preface of T1443 fascicle 1 announces the bhikṣuṇī-rules as "prātimokṣa-sūtra" (別解脫經) and notes the "eight pārājika rules" as the bhikṣuṇī-code's opening marker.¹⁵ But the eight garudhammas themselves, by the time we reach fascicle 19, no longer have the textual identity of a consolidated block.

The same Vinaya tradition, the same translator, the same translation project, the same decade: and yet two genuinely distinct registers stand side by side. T1451's narrative form develops garudhamma material along the units of person–scene–occasion; T1443's dispersed-rule form develops the same material as numbered śikṣāpada. This internal doubling is not a cross-tradition phenomenon. It is not shared by the other four recensions. It is fully internal to Mūlasarvāstivāda — and crucially, its detection does not require cross-recension comparison. It is visible simply by reading the two T-numbers of the same translator side by side.

A second terminological note. The Sarvāstivāda-vinaya (T1435, Daśa-bhāṇavāra, translated by Puṇyatara and Kumārajīva, 404–409) and the Mūlasarvāstivāda-vinaya cluster (T1442 / T1443 / T1450 / T1451, translated by Yijing, 700–712) are not the same tradition despite the proximity of their names. Yijing himself, in T2125, draws the distinction and corrects what was already a common Chinese confusion ("the Daśa-bhāṇavāra is the Mūlasarvāstivāda Vinaya"). We return to this point in §V.

III. The Five-Recension Architectural Triangulation

3.1 The five-recension comparison table

Extending the comparison sample to six witnesses — the Pāli of the Theravāda, the Mahīśāsaka-vinaya (T1421), the Mahāsāṅghika-vinaya (T1425), the Sarvāstivāda-vinaya (T1435), the Dharmaguptaka-vinaya (T1428), and the Mūlasarvāstivāda cluster (T1442 / T1443 / T1450 / T1451) — the placement of the eight garudhammas distributes as follows.

Recension	Principal text	Placement of the eight garudhammas	Textual form
Theravāda	Pāli Cullavagga X	Bhikkhunī-khandhaka opening consolidated block	Consolidated; Mahāprajāpatī's ordination narrative as frame
Mahīśāsaka	T1421 fascicle 29	"Section Five, Eight Rules for Bhikṣuṇīs" consolidated block	Consolidated; Gotamī ordination narrative as frame; eight non-transgressible rules in a continuous block
Mahāsāṅghika	T1425 fascicles 30 & 32	Fascicle 30 Bhikṣuṇī-dharma opening (line 359 onward)	Strongest consolidated form; each rule with its own "transgression-of-gurudharma" gloss; sealed with "the Nth gurudharma ends"
Sarvāstivāda	T1435 fascicles 42–47	Within the Bhikṣuṇī-vinaya section; fascicle 47 separates an independent garudharma chapter (line 523)	Double consolidated (within section + standalone chapter); embedded in fortnightly-instruction (ardhamāsa-uddesa) frame, not ordination-narrative frame
Dharmaguptaka	T1428 fascicle 48	Bhikṣuṇī-khandhaka consolidated block	Consolidated; uniform closure formula per rule; five-hundred-year prediction as block-end seal
Mūlasarvāstivāda	T1451 fascicles 29–30 + T1443 fascicle 19	Narrative form in T1451; dispersed-rule form in T1443 śikṣāpada 126–129	Doubled register internal to the tradition; no single consolidated block

At the layer of garudhamma-placement, five of the six witnesses (Theravāda, Mahīśāsaka, Mahāsāṅghika, Sarvāstivāda, Dharmaguptaka) share the consolidated form. Mūlasarvāstivāda alone exhibits the doubled-register pattern — one of six. The distribution is five-to-one, not three-to-three and not even. The asymmetry itself is the principal finding of the comparison.

3.2 Narrative and dispersed-rule registers coexisting within a single tradition

The Mūlasarvāstivāda internal split carries a further structural implication. The same eight rules can appear, within a single tradition, in two textually distinct registers — narrative (T1451) and dispersed-rule (T1443). This is not a cross-tradition difference but an intra-tradition register difference. The Vinaya in this tradition does not present itself as a single textual genre. It is not the case that vinaya names a single text-type; rather, in this one tradition, vinaya simultaneously takes the form of a person-scene-occasion narrative and the form of a numbered rule-table.

This phenomenon stands in structural correspondence with the Suttavibhaṅga-versus-Pāṭimokkha split discussed in the previous paper (P09). There, within a single tradition, the Suttavibhaṅga developed the four-component anatomy of each rule (nidāna, sikkhāpada, padabhājanīya, vinīta-vatthu) — already a composite of narrative and rule elements — while the Pāṭimokkha compiled the regulatory clauses alone. P09 displayed this narrative-versus-rule split as an intra-tradition phenomenon at the Suttavibhaṅga/Pāṭimokkha level. The present paper extends the observation: in Mūlasarvāstivāda, the same intra-tradition split appears at the Vinaya-text level, with two separate texts in the same translation project occupying the two registers — T1451 takes the narrative pole, T1443 takes the rule pole. This intra-tradition register-doubling is, among the five recensions, found only in Mūlasarvāstivāda.

3.3 Stable core and variable periphery as a dual architecture

The five-recension comparison discloses a dual-layer architecture.

The first layer is the stable core. The eight-category prātimokṣa sequence — pārājika, saṅghādisesa, aniyata, naiḥsargika-pāyantika, pāyantika, pratideśanīya, śaikṣa, adhikaraṇa-śamatha — admits one-to-one cross-mapping across five of the six witnesses (Theravāda, Mahīśāsaka, Dharmaguptaka, Sarvāstivāda, Mūlasarvāstivāda). The Mahāsāṅghika carries an added ninth heading, but its first eight match the same sequence (see §3.4 below). The category-names are linguistically diverse across recensions (three Chinese renderings of the same pāyantika root, as noted at §2.2), but their category-rule-vinīta-vatthu identity is structurally stable. This stability is the cross-recension convergence baseline for comparative Vinaya work.

The second layer is the variable periphery. The placement of the eight garudhammas — whether they are consolidated, whether they are narrativised, whether they are dispersed into the numbered śikṣāpada — diverges across the recensions. Importantly, this divergence is not a two-versus-three split. It is a four-to-one asymmetry: four recensions converge on consolidation, one alone diverges into doubled register. The variation at the periphery does not destabilise the convergence at the core. The stable core is not the whole Vinaya; the variable periphery is not the whole Vinaya either. The two coexist at different strata of the same text — core convergent, periphery divergent. That is the dual finding of comparative Vinaya at the architectural level.

At the practice-and-scholarship interface, the implication of this dual structure may be stated as follows: the cross-tradition variation of the Vinaya is not uniformly distributed. The stability of the core layer secures that the three Vinaya-lineages point to the same teaching; the variation of the periphery layer displays that the same teaching can take more than one historical embodiment. Textual diversity is in service of doctrinal coherence, not the other way around.

3.4 The Mahāsāṅghika ninth-rubric variant axis

The Mahāsāṅghika-vinaya (T1425), in fascicle 32, adds beyond the shared eightfold base a ninth heading — dharma-anudharma ("the Dharma and what conforms to the Dharma", 法隨順法). Its first eight categories correspond fully with the other four recensions; the ninth is unique to the Mahāsāṅghika.¹⁶ This added heading sits inside the Mahāsāṅghika tradition's historical distinctiveness as a separate school from the Sthavira side, and provides a textual witness to that split at the level of prātimokṣa organisation — the first eight as the shared cross-recension base, the ninth as the recension-specific addition. The eight-plus-one pattern is itself a further instance of the "stable core plus variable periphery" architecture on another axis.

One further note. The Sarvāstivāda placement of the eight garudhammas, while consolidated, sits inside the fortnightly-instruction (ardhamāsa-uddesa) framing rather than the ordination-narrative framing.¹⁷ The placement-layer divergence among the four consolidated recensions is itself further textured at the level of framing. This paper does not expand that sub-layer in detail; we register it here as a refinement, not as a counterexample.

IV. Three-Scholar Dialogue

4.1 Bhikkhu Anālayo (Pāli leg)

Among contemporary scholars, the most systematic treatment of the cross-tradition textual stability of the bhikkhunī-rules is the work of Bhikkhu Anālayo. The Foundation History of the Nuns' Order (Hamburg University Press, 2016), together with the trilogy of articles in the Journal of Buddhist Ethics between 2014 and 2017, brings the cross-recension philology of the eight garudhammas to the mature current standard of the field.¹⁸ Anālayo's core method is historical-foundationalist: cross-recension comparison is deployed to identify which elements belong to the early stratum and which are later redactional additions, with the aim of recovering the historical layers of the garudhamma material.

The present paper acknowledges the philological baseline Anālayo has set, and pushes one step along a different methodological axis. The historical-foundationalist question is when: which form is earliest, which late, what is the redactional history? The question of the present paper is how the redactors arranged: why does the placement-layer distribute in a four-to-one asymmetry across the five recensions? The two registers — placement-as-consolidated versus placement-as-dispersed — are visible to the redactor's eye, not the origin-historian's eye. The two approaches do not conflict; they ask different questions. The first asks about temporal layering ("earlier or later"); the second asks about synchronic arrangement ("how laid out").

4.2 Ann Heirman (Dharmaguptaka leg)

The most systematic contemporary study of the Dharmaguptaka bhikṣuṇī-code is Ann Heirman's Rules for Nuns according to the Dharmaguptakavinaya, three volumes (Motilal Banarsidass, Buddhist Tradition Series 47–49, 2002).¹⁹ This is the strongest contemporary Western monograph on the Dharmaguptaka nuns' code, and the starting point for any East Asian Vinaya study that engages this material.

The present paper acknowledges Heirman's single-recension depth and supplements it along the cross-recension breadth axis. Heirman handles the Dharmaguptaka as a self-contained study within one tradition; the present paper places the Dharmaguptaka within a five-recension comparative frame, where its consolidated form sits as "one of five, four of which share the consolidated pattern" — a majority pattern, not a default baseline. The consolidated form of the Dharmaguptaka, viewed from inside the single tradition, is its native architectural feature; viewed across the five-recension array, it is one instance of a shared four-recension pattern. Heirman's single-recension depth makes the cross-recension breadth of the present paper available; the breadth would be unreadable without the depth.

4.3 Shayne Clarke (Mūlasarvāstivāda leg)

The principal contemporary study of the Mūlasarvāstivāda Vinaya is Shayne Clarke's Family Matters in Indian Buddhist Monasticisms (University of Hawai'i Press, 2014), together with several Mūlasarvāstivāda-specific articles including "Lost in Tibet, Found in Bhutan".²⁰ Clarke's principal thesis is that the narrative density of the Mūlasarvāstivāda — its tendency to develop Vinaya material through persons, scenes, and occasions rather than through abstracted code — is the defining textual feature of this tradition.

The present paper acknowledges Clarke's observation of narrative density and supplements it with a complementary register internal to the same tradition. Clarke's focus is on the narrative side (the Kṣudrakavastu tradition exemplified by T1451). The present paper adds the observation of the dispersed-rule side (T1443 Bhikṣuṇī-vinaya fascicle 19). The full picture of the Mūlasarvāstivāda is not narrative alone; it is the doubled register, narrative (T1451) and dispersed-rule (T1443) coexisting in the same tradition. The internal register-split is a structural feature that single-axis narrative-focus misses; once the second register is included, the Mūlasarvāstivāda's picture becomes architecturally complete.

4.4 Brief acknowledgement and an honest disclosure

The three scholars carried over from P09 — Bhikkhu Sujato's Pāli Vinaya English translation and Bhikkhunī-vibhaṅga introduction; Oskar von Hinüber's A Handbook of Pāli Literature (de Gruyter, 1996) on the redactional stratigraphy of the Pāli Vinaya; and W. Pachow's A Comparative Study of the Prātimokṣa (Sino-Indian Studies V, 1955), the foundational figure of all modern cross-recension Prātimokṣa comparison — each receive one-line acknowledgement here without further development.²¹ Pachow in particular is the figure on whose 1955 foundation the entire subsequent comparative Vinaya project (including the work of Anālayo, Heirman, and Clarke) rests.

One honest disclosure is owed. The specific finding of the present paper — that across the five-recension array the Mūlasarvāstivāda is the sole exception, and that its defining feature in this comparison is the doubled register — does not yet appear, to our knowledge, as a focused monograph treatment in contemporary Western scholarship. Anālayo, Heirman, and Clarke each work within deep specialisation, predominantly within a single tradition or a particular two-tradition comparison. The five-recension architectural reading of the present paper attempts a positioning in a region of relatively sparse contemporary attention rather than an inheritance of an already-mapped comparative tradition.

V. Indigenous Cross-Tradition Awareness in the Chinese Commentaries

§V is the distinctive contribution of the present paper. The cross-recension comparative work of Daoxuan and Yijing demonstrates that comparative Vinaya scholarship is not an external imposition by twentieth-century Western academy but an indigenous concern that East Asian Vinaya masters had already taken up in the seventh and eighth centuries.

5.1 Daoxuan's six-recension classification in the Xingshichao

In fascicle 1 of his Sifenlü Shanfan Buque Xingshichao (T1804), Daoxuan offers an explicit classification of the cross-recension situation in his day. His enumeration of the "six authentic recensions" runs in order: Saṃghika (Mahāsāṅghika), Tánwúdé (Dharmaguptaka), Sarvāstivāda, Mīshāsai (Mahīśāsaka), Jiāyèyí (Kāśyapīya), and Pólúfúluó (Vātsīputrīya).²² The six-recension list is set down with the Dharmaguptaka as the principal house tradition, but the cross-recension inventory itself is explicit and named. In the same fascicle Daoxuan also lays out a typology of "the six kinds of Vinaya-holders in the world" and self-discloses a comparative practice: "the texts list four readings; the Daśa-bhāṇavāra is in seal-print agreement in meaning" — that is, when comparing four recensions on the same rule, the rendering in the Daśa-bhāṇavāra (T1435, Sarvāstivāda) agrees in translation and in meaning. This self-disclosure is itself a seventh-century indigenous practice of cross-recension textual comparison — methodologically equivalent to what modern scholarship calls textual collation, even if it does not use the modern terminology.

5.2 The continuation in word-level collation: Yuanzhao's Zichiji

The comparative work Daoxuan opened is carried to a finer-grained register by the Song Nanshan master Yuanzhao in his Sifenlü xingshichao zichi ji (《四分律行事鈔資持記》, T1805), a sub-commentary on Daoxuan's Xingshichao.²³ In one passage on the "Mahīśāsaka and Daśa-bhāṇavāra general agreement" (五分十誦大同), he compares the upasampadā procedure across the Mahīśāsaka and Sarvāstivāda texts; in a passage on the Vinaya-time-and-place categories of the Mingliao lun (《明了論》, Sammitīya Vinaya commentary), he traces transmission marks of the Sammitīya school; and in a passage on the phrase "these two characters refer to the two schools" he does word-level cross-school disambiguation. This word-level collation extends the cross-recension comparison Daoxuan established in the Xingshichao (seventh century) and refines it within the Song Nanshan line; its granularity is methodologically continuous with modern comparative Vinaya scholarship — again without the modern terminology, but with the same substantive method. The indigenous comparative tradition thus runs unbroken from Daoxuan in the Tang to Yuanzhao in the Song.

5.3 The three indigenous segments in Yijing's Nanhai jigui

Yijing's indigenous cross-tradition perspective rests on three textual segments in the Nanhai jigui neifa zhuan.

The first is the geographical-distribution segment. Yijing notes, in T2125, that the northern regions of Tang China principally use the Dharmaguptaka, while the southern Jiangnan area carries a mixture of several recensions; the geographical variation is itself an indigenous datum for comparative Vinaya observation.²⁴

The second is Yijing's self-disclosed Mūlasarvāstivāda standpoint with historical-philological precision. In the same text he states plainly that his own translations all follow the Mūlasarvāstivāda Vinaya, and in the same passage explicitly corrects what was already a current Chinese confusion ("the Daśa-bhāṇavāra is the Mūlasarvāstivāda Vinaya") and supplies the historical-philological frame "the Sarvāstivāda group divides into three: Dharmaguptaka, Mahīśāsaka, Kāśyapīya".²⁵ The historical significance of this self-disclosure is that it is a seventh-century contemporary observation by an East Asian Vinaya master with twenty-five years of direct Indian experience — not a retrospective sectarian positioning. This passage must be presented in a descriptive-historical register; it should not be expanded into a contemporary advocacy for Mūlasarvāstivāda superiority.

The third is the critical segment already quoted in §I: the "various Vinaya-holding schools were pulling on each other" and "tangled and disorderly in its commentaries and excerpts". The context of this critique is the same East Asian Vinaya master, returning from long observation of Indian Vinaya practice, commenting on the conditions he found at home — the coexistence of multiple recensions with blurred boundaries. What Yijing performs here is itself a piece of comparative Vinaya work: his stance is descriptive-historical, not ranking-prescriptive.

5.4 A four-layered indigenous-evidence summary

Compiling the comparative work of Daoxuan and Yijing into a four-layered evidence summary for the claim that comparative Vinaya is an indigenous East Asian concern rather than an externally imposed twentieth-century framework:

First layer: Daoxuan's six-recension enumeration in the Xingshichao establishes the cross-recension factual inventory at the level of recension identification and naming.

Second layer: Yuanzhao's word-level collation in the Zichiji (a Song sub-commentary on Daoxuan's Xingshichao) extends the cross-recension comparative method Daoxuan opened to the level of word-level textual scholarship.

Third layer: the Abhidharma stratum — the Abhidharmakośa (T1558) and the Abhidharma-samuccaya (T1605) treatments of the three-restraint architecture — provides recension-aware doctrinal classification at the śāstra level.

Fourth layer: Vasubandhu's eightfold prātimokṣa-saṃvara inventory in Abhidharmakośa fascicle 14 (treated in P09) provides cross-recension synthesis at the śāstra pivot point.

The combined conclusion is that comparative Vinaya scholarship was already a working indigenous concern in seventh- and eighth-century East Asia — Daoxuan's Vinaya-text recension-inventory, Yijing's transregional observation, and the recension-aware śāstra stratum — and was carried to word-level granularity at the Prātimokṣa-text stratum by the Song Nanshan line (Yuanzhao's Zichiji); an unbroken thread from Tang to Song. The contribution of the present paper sits inside this indigenous tradition's extension, not outside it. What we add is one further layer of synchronic structural reading — the placement-distribution pattern across five recensions — not a new external framework.

5.5 The Zhiyi-equivalent translation thread (carried forward)

The previous paper treated the cross-recension equivalence-of-translation work of Zhiyi and his successors on the "everywhere-liberation" (處處解脫) and "obtainable-liberation" (保得解脫) renderings of prātimokṣa. We pick this up here in one line and do not extend it further: the Tiantai and Nanshan Vinaya schools also contributed to this cross-tradition equivalence-translation work, but their principal venues are the preceding and following papers in the series, not the present one.

VI. Conclusion

On the principal stage of the three Vinaya-lineages (Theravāda Pāli, Dharmaguptaka Sifenlü, Mūlasarvāstivāda), with the placement of the eight garudhammas as the primary comparative axis, the eight-category prātimokṣa sequence as the stable-core baseline, and the five-recension array as the comparison sample, three structural findings emerge. First, the eight-category prātimokṣa sequence is stable across all five recensions. Second, the placement of the garudhammas is consolidated in four of the five recensions, with Mūlasarvāstivāda alone exhibiting the doubled register (narrative T1451 plus dispersed-rule T1443). Third, the Mahāsāṅghika carries a ninth rubric (dharma-anudharma) beyond the shared eightfold base. The dual architecture of a stable core and a variable periphery is a redactor's-eye finding internal to the comparative Vinaya tradition itself — neither a ranking of one recension above another, nor a contemporary normative critique.

At the practice-and-scholarship interface, the meaning of this structural finding may be stated as follows. The comparison of the three Vinaya-lineages is not a project of adjudicating which is true and which false; it is a way of observing the same self-legislating categorial logic in more than one historical embodiment. Textual diversity is in service of doctrinal coherence. The three Vinaya-lineages differ in detail but converge on the same sikkhā axis and on the same sīla-to-paññā developmental sequence anchored at the head of the AN 11.1 chain — the same teaching in multiple textual instantiations.

At the scholarship-and-institution interface, the meaning of the same finding may be stated as follows. Comparative Vinaya scholarship is not an external twentieth-century imposition by Western academy; it is the historical extension of an indigenous East Asian Vinaya-master concern — Daoxuan, Yijing, and others — into the seventh and eighth centuries. The present paper works within the continuation of that indigenous lineage. What it adds is one layer of synchronic-structural pattern-recognition, not the introduction of a new framework.

The present paper has focused on the synchronic comparison of the three Vinaya-lineages — the placement of the eight garudhammas and the eight-category prātimokṣa sequence as the dual architecture of a stable core and a variable periphery. The next paper (SĪLA-P11, The Second Council and the Devadatta Episodes) will take up the diachronic narrative axis: the Second Council and the seven hundred at Vaiśālī, and the inversion of the fifth of Devadatta's five demands between the Mūlasarvāstivāda and the Pāli traditions — where the Mūlasarvāstivāda Saṅghabhedavastu (T1450 fascicle 10) frames the fifth demand as "dwelling in village dwellings" (an anti-eremitic, comfort-seeking demand), while the Pāli tradition frames it as a strict forest-only demand (an ascetic intensification). The diachronic axis observes the same cross-recension discipline as the present paper's synchronic axis, but along a different time-arrow. The two papers together offer a more complete picture of the Vinaya in both static and dynamic aspects.

One open question is left for the reader and for subsequent papers: if the stable core secures that the three Vinaya-lineages point to the same teaching, and the variable periphery displays that the same teaching can have multiple historical embodiments, does this dual structure replicate at the periphery of other non-bhikkhunī Vinaya domains — upasampadā procedure, uposatha observance, rains-residence institution, kaṭhina-robe regulation, and so on? Does the four-to-one asymmetry repeat, or vary? The answer must wait for further point-by-point cross-recension work in those domains. The present paper's single-axis finding at the garudhamma placement is a first step of that broader comparative project, not the whole of it.

References

Primary Sources — Pāli

Cullavagga X, Bhikkhunī-khandhaka: PTS Vin ii 253–285. Edition used: SuttaCentral CC0 public-domain edition (Bhikkhu Brahmali English, Mahāsaṅgīti Pāli ms), UID pli-tv-kd20.
Cullavagga XI, Pañcasatika-khandhaka (First Council): PTS Vin ii 284–308. Edition used: SuttaCentral CC0, UID pli-tv-kd21.
Suttavibhaṅga, Bhikkhu-vibhaṅga, Pācittiya 72 Vilekhanasikkhāpada: PTS Vin iv 142–144, SuttaCentral UID pli-tv-bu-vb-pc72.
Aṅguttara-nikāya 4.245 Sikkhānisaṃsa-sutta: PTS AN ii 243, SuttaCentral UID an4.245. (cross-paper anchor)
Aṅguttara-nikāya 11.1 Kimatthiya-sutta: PTS AN v 1–2, SuttaCentral UID an11.1. (cross-paper anchor)

Primary Sources — Chinese (CBETA)

T1421: Mahīśāsaka-vinaya (《彌沙塞部和醯五分律》), trans. Buddhajīva (佛陀什) and Zhu Daosheng (竺道生), Liu-Song (423–424). CBETA T22 No. 1421.
T1425: Mahāsāṅghika-vinaya (《摩訶僧祇律》), trans. Buddhabhadra (佛陀跋陀羅) and Faxian (法顯), Eastern Jin (416–418). CBETA T22 No. 1425.
T1428: Dharmaguptaka-vinaya (《四分律》), trans. Buddhayaśas (佛陀耶舍) and Zhu Fonian (竺佛念), Later Qin Hongshi (410–413). CBETA T22 No. 1428.
T1429: Dharmaguptaka-bhikṣu-prātimokṣa (《四分律比丘戒本》), trans. Buddhayaśas, Later Qin. CBETA T22 No. 1429.
T1435: Sarvāstivāda-vinaya (《十誦律》, Daśa-bhāṇavāra), trans. Puṇyatara (弗若多羅) with Kumārajīva (羅什), Dharmaruci (曇摩流支), and Vimalākṣa (卑摩羅叉), Later Qin (404–409). CBETA T23 No. 1435.
T1442: Mūlasarvāstivāda-bhikṣu-vinaya (《根本說一切有部毘奈耶》), trans. Yijing (義淨), Tang (700–712). CBETA T23 No. 1442.
T1443: Mūlasarvāstivāda-bhikṣuṇī-vinaya (《根本說一切有部苾芻尼毘奈耶》), trans. Yijing, Tang (700–712). CBETA T23 No. 1443.
T1450: Mūlasarvāstivāda-saṅghabhedavastu (《根本說一切有部毘奈耶破僧事》), trans. Yijing, Tang. CBETA T24 No. 1450.
T1451: Mūlasarvāstivāda-kṣudrakavastu (《根本說一切有部毘奈耶雜事》), trans. Yijing, Tang (710–711). CBETA T24 No. 1451.
T1558: Abhidharmakośa-bhāṣya (《阿毘達磨俱舍論》), composed by Vasubandhu (世親), trans. Xuanzang (玄奘), Tang. CBETA T29 No. 1558.
T1605: Abhidharma-samuccaya (《大乘阿毘達磨集論》), composed by Asaṅga (無著), trans. Xuanzang, Tang. CBETA T31 No. 1605.
T1606: Abhidharma-samuccaya-vyākhyā (《大乘阿毘達磨雜集論》), composed by Sthiramati (安慧), trans. Xuanzang, Tang. CBETA T31 No. 1606.
T1804: Sifenlü Shanfan Buque Xingshichao (《四分律刪繁補闕行事鈔》) of Daoxuan (道宣), Tang seventh century. CBETA T40 No. 1804.
T1805: Sifenlü xingshichao zichi ji (《四分律行事鈔資持記》) of Yuanzhao, Song (sub-commentary on Daoxuan's Xingshichao). CBETA T40 No. 1805.
T2125: Nanhai jigui neifa zhuan (《南海寄歸內法傳》) of Yijing, Tang late seventh to early eighth century. CBETA T54 No. 2125.

Modern Scholarship

Anālayo, Bhikkhu. The Foundation History of the Nuns' Order. Hamburg Buddhist Studies 6. Hamburg: Hamburg University Press, 2016.
Anālayo, Bhikkhu. "On the Bhikkhunī Ordination Controversy." Sri Lanka International Journal of Buddhist Studies 3 (2014).
Anālayo, Bhikkhu. "The Cullavagga on Bhikkhunī Ordination." Journal of Buddhist Ethics 22 (2015): 1–34.
Anālayo, Bhikkhu. "Theravāda Vinaya and bhikkhunī Ordination." Journal of Buddhist Ethics 24 (2017): 1–32.
Clarke, Shayne. Family Matters in Indian Buddhist Monasticisms. Honolulu: University of Hawai'i Press, 2014.
Clarke, Shayne. "Lost in Tibet, Found in Bhutan: The Unique Nature of the Mūlasarvāstivādin Law Code for Nuns." Buddhism, Law & Society 2 (2016–17): 199–292.
Heirman, Ann. Rules for Nuns according to the Dharmaguptakavinaya. Buddhist Tradition Series 47–49. Delhi: Motilal Banarsidass, 2002, three volumes.
Heirman, Ann. "Where is the Probationer in the Chinese Buddhist Nunneries?" Zeitschrift der Deutschen Morgenländischen Gesellschaft 158:1 (2008): 105–137.
von Hinüber, Oskar. A Handbook of Pāli Literature. Indian Philology and South Asian Studies vol. 2. Berlin and New York: Walter de Gruyter, 1996.
Pachow, W. A Comparative Study of the Prātimokṣa. Sino-Indian Studies V. Santiniketan: Sino-Indian Cultural Society, 1955.
Sujato, Bhikkhu. Bhikkhunī Vibhaṅga: A Translation of the Bhikkhunī Vibhaṅga Section of the Pali Vinaya. SuttaCentral, 2014.

Cross-References

Within-volume back-links (SĪLA earlier papers)

SĪLA-P09 (Part III #1, immediately prior): Architecture of the Pāṭimokkha — A Self-Legislating Casuistic Case Law. The single-tradition architectural reading of the Pāṭimokkha as a self-legislating casuistic case law. The present paper extends that reading from single-tradition architecture to the cross-tradition synchronic comparison at the five-recension array. P09's Suttavibhaṅga-versus-Pāṭimokkha intra-tradition split corresponds, in the present paper, to the T1451-versus-T1443 intra-tradition register-split within Mūlasarvāstivāda (§3.2).
SĪLA-P08 (Part II SEALED): Sīla-gandha — The Fragrance That Travels Against the Wind. AN 4.245 / AN 11.1 cross-paper anchor carry. The outward social-domain framing of sīla at the lay–saṅgha interface is the complement to the inward monastic-internal frame of the Vinaya explored here.
SĪLA-P07 (Part II): Uposatha and the Genius of Intermittent Monasticism. AN 11.1 cross-paper anchor carry; the fortnightly-recitation institutional frame.
SĪLA-P01 (Part I opener): What Sīla Actually Means. The three-axis (Pāli / Sanskrit / Chinese) reading foundation on which the present paper's five-recension comparison rests.

Within-volume forward links (SĪLA later papers)

SĪLA-P11 (Part III #3, immediately next): The Second Council and the Devadatta Episodes. The present paper focuses on synchronic structure; P11 turns to the diachronic narrative axis — the Second Council at Vaiśālī and the Mūlasarvāstivāda–Pāli inversion of the fifth of Devadatta's five demands (T1450 fascicle 10 "dwelling in village dwellings" against the Pāli forest-only requirement).
SĪLA-P12 (Part IV opener): The Threefold Pure Precept-Collection. The three-restraint architecture (saṃvara-śīla / kuśaladharma-saṃgrāhaka-śīla / sattvārthakriyā-śīla); P09's eightfold prātimokṣa-saṃvara and the present paper's cross-recension prātimokṣa stability together form the saṃvara-śīla foundation rung.
SĪLA-P17 (Part V): Daoxuan's Sifenlü Xingshichao Unfolded. The present paper engages Daoxuan only at the cross-recension comparative-method layer; P17 is the venue for Daoxuan's full Vinaya-compendium exegesis.

Cross-volume polyphony anchors

SAMĀDHI volume forward-link: AN 11.1's eleven-link chain from sīla through paññā is acknowledged here as cross-paper anchor; its full developmental treatment is the venue of the SAMĀDHI volume (P24 in particular).
NIAN volume cross-reference: the Pāṭimokkha's recursive self-reference at the recitation event (P09 §V; here §2.1) bears on NIAN's treatment of anussati as a recursive recollective mode; the venue for that cross-mirror is the NIAN volume.

Recurring cross-paper anchors

AN 4.245 Sikkhānisaṃsa-sutta: cross-paper recurrence anchor, carried in SĪLA-P01, P07, P09, and the present paper.
AN 11.1 Kimatthiya-sutta: cross-paper recurrence anchor, carried in SĪLA-P02, P07, P08, P09, and the present paper.

CBETA collation: The three principal Chinese-translation Vinaya anchors of this paper are Mahīśāsaka-vinaya (T1421 Mahīśāsaka · trans. Buddhajīva and Zhu Daosheng · Liu-Song 423–424 · fascicle 29 consolidated garudharma block), Mahāsāṅghika-vinaya (T1425 Mahāsāṅghika · trans. Buddhabhadra and Faxian · Eastern Jin 416–418 · fascicle 30 garudharma + fascicle 32 nine-heading sequence with the ninth rubric dharma-anudharma), Dharmaguptaka-vinaya (T1428 Dharmaguptaka · trans. Buddhayaśas and Zhu Fonian · Later Qin Hongshi 410–413 · fascicles 48–49 Bhikṣuṇī-khandhaka consolidated garudharma block + fascicle 54 dual-council eight-category recitation), Dharmaguptaka-bhikṣu-prātimokṣa (T1429 Dharmaguptaka Prātimokṣa · Later Qin · the 半月半月說、戒經中來 framing-formula at nine occurrences), Sarvāstivāda-vinaya (T1435 Sarvāstivāda · trans. Puṇyatara with Kumārajīva · Later Qin 404–409 · fascicles 42–47 Bhikṣuṇī-vinaya doubled consolidation), Mūlasarvāstivāda-bhikṣu-vinaya (T1442 Bhikṣu-vinaya · trans. Yijing · Tang 700–712 · rule-form register), Mūlasarvāstivāda-bhikṣuṇī-vinaya (T1443 Bhikṣuṇī-vinaya · trans. Yijing · fascicle 19 śikṣāpada 126–129 dispersed-rule pāyantika numbering · dispersed-rule principal anchor), Mūlasarvāstivāda-saṅghabhedavastu (T1450 · trans. Yijing), and Mūlasarvāstivāda-kṣudrakavastu (T1451 · trans. Yijing · fascicles 29–30 garudharma narrative form · narrative-register principal anchor · corrected from fascicle 40). All citations have been collated against the CBETA Electronic Tripiṭaka Collection 2024. The Chinese-commentary anchors Sifenlü Shanfan Buque Xingshichao (T1804 · Daoxuan · seventh-century Tang · fascicle 1 pt1 "six authentic recensions" classification + "six kinds of Vinaya-holders in the world" typology), Sifenlü xingshichao zichi ji (T1805 · Yuanzhao · Song · sub-commentary on Daoxuan's Xingshichao · "Mahīśāsaka–Daśa-bhāṇavāra general agreement" + Mingliao lun time-and-place + "these two characters refer to the two schools" word-level cross-recension collation · extending Daoxuan's comparison · NB T40n1805 = Yuanzhao's Zichiji, not Daoxuan's Hanzhu jieben shu [the latter is X0714]), and Nanhai jigui neifa zhuan (T2125 · Yijing · Tang 691–712 · collated against the CBETA XML edition (T54n2125) · the three indigenous-awareness segments: geographical distribution of four Vinaya recensions in Tang China, Yijing's self-disclosed Mūlasarvāstivāda standpoint with the Daśa-bhāṇavāra-disambiguation, and the "tangled and disorderly" indigenous critique) have also been collated against CBETA. The Pāli principal anchors — Cullavagga XI's First Council eight-category Pāṭimokkha recitation and Mahākassapa's apaññattaṁ nappaññapeyya paññattaṁ na samucchindeyya transmission discipline; the Suttavibhaṅga Pācittiya 72 Vilekhanasikkhāpada recursive self-reference (P09 carry) — are cited by PTS reference, with the SuttaCentral Brahmali/Sujato edition of the Mahāsaṅgīti recension (CC0) as the reading edition. The Cullavagga X Bhikkhunī-khandhaka eight garudhammas Pāli baseline is a segment, supplied from SuttaCentral CC0 (UID pli-tv-kd20 segments 1.4.1–1.4.21 · PTS Vin ii 253–258 · Brahmali CC0 English + Mahāsaṅgīti CC0 Pāli). The Aṅguttara-nikāya 4.245 Sikkhānisaṃsa-sutta and 11.1 Kimatthiya-sutta serve as cross-paper recurrence anchors (carried in SĪLA-P01 / P02 / P07 / P08 / P09). The śāstra-stratum Abhidharmakośa (T1558 fascicle 14), Abhidharma-samuccaya (T1605), and Abhidharma-samuccaya-vyākhyā (T1606) appear here as P09 carry only and are not expanded in the present paper.

指月 Six Practice Gates · SĪLA Volume · Paper P10 · May 2026

Author: 釋慧鏡 (Shi Huijing) Project: 指月 Point-to-the-moon · 六行門 Six Practice Gates License: CC BY-NC-SA 4.0 Repository: github.com/pointing-at-the-moon/six-practice-gates

Footnotes

Yijing, Nanhai jigui neifa zhuan (《南海寄歸內法傳》), fascicle 1, Taishō T54 No. 2125, T54 p.0205c20–0206a04. The two key phrases of Yijing's critique are 神州持律諸部互牽 ("the various Vinaya-holding schools in this divine land were pulling on each other") and 章鈔繁雜 ("[their] commentaries and excerpts had become tangled and disorderly"). Citations verified against the CBETA Electronic Tripiṭaka Collection. ↩
Daoxuan, Sifenlü Shanfan Buque Xingshichao (《四分律刪繁補闕行事鈔》), fascicle 1, Taishō T40 No. 1804. The six-recension list at fascicle 1 upper, line 354; the "six kinds of Vinaya-holders in the world" typology at fascicle 1 upper, lines 261–264; the "texts list four readings; the Daśa-bhāṇavāra is in seal-print agreement in meaning" passage at fascicle 1 upper, lines 247–253. ↩
Shi Huijing, "Architecture of the Pāṭimokkha: A Self-Legislating Casuistic Case Law", Six Practice Gates series, SĪLA-P09, 2026, github.com/pointing-at-the-moon/six-practice-gates. ↩
Pāli Cullavagga XI, the First Council scene of Upāli's recitation of the Pāṭimokkha in eight categories: PTS Vin ii 287, SuttaCentral UID pli-tv-kd21, Brahmali editorial recension of the Mahāsaṅgīti edition (CC0). ↩
Cullavagga XI, PTS Vin ii 288, SuttaCentral UID pli-tv-kd21. The Pāli runs: apaññattaṁ nappaññapeyya paññattaṁ na samucchindeyya yathāpaññattesu sikkhāpadesu samādāya vatteyya. ↩
Bhikkhu-pāṭimokkha, Pācittiya 72, Vilekhanasikkhāpada: PTS Vin iv 142, SuttaCentral UID pli-tv-bu-vb-pc72. The recursive self-reference of the Pāṭimokkha to its own recitation event (pātimokkhe uddissamāne) is the structural attestation; the key term khuddānukhuddaka appears in this rule and is shared with the Buddha's final allowance at Dīgha-nikāya 16 §6.3 (P09 carry). ↩
Pāli Cullavagga X, Bhikkhunī-khandhaka: PTS Vin ii 253–258. The passage is cited here from the SuttaCentral CC0 edition (Bhikkhu Brahmali English, Mahāsaṅgīti Pāli ms), SuttaCentral UID pli-tv-kd20 segments 1.4.1–1.4.21. The Cullavagga X material is absent from the Pāli vault and was retrieved via the SuttaCentral CC0 fallback. ↩
Dharmaguptaka-vinaya (《四分律》), trans. Buddhayaśas and Zhu Fonian, Later Qin Hongshi 12–15 (410–413 CE), fascicle 1 (序分): CBETA Taishō T22 No. 1428. The dasa atthavase paṭicca / 十句義 formulation of the rationale for laying down the precepts appears at the opening of fascicle 1. ↩
Dharmaguptaka-vinaya fascicle 54, T22 No. 1428, T22n1428_p0966–p0971. Upāli's First Council recitation in eight categories at lines 1–290; Mahākāśyapa's Second Council ruling on the "lesser and minor" rules at lines 291–560. (Note: an earlier project document mistakenly placed the Second Council material at fascicle 60; in fact both councils sit in fascicle 54, the second immediately following the first.) ↩
Dharmaguptaka-vinaya fascicle 48, Bhikṣuṇī-khandhaka consolidated garudharma block: CBETA T22 No. 1428, T22n1428_p0922b–c. Each rule closed by the formula "this rule is to be respected, revered, and praised; it is not to be transgressed for as long as life endures"; block sealed by the five-hundred-year saddhamma prediction. ↩
Dharmaguptaka-bhikṣu-prātimokṣa (《四分律比丘戒本》), trans. Buddhayaśas: CBETA T22 No. 1429. The framing-formula 半月半月說、戒經中來 ("fortnight by fortnight to be recited, drawn from the Bhikṣu-prātimokṣa") occurs nine times across the text, covering all eight categories. Pāli parallel: imāni kho panāyasmanto cattāro pārājikā dhammā uddesaṃ āgacchanti ("venerable ones, these four pārājika rules come up for recitation in the Pāṭimokkha"). ↩
Mūlasarvāstivāda-Kṣudrakavastu (《根本說一切有部毘奈耶雜事》, T1451), trans. Yijing, Tang (710–711), fascicle 29 (end) through fascicle 30 (opening): CBETA T24 No. 1451, T24n1451_p0350b–p0351b. The passage contains Mahāprajāpatī's three requests, the Buddha's three hesitations with the three similes (household with few men / hailstone-struck crop / sugarcane with red joints), Ānanda's intercession, the eight garudhammas each addressed with "Ānanda!", and the marriage-garland ordination posture. (Project correction: an earlier note placed this narrative at fascicle 40; fascicle 40 is in fact Second Council material that belongs to P11.) ↩
Kṣudrakavastu (T1451) fascicle 1 opening: the text names itself as kṣudraka-vastu (《雜事》) and announces its organisation by uddāna-summary topical clusters rather than by numbered śikṣāpada. CBETA T24 No. 1451. ↩
Mūlasarvāstivāda-bhikṣuṇī-vinaya (《根本說一切有部苾芻尼毘奈耶》, T1443), trans. Yijing, Tang (700–712), fascicle 19: CBETA T23 No. 1443. Śikṣāpada 126 "rule on requesting instruction" corresponds to Pāli garudhamma 3; śikṣāpada 127 "rule on holding the uposatha where there are no bhikṣus" corresponds to the rule that uposatha must be performed in the presence of bhikṣus; śikṣāpada 128 "rule on entering the rains-retreat where there are no bhikṣus" corresponds to Pāli garudhamma 2; śikṣāpada 129 "rule on not performing the pravāraṇā three observances at the joint assembly" corresponds to Pāli garudhamma 4. Each is closed with the pāyantika (波逸底迦) category-marker and embedded in the continuous pāyantika numbering of fascicles 15–20, which cover śikṣāpada 38 through 180 and beyond. ↩
Mūlasarvāstivāda-bhikṣuṇī-vinaya (T1443) fascicle 1, Vinaya-preface (毘奈耶序), line 99. The preface, in uddāna-summary form, announces the bhikṣuṇī-code as "prātimokṣa-sūtra" (別解脫經) and identifies the "eight pārājika rules" as the opening marker of the bhikṣuṇī-code. The bhikṣuṇī eight pārājikas (against the bhikṣu four) is a cross-recension stable feature of the nuns' code. ↩
Mahāsāṅghika-vinaya (《摩訶僧祇律》, T1425), trans. Buddhabhadra and Faxian, Eastern Jin (416–418), fascicle 32: CBETA T22 No. 1425. Fascicle 32 explicitly establishes a "nine-heading sequence" (九法序), adding the ninth heading dharma-anudharma (dhamma-anudhamma, 法隨順法) beyond the shared eight-category base. The bhikṣuṇī-rules consolidated block sits in fascicle 30 opening (line 359 onward), in the strongest consolidated form of the five recensions: each rule glossed with an "exceeding-of-gurudharma" (越敬法) explanation, and sealed with the formula "the Nth gurudharma ends". ↩
Sarvāstivāda-vinaya (《十誦律》, T1435), trans. Puṇyatara with Kumārajīva, Dharmaruci, and Vimalākṣa, Later Qin (404–409): CBETA T23 No. 1435, fascicles 42–47. The eight garudhammas are placed within the bhikṣuṇī-section and, additionally, in an independent garudharma chapter at fascicle 47 line 523. The framing is the fortnightly-instruction (ardhamāsa-uddesa) ritual rather than the ordination narrative — a sub-pattern distinct from the four other consolidated recensions. ↩
Anālayo, Bhikkhu. The Foundation History of the Nuns' Order. Hamburg Buddhist Studies 6. Hamburg: Hamburg University Press, 2016. See also Anālayo's series of articles on the eight garudhammas and bhikkhunī-rules in the Journal of Buddhist Ethics vols. 21 (2014), 22 (2015), and 24 (2017). ↩
Heirman, Ann. Rules for Nuns according to the Dharmaguptakavinaya. Buddhist Tradition Series 47–49. Delhi: Motilal Banarsidass, 2002, three volumes. See also Heirman, "Where is the Probationer in the Chinese Buddhist Nunneries?", Zeitschrift der Deutschen Morgenländischen Gesellschaft 158:1 (2008), pp. 105–137. ↩
Clarke, Shayne. Family Matters in Indian Buddhist Monasticisms. Honolulu: University of Hawai'i Press, 2014. See also Clarke, "Lost in Tibet, Found in Bhutan: The Unique Nature of the Mūlasarvāstivādin Law Code for Nuns", Buddhism, Law & Society 2 (2016–17), pp. 199–292. ↩
Bhikkhu Sujato is the principal contributor of English translations of the Pāli Vinaya on SuttaCentral and the author of the Bhikkhunī-vibhaṅga introduction. Oskar von Hinüber, A Handbook of Pāli Literature (Berlin: de Gruyter, 1996), §31–43, on the redactional stratigraphy of the Vinaya-piṭaka. W. Pachow, A Comparative Study of the Prātimokṣa, Sino-Indian Studies V (Santiniketan: Sino-Indian Cultural Society, 1955) — the foundational figure of all modern cross-recension Prātimokṣa comparison. ↩
As n. 2. The Xingshichao fascicle 1 upper, line 354, "six authentic recensions" enumeration: Saṃghika / Tánwúdé / Sarvāstivāda / Mīshāsai / Jiāyèyí / Pólúfúluó. ↩
Yuanzhao, Sifenlü xingshichao zichi ji (《四分律行事鈔資持記》, T1805), a Song sub-commentary on Daoxuan's Xingshichao, Taishō T40 No. 1805. The "Mahīśāsaka and Daśa-bhāṇavāra general agreement" (五分十誦大同) passage at fascicle 1 lines 156–162; the Mingliao lun (《明了論》) on time-and-place categories of Vinaya at lines 991–1003; the "these two characters refer to the two schools" passage at lines 1007–1016. Note: Taishō 40 No. 1805 is by Yuanzhao (Song), NOT Daoxuan; Daoxuan's own Hanzhu jieben shu (戒疏) is Manji Zokuzōkyō X0714. These word-level collation passages are Yuanzhao's, glossing Daoxuan's Xingshichao and extending the cross-recension comparison Daoxuan opened. ↩
As n. 1. Nanhai jigui T54 p.0205b26–c04: geographical distribution of the four Vinaya recensions in Tang China — northern regions principally Dharmaguptaka, Jiangnan southern regions a mixture. ↩
As n. 1. Nanhai jigui T54 p.0206b28–c04: Yijing's self-disclosure that his translations follow the Mūlasarvāstivāda Vinaya; explicit correction of the common Chinese confusion "the Daśa-bhāṇavāra is the Mūlasarvāstivāda Vinaya"; and the historical-philological framing "the Sarvāstivāda group divides into three: Dharmaguptaka, Mahīśāsaka, Kāśyapīya". ↩

Source on GitHub ↗