Voice of 

the Buddha
 Voice of
the Buddha
Buddhaghosa on
the Immeasurable Words

MARIA HEIM

1
 1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.

Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.

© Oxford University Press 2018

All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.

You must not circulate this work in any other form

and you must impose this same condition on any acquirer.

Library of Congress Cataloging-​in-​Publication Data

Names: Heim, Maria, 1969– author.
Title: Voice of the Buddha : Buddhaghosa on the immeasurable words / Maria Heim.
Description: New York, NY : Oxford University Press, [2018] |
Includes bibliographical references and index.
Identifiers: LCCN 2018016558 (print) | LCCN 2018036146 (ebook) |
ISBN 9780190906665 (updf ) | ISBN 9780190906672 (epub) |
ISBN 9780190906689 (online content) | ISBN 9780190906658 (cloth)
Subjects: LCSH: Buddhaghosa. | Theravāda Buddhism.
Classification: LCC BQ366 (ebook) | LCC BQ366 .H45 2018 (print) |
DDC 294.3/91092—dc23
LC record available at https://lccn.loc.gov/2018016558

1 3 5 7 9 8 6 4 2
Printed by Sheridan Books, Inc., United States of America
For Soren
 Contents

Acknowledgments  ix
Abbreviations for Pali Texts  xi

Introduction 1

PART I :  Building Blocks for an Interpretative Program

1. The Buddha’s Omniscience and the Immeasurability of Scripture 33

2. Scripture, Commentary, and Exegetical Distinctions 60

PART II :  Interpreting the Three Piṭakas

3. The Contexts and Conditions of Buddhavacana in the Suttanta 109

4. Disentangling the Tangle: Abhidhamma as Phenomenological Analysis 144

5. The “Completely Pleasing” Exegesis on the Vinaya 184

Conclusion 218

Appendix A: The Recollection of the Dhamma  223

Appendix B: Commentary on the Section on Verañja Starting the Vinaya  229
Appendix C: Four Oceans and Three Piṭakas  239
Bibliography  253
Index  265
 Acknowledgments

I am grateful to many people who have helped in large and small ways
with this project. In many respects the book is the outcome of reading practices
I  began to learn with Charles Hallisey while studying Pali commentary over
twenty years ago at Harvard. I remain grateful to Charlie for his conversation
and mentorship ever since, and for reading and commenting on an early version
of the manuscript. The book is better also for his inviting me to join, via Skype,
his “Readings in Pali Commentary” class at Harvard Divinity School for the aca-
demic year 2016–​2017; thanks to him and the students of that course for sharing
this experience with me.
I am also very grateful to Chakravarthi Ram-​Prasad for commenting metic-
ulously on the entire manuscript, and for the many conversations over the years
we’ve had about Buddhaghosa and Indian thought more broadly. I  appreciate
also his student Gilda Darlas for many wonderful exchanges about Buddhaghosa
and her ideas for his practical applications.
I am blessed with brilliant and supportive colleagues at the Five Colleges who
have made this community such a stimulating intellectual home. I am particu-
larly indebted to Andy Rotman for coordinating our faculty seminar, and to the
participants of the discussion of the Abhidhamma chapter in late 2016:  Andy
Rotman, Jay Garfield, Steve Heim, Sandy Huntington, Reiko Ohnuma, William
Edelglass, and Andrew Olendzki. Jay Garfield has been particularly supportive
and helpful in reading drafts of my work.
Other portions of this project have benefited from being presented at confer-
ences and workshops. I particularly appreciate the participants of my 2016 work-
shop at Amherst College on Buddhaghosa:  Rupert Gethin, Charles Hallisey,
Chakravarthi Ram-​Prasad, Janet Gyatso, Alastair Gornall, Aleix Ruiz-​Falqués,
Tari Shulman, and Richard Nance. Many thanks to the Hamilton Fund and the
Religion Department for supporting the workshop, and to Lisa Ballou for her
able administrative assistance. I am grateful also to Jack Petranker for inviting me
to two workshops at the Mangala Research Center for Buddhist Languages, one
x Acknowledgments

on Abhidhamma/​Abhidharma traditions and the other on the work on Pierre

Hadot, both of which were helpful in the early stages of my thinking for this book.
I learned a lot from the panel put together by Rafal Stepien for the American
Comparative Literature conference, where I was able to discuss some of the ideas
contained herein. I  thank Asanga Tilakaratne and the Sri Lanka Association
of Buddhist Studies for inviting me to the “Theravāda and Buddhaghosa” con-
ference in 2016, from which I learned much. Thanks very much also to Vanessa
Sasson for reading the whole manuscript and her enthusiastic support of it, to
Natalie Gummer for the many conversations over the years we have had about
texts, and to Margaret Cone for her helpful replies to my queries. Two anon-
ymous reviewers for Oxford University Press offered sympathetic readings and
careful critiques of the manuscript, which helped me produce a better book.
I owe many thanks to Amherst College for honoring me with the Elizabeth
W. Bruss Readership, for the sabbatical support in spring 2017 that allowed me
to finish the book, and in general for providing me a supportive and congenial
workplace.
And, as always, I thank Steve for reading the whole manuscript and for his
constant help and support of my career. And my deep love and gratitude go to
him, Soren, and Zack for being who they are and for forbearing this obsession
with Buddhaghosa with grace and good humor.
 Abbreviations for Pali Texts

A Aṅguttaranikāya
As Atthasālinī (Dhammasaṅgaṇī-​aṭṭhakathā)
D Dīghanikāya
Dhp Dhammapada
Dhp-​a Dhammapada-​aṭṭhakathā
Dhs Dhammasaṅgaṇī
Ja Jātaka-​aṭṭhakathā
Khp Khuddakapāṭha
Kv Kathāvatthu
Kv-​a Kathāvatthu-​aṭṭhakathā
Mil Milindapañha
Mp Manorathapūraṇi (Aṅguttaranikāya-​aṭṭhakathā)
Netti Nettippakaraṇa
Paṭis Paṭisambhidāmagga
Paṭis-​a Paṭisambhidāmagga-​aṭṭhakathā
Pj I Paramatthajotikā I (Khuddakapāṭha-​aṭṭhakathā)
Ps Papañcasūdanī (Majjhimanikāya-​aṭṭhakathā)
Pv-​a Paramatthadīpanī (Petavatthu-​aṭṭhakathā)
S Saṃyuttanikāya
Sp Samantapāsādikā (Vinaya-​aṭṭhakathā)
Spk Sāratthappakāsinī (Saṃyuttanikāya-​aṭṭhakathā)
Sv Sumaṅgalavilāsinī (Dīghanikāya-​aṭṭhakathā)
Vibh Vibhaṅga
Vibh-​a Sammohavinodanī (Vibhaṅga-​aṭṭhakathā)
Vism Visuddhimagga
Vism-​mhṭ Visuddhimagga-​mahāṭīkā
 Introduction

According to chronicle and legend, there was once a precocious

Brahmin scholar, deeply learned in the Vedic and grammatical systems of ancient
India and renowned for his skills in debate. In his wanderings he encounters a
Buddhist monk who, spotting the scholar’s talent, recites his own knowledge of
the Abhidhamma, one of the three branches of Buddhist scripture. Fascinated,
the scholar asks to acquire this knowledge but is told he must ordain as a Buddhist
monk to receive it. Out of pure curiosity, he does so, and is then permitted to
study all three branches of Buddhist scripture, whereupon he is said to have
become a committed believer. His subsequent contributions to advancing
Buddhist knowledge earn him the monastic title Buddhaghosa, “Voice of the
Buddha,” “because his speech was profound” (or “his voice was deep”) like that of
the Buddha, with the aspiration that his words, like those of the Buddha himself,
“might be voiced over the surface of the earth.”1
From India, Buddhaghosa is sent by his teachers to the island of Lanka to take
the ancient Sinhala commentary on the scriptures and produce it in Māgadha,
the language we know today as Pali. He stays at the Mahāvihāra monastery and
studies the ancient commentaries, but before he is given leave by the monastic
authorities to edit and translate the commentarial corpus, he is asked to produce a
commentary on two stanzas of the Buddha’s words that might serve as a summary
of all three branches of scripture. He produces the magisterial handbook, the
Path of Purification (Visuddhimagga), which he reads out under the Bodhi Tree

1. This summary is taken from Bhikkhu Ñāṇamoli’s translation of the passages on Buddhaghosa
in the Mahāvaṃsa, The Path of Purification, xxxiv–​xxxv. See also Geiger, trans., Cūlavaṃsa,
ch. 37, vv. 215–​47, and Gray’s translation of the thirteenth-​century legend of Buddhaghosa
called Buddhaghosuppatti; Finot, “The Legend of Buddhaghosa”; Law, Buddhaghosa;
Buddhadatta, “Who Was Buddhaghosa?” 142–​57. I offer an extensive annotated bibliography
on Buddhaghosa in the Oxford Bibliographies (Heim, “Visuddhimagga/​Buddhaghosa”).
2 In t rodu ct ion

on the precincts of the ancient monastery at Anuradhapura. Yet he is subjected

to a further trial: the manuscript of this huge text vanishes (the gods steal it), and
he must reproduce it. And a second time, it is lost, and again reproduced. The
gods return the first two manuscripts, and all three are read together and found
to be identical in every word and phrase. For his mastery and unerring consist-
ency, Buddhaghosa is celebrated as “the future Buddha Metteyya himself,” and
entrusted with editing and translating the commentaries, whereupon his work
comes to be accepted by the elders of the tradition “as equal in authority with the
canonical texts themselves.”2
Such legends have texts and textual practice at their center, and their hero
is a master of unremitting curiosity and erudition. While we probably cannot
infer from these legends what exactly it was that piqued the fifth-​century CE
scholar’s curiosity about the Abhidhamma (to the extent that he was willing
to ordain as a monk to get access to it), or how it was that he was allegedly
capable of reproducing thrice over and verbatim the nearly eight-​hundred-​page
Visuddhimagga, we can learn from the body of textual work that he produced
how he thought about his scholarly project. In fact, when we turn to the prodi-
gious intellectual achievement traditionally attributed to him, we find him very
forthcoming about what he took scripture and textual practice to be. He tells us,
in various ways, that he finds the Buddha’s words beautiful, highly pleasurable,
universally applicable, and highly transformative. Further, the remarkable claims
in the legends that liken Buddhaghosa’s voice to that of the Buddha, that find his
work to be as authoritative as that of the canonical sources, and that identify him
with the future Buddha Metteyya suggest a reading of commentary as an echo
or extension of the Buddha’s words. These assertions find support in the reading
practices Buddhaghosa discusses and demonstrates. For him, the Buddha’s words
are “immeasurable” and the practices they generate are the ongoing unfolding
of the Buddha’s omniscient ken. Commentary is textual practice that extends
scripture.
This book attempts to find out how this may be so and what is required of us to
come to see this. The pages and chapters that follow explore Buddhaghosa’s pro-
grammatic accounts of scriptural interpretation to discern the nature and scope of
his hermeneutics and the contours of his intellectual project as a whole. In these
pages I attempt to read with Buddhaghosa to see how he encountered Buddhist
scripture; the result is a “new” (new for us moderns, anyway) reading of the
Buddha’s words and the possibilities for what scholarly commentary and learning
practices can be. I  suggest two key points. First, I  explore how Buddhaghosa

2. Ñāṇamoli, The Path of Purification, xxxv, a translation of the Mahavāṃsa, ch. 37, vv. 215–​47.
 Introduction 3

found in Buddhist texts not totality, but infinity. That is, for Buddhaghosa, any
particular expression in the Buddhist canon cannot be exhaustively described,
and the interpretive task is to show the “immeasurability” of the Buddha’s words
and to shape the ideal reader’s response to it. Second and relatedly, Buddhaghosa’s
interpretative assumptions often treat the Buddha’s teachings not as declarative
or discursive utterances so much as practices. The Buddha’s words were uttered
not so much to give final propositional accounts of the way things are, but to in-
itiate a series of practices that are themselves the very work of insight and under-
standing that enact the tradition’s therapeutic and soteriological aims.

Disciplinary Philology and Pali Theories of Texts

As an interpreter, Buddhaghosa was explicit and systematic about the tools and
practices he deployed, and we can suggest that his remarks about them, as well as
what we can understand from observing his practices, amount to a theory of scrip-
ture and exegesis. One way to identify Buddhaghosa’s theory, and its implications
for reading canon and commentary, is to situate them in a larger framework of
disciplinary philology. We can see Buddhaghosa practicing a philology that ad-
vanced a theory of scripture and the interpretative project. I have found it useful
to locate my understanding of this project in terms of Sheldon Pollock’s concep-
tion of “future philology.”
In his penetrating and far-​reaching vision for a revitalized philology that is
critical, disciplined, and reflexive, Sheldon Pollock urges that we recover “the
initiatives, theories, methods, and insights of scholars across time and across the
world in making sense of texts.” 3 As he frames it, in its most expansive form phi-
lology is the “discipline of making sense of texts,”4 and concerns “the practices
of reading as such.”5 In addition, philological practice, whether past or present,
involves implicit or explicit theories about how texts make meaning. Pollock
argues that such theoretical accounts, including those found in the ancient texts
we study, should be objects of inquiry for modern textual scholars. Indeed, philo-
logical theory and practice must be comparative and globalized if we are to make
sense of texts from worlds far removed in space and time from us. A broader com-
parative perspective can help us recognize that historical-​critical philology is not
a universal practice but rather is underpinned by theoretical assumptions devel-
oped in modern Europe. In a truly globalized philology, then, it is not enough to

3. Pollock, “Future Philology? The Fate of a Soft Science in a Hard World,” 949.

4. Pollock, “Introduction” to World Philology, 22.
5. Pollock, “Introduction,” 20.
4 In t rodu ct ion

bring modern hermeneutic assumptions, theories, and schemas to encounter and

interpret texts from the distant past; rather our imaginations and practices may
also be informed by the ways that texts and reading practices were conceived by
authors within earlier text communities. Such work can help historicize our own
assumptions about what texts are and how they make meaning, even as they give
us new tools and possibilities for learning from the past. Perhaps most provoca-
tively they may give us “new” texts in suggesting fresh ways to encounter texts we
might think we already understand.
One substantial piece of this much broader program involves turning to
commentaries for their philological views, that is, what they thought was neces-
sary for the discipline of making sense of texts.6 How did ancient commentators
conceive of the texts on which they commented and the compositions and
practices they themselves were involved in? Commentaries sometimes tell us a
great deal about their hermeneutical theories and practices as they articulate their
particular conceptions of meaning, genre, and discourse. Scholars of premodern
Indian scriptural commentary from various genres can turn to such commentaries
as objects of reflexive philological deliberation, asking how they conceive of the
nature of meaning, genre, transmission, and reception of the texts they discuss.7
The Pali textual tradition provides ample resources for this kind of critical phi-
lology. In addition to providing multiple layers of scholarly tradition and reflec-
tion on texts, its exegetes advanced in systematic fashion formal hermeneutical
principles and sophisticated and explicit discussions of how texts make meaning.
Even at the canonical level, features of the Buddha’s words, registers of speech,
styles of discourse, and contexts for understanding were noticed and commented
on in the sutta and vinaya corpuses, and certainly exegetical practice (vibhaṅga,
niddesa, veyyākaraṇa) was present and recognized as such in the canonical mate-
rial; even the canonical Abhidhamma is seen, in some important sense, as a body
of exegetical work. As time went by, many exegetical protocols were formalized,
most obviously in two hermeneutical handbooks, the Nettippakaraṇa and the

6. Pollock, “Future Philology?” 934.

7.  Pollock, “Future Philology?” 949. For attention to such philological practice in Sanskrit
commentaries, see Pollock’s article on literary and Vedic commentaries in his edited volume
World Philology (114–​36). See Kahrs, Indian Semantic Analysis, for a study of the nirvacana
analysis in Sanskrit commentarial texts in a way that takes seriously “indigenous concepts and
thought patterns” (8). In Buddhist studies, see Cox’s article on commentaries functioning as a
“mediating hermeneutics” that becomes a “necessary lens” where the “ ‘text’ becomes the vision
produced through their ordered conception” (“The Unbroken Treatise,” 149). See also Nance’s
work on commentators among the north Indian traditions (Speaking for Buddhas: Scriptural
Commentary in Indian Buddhism and “Mindsets and Commentarial Conventions among
Indian Buddhists,” 210–​35.)
 Introduction 5

Peṭakopadesa. They were developed further by the aṭṭhakathā commentarial tra-

dition, advanced by Buddhaghosa (or rather, to be more precise, the tradition
he represents8) and Dhammapāla.9 Indeed, the layer of aṭṭhakathā—​a genre of
commentary unique to the Pali tradition—​offers one of the most hermeneuti-
cally explicit treatments of an encounter with textual knowledge in all of early
Indic literature.
Buddhaghosa is most explicit about textual practice in his introductory
sections (nidānas) of each of the three collections of canonical material preserved
in the Pali tradition—​the Suttas, the Vinaya, and the Abhidhamma—​and these
constitute the core of my investigations. Here he provides the aṭṭhakathā layer’s
most thoroughgoing reflections on philological practice. The usual commentarial
preoccupation with sentence-​level exegesis is set aside as he raises his eyes from
words, sentences, and passages to consider larger units of texts, including entire
genres vis-​à-​vis other genres, and the Buddha’s words as a whole. These nidānas
to each of the three corpuses occur at the beginning of each piṭaka: at the start of
the Sumaṅgalavilāsiṇī (Sv, the commentary on the Dīgha Nikāya [D]‌, the first
book of the Suttanta); the introduction to the Atthasālinī (As, the commentary
on the Dhammasaṅgaṇī [Dhs], the first book of the Abhidhamma); and the
Samantapāsādikā (Sp) on the Vinaya (Vin).10 Intertextuality is everywhere in this

8. Buddhaghosa was an editor and translator of many of the aṭṭhakathā; the extent to which his
contributions to them were his own, or even which of the commentaries attributed to him are
his, is unclear. I discuss this issue later.
9. Dhammapāla is usually dated to the sixth century and describes himself as living in south
India, but there is some question, given the large body of works ascribed to him, of whether
there were two Dhammapālas, one who wrote aṭṭhakathās (on most of the books of the
Khuddaka-​nikāya) in the sixth century, and the other who wrote ṭīkās (on the Visuddhimagga
and on the commentaries on the Dīgha, Majjhima, Saṃyutta, Jātaka, Budddhavaṃsa, and
Nettippakaraṇa) no later than the tenth century (see Warder, “Some Problems of the Later
Pali Literature,” 198–​202); Norman, following the nineteenth-​century Gandhavaṃsa, suggests
the Dhammapāla who wrote the Khuddaka aṭṭhakathās was also responsible for the ṭīkā on
the Visuddhimagga (Pāli Literature, 133–​34, 148–​49). De Silva argues that the author of the
ṭīkās and the aṭṭhakathās was likely the same Dhammapāla (De Silva, “Introduction” to her
edition of Dīghanikāya-​aṭṭhakathā-​tīkā Līnatthavaṇṇanā). See also Cousins, “Dhammapāla
and the Tīkā Literature,” 159–​65. There is also a sub-​subcommentary (anuṭīkā) on the entire
Abhidhamma ascribed to a Dhammapāla.
10.  For a translation of parts of this first part of Sv, see Bodhi, The All-​Embracing Net of
Views:  The Brahmajāla and Its Commentaries; for a translation of As, see Pe Maung Tin,
trans., The Expositor; for the “external introduction” of Sp, see Jayawickrama, The Inception of
the Discipline and the Vinaya Nidāna (the rest of the Pali Vinaya commentary has not been
translated, but there is an English translation from a Chinese version of the Samantapāsādikā
in Bapat and Hirakawa, trans., Shan-​Chien-​P’i-​P’o-​Sha). Some of this material is also present in
the Paramatthajotikā I as well (see Ñāṇamoli, Minor Readings and Illustrator, for a translation
and discussion of this text and its exegetical reflections).
6 In t rodu ct ion

material, and these introductions share large portions of text with one another,
with some changes of emphasis according to the distinctive qualities of each of
the three branches of Buddhist scriptural knowledge they comment on; together
they constitute the conceptual framework for the entire body of knowledge on
which the commentaries provide exegesis. In these sections Buddhaghosa speaks
to how texts, “genres” (in the sense of piṭaka), and Buddhist knowledge as a
whole, are to be construed. He describes and theorizes genre, registers of dis-
course, “readerly” response, the nature of scripture and its transmission, and how
various types of Buddhist knowledge and pedagogy work.
The theories of Buddhist texts that Buddhaghosa puts forward in his nidānas
are programmatic for his exegetical methods, and he remains faithful to them in
commenting on particular texts and in the Visuddhimagga. The disciplinary phil-
ological tasks they describe become the main hermeneutical lens of his project.
Specifically, he took each genre of scripture (piṭaka) to be a type of “method”
(naya), and so his commentaries on particular scriptural texts, whether sutta,
Abhidhamma list, or vinaya rule, explore in a dynamic fashion the methods of
thought they make possible. The Buddha’s words were preserved in three piṭakas
(we can call them “genres” in Ricoeur’s sense of “generative discourses,” as I dis-
cuss in ­chapter 2), that are seen by Buddhaghosa as distinctive modes of thought
or methods that require training necessary for receiving the Buddha’s “well-​
spoken words.” Each piṭaka is not only a collection of teachings organizing the
discourses that the Buddha taught (Suttanta), the elaboration of the monastic
rules (Vinaya), and the higher, expanded Dhamma (Abhidhamma), but each is
also a method (naya) and area of expertise (pariyatti).11
These methods require different skills to interpret. Suttanta knowledge is dia-
logical and contextual, often given in a conventional and accessible idiom and to
people according to their various and particular inclinations.12 As we come to see,
a sutta is always embedded in a narrative context that Buddhaghosa deemed es-
sential for interpreting its doctrinal teachings. Buddhaghosa draws a distinction

11. As 11: “Which is the ocean of method? The Buddha’s words [that are] the three piṭakas”
(Katamo nayasāgaro? Tepiṭakaṃ buddhavacanaṃ). As 19; Sv i.18; Sp i.20:  “People learned
about the meaning of ‘piṭaka’ refer to piṭaka in the sense of an area of expertise and basket”
(Piṭakaṃ piṭakatthavidū pariyattibbhājanatthato āhu). All abbreviations are standard for the
Pali Text Society and can be found on the Abbreviations page. All translations are my own
unless otherwise noted, and are from the editions of the Chaṭṭha Saṅgāyana.
12. As 21; Sv i.19; Sp i.21: “The Suttanta Piṭaka is a colloquial teaching because of being taught by
the Bhagavan who is skilled in the colloquial,[and] because it has a preponderance of the col-
loquial” (suttantapiṭakaṃ vohārakusalena bhagavatā vohārabāhullato desitattā vohāradesanā);
and it is given “according to beings’ various inclinations, dispositions, behaviors, and intentions”
(anekajjhāsayānusayacariyādhimuttikā sattā yathānulomaṃ).
 Introduction 7

between narrative Suttanta discourse and Abhidhamma discourse, which he saw

as abstracted and technical, with its own specialized idioms and use of language
in its “furthest sense.”13 To some degree, Abhidhamma knowledge is decontex-
tualized, though this is more nuanced in actual practice than it might initially
appear. Finally, the injunctives and monastic rules of the Vinaya were theorized
as teachings about disciplinary practices of restraint aimed at coping with monks’
and nuns’ transgressions.14
Just as understanding each genre is required for gaining access to the Buddha’s
teaching, so too working with passages, sentences, and words is required for
grasping the different registers, techniques, idioms, and methods of the Buddha’s
speech that Buddhaghosa identifies as indispensable to interpretative practice.
As the scholastic enterprise developed, the exegetical tradition further expanded
canonical distinctions about words and texts as well as its own categories into
formal hermeneutic terms to try to capture the complexities of the Buddha’s words
(buddhavacana). These distinctions convey qualities of the Buddha’s words that
invoke specific exegetical cues. The Buddha was said to have sometimes spoken in
brief (saṅkittena) and other times in detail (vitthārena), in short outlines (such as
uddesa and mātikā), and in extended exposition (such as vibhaṅga and niddesa).
Sometimes he spoke in an explicit, definitive manner (nītattha) and other times in
a manner requiring interpretation to draw out the meaning (neyyattha); often the
first interpretative task is to discern which is which. He was said to have spoken
in a conventional (sammuti) idiom on some occasions and on some subjects,
while reserving the use of language described as “furthest sense” (paramattha)
for others. His speech was sometimes illustrative with narratives and examples to
make his points, a kind of discourse reliant on context called pariyāya; at other
times, he spoke more abstractly (nippariyāya), with less illustration. His words
may be analyzed according to meaning (attha) and phrasing (byañjana). All of
these distinctions—​which we consider in detail in c­hapter  2—​were regarded
by Buddhaghosa as crucial for his interpretative task: grasping how the Buddha
taught was prerequisite to grasping what he taught.

13. Buddhaghosa contrasts the colloquial teaching of the Suttanta, with the “furthest sense”
teaching of the Abhidhamma: “the Abhidhamma Piṭaka is a furthest-​sense teaching because
of being taught by the Bhagavan who is skilled in furthest-​sense [language] and because it has
a preponderance of the furthest sense” (abhidhammapiṭakaṃ paramatthakusalena bhagavatā
paramatthabāhullato desitattā paramatthadesanā; As 21; Sv i.19; Sp i.21). We discuss this dis-
tinction in detail in ­chapter 2.
14.  As 21; Sv i.19; Sp i.21:  Vinaya is a “teaching according to monastic offence”
(yathāparādhasāsanaṃ) and a “discourse on various kinds of restraints” (saṃvarāsaṃvara­kathā).
8 In t rodu ct ion

Words and Texts in Practice: Some Demonstrations

It is time for us to turn to Buddhaghosa’s commentaries to see in practice how he
develops a theory of scripture through the exegetical act itself. Buddhaghosa is
interested in certain canonical praises of the Buddha’s words not only because he
saw them as expressions of admiration and worship, but also, perhaps even more
urgently for his purposes, because he takes them as cues for interpretative prac-
tice. To begin to show what I mean, we can look at how Buddhaghosa develops
what he regarded as unique properties of the Buddha’s words, which can exem-
plify some of the major outlines of his theory of scripture and what is required
to interpret it. Three examples can illustrate his way of drawing out interpretative
cues from the canonical material, as well as begin to establish some of his main
ideas about what is going on in scripture. Praises of the Buddha’s words tell us to
look for the immeasurable, seek beauty, and notice the impact of the Buddha’s words.

Look, in Every Instance, for the Immeasurable

Let us consider first the claim that the Buddha’s words are immeasurable.
A sutta in the canonical Aṅguttara describes how the Buddha was once staying
at Rājagaha talking with a certain Upaka Maṇḍikāputta about the bad (akusala,
that is, blameworthy, unskillful, and unhealthy) practices of carping against other
people’s views. This assertion is made:

The Tathāgata has immeasurable teachings of the Dhamma about this, with
immeasurable words and immeasurable phrasings.15

Buddhaghosa’s interpretation of this line reads:

“Immeasurable words about this” means that when this is declared to

be “bad,” even the words, the letters, and the Teaching of the Dhamma
are immeasurable. When there is the declaration of bad in this way “this
is bad,” the ways it is received also are immeasurable:  “this is bad, this
also is bad, it is bad for this reason, or for this reason.” Moreover, be-
cause the Tathāgata would teach the Dhamma by different modes, the
Teaching would in this way also become immeasurable. For it is said that
“the teaching of the Dhamma by the unlimited Tathāgata is unlimited in

15.  A  ii.181:  Tattha aparimāṇā padā aparimāṇā byañjanā aparimāṇā tathāgatassa

dhammadesanā.
 Introduction 9

Dhamma, word, and phrasing.” The meaning should be understood with

this approach in every case.16

Buddhaghosa is packing a great deal into this interpretation. The canonical claim
that the teaching of the Dhamma, in its words and phrasings, is immeasurable,
indicates that the actual words and letters—​the ways of putting the Dhamma—​
are immeasurable: there are many ways to describe problematic practices or views.
Additionally, there are many ways an audience can receive such a declaration and
apply it (presumably as many receptions of a teaching as there are audiences).
Additionally, the modes of teaching are immeasurable:  the Buddha taught in
different registers, styles, and pedagogical methods (which we will begin to ex-
plore throughout this book). And finally, since the Tathāgata is himself “unlim-
ited,” what he taught is unlimited or unexhausted. And Buddhaghosa thinks
this interpretation is to work as a general claim, to be deployed in all cases: the
Buddha’s words are immeasurable in how they may be worded and phrased, how
they may be received, the modes in which they may be taught, and because they
issue from an omniscient ken.

Seek Beauty in Parts and Wholes

For another slightly longer example of canonical praises of buddhavacana that
Buddhaghosa develops into interpretative practices, let us turn to a widely cited
canonical claim about how the Buddha taught:

The Dhamma is well-​spoken by the Bhagavan, visible here and now, timeless,
inviting one to come and see, leading forward, and to be experienced by the
wise for themselves.17

In the commentaries, Buddhaghosa refers his readers to a lengthy discussion

of this sentence in the Visuddhimagga, where the exegesis of what it means
functions also as a contemplative exercise called “recollection of the Dhamma,”

16.  Mp iii.167:  Tattha aparimāṇā padātiādīsu tasmiṃ akusalanti paññāpane padānipi

akkharānipi dhammadesanāpi aparimāṇāyeva. Itipidaṃ akusalanti idampi akusalaṃ
idampi akusalaṃ imināpi kāraṇena imināpi kāraṇena akusalanti evaṃ akusalapaññattiyaṃ
āgatānipi aparimāṇāni. Athāpi aññenākārena tathāgato taṃ dhammaṃ deseyya, evampissa
desanā aparimāṇā bhaveyya. Yathāha—​ ‘‘apariyādinnāvassa tathāgatassa dhammadesanā,
apariyādinnaṃ dhammapadabyañjana’’nti. Iminā upāyena sabbavāresu attho veditabbo.
17. Svākkhāto bhagavatā dhammo sandiṭṭhiko akāliko ehipassiko opaneyyiko paccattaṃ veditabbo
viññūhīti (D ii.93, 217, 222; D iii.5, 228; M i.137, et cetera.).
10 In t rodu ct ion

part of a program of developing concentration and calm.18 For purposes of the

example, we need not attend to all of the pages of his extensive commentary on
this sentence (an expansion of possibilities which itself might be said to be per-
forming immeasurability), and can focus on just some of these descriptions of the
Dhamma—​that it is “well-​spoken,” “visible here and now,” and “timeless” (For a
translation of the full passage, see Appendix A).
We can begin with the idea that the Buddha’s words are “well-​spoken.”
Buddhaghosa describes many ways that the Buddha’s words can be said to be
“well-​spoken” by bringing in another recurrent formula from the canonical
sources that the Buddha

teaches a teaching beautiful in the beginning, beautiful in the middle, and

beautiful in the end with meaning and with phrasing; he makes known
the entirely perfect and pure religious life.19

In a manner similar to earlier hermeneutical guidelines from the Nettippakaraṇa,

he takes this claim to encourage a specific interpretative exercise.20 If the Buddha’s
words are beautiful (Buddhaghosa adds “good” and “blameless” to this quali-
fier21) in these three parts, we need to know how to look for and experience their
beauty in the beginning, middle, and end. Exegetical practice must be sensitive
to the ways that each unit of text is beautiful in these three places. But what is to

18.  Vism 213 [VII.69–​86]; cf. Sp i.126–​27. Sv ii.246 and Mp i.244 refer the reader to the
Visuddhimagga for the commentary. Readers will always benefit from looking at Ñāṇamoli’s
masterful translation of the Visuddhimagga (this passage is found on pages 209–​14), but I give
my translations unless otherwise noted.
19. Vism 213 [VII.69]: So dhammaṃ deseti ādikalyāṇaṃ majjhekalyāṇaṃ pariyosānakalyāṇaṃ
sātthaṃ sabyañjanaṃ; kevalaparipuṇṇaṃ parisuddhaṃ brahmacariyaṃ pakāseti (M i.179; D
i.63). The “religious life” (brahmacariya) is glossed as giving, doing service, practicing the five
moral precepts, practicing the four immeasurables (loving-​kindness, etc.), abstaining from
sex, being satisfied with one’s own spouse, having vigor, taking the Uposatha observances, and
following the Ariya path and the Dispensation (Sv i.177). Note that Buddhaghosa sometimes
sees kevalaparipuṇṇa and parisuddha as modifying dhamma, rather than brahmacariya (as he
does in Sv i.177 and Ps ii.203, but not Vism 214 [VII.70]), but the suttas themselves go on to
describe brahmacariya in these terms, suggesting that these adjectives stand in apposition to
brahmacariya.
20. The hermeneutical guide, the Nettippakaraṇa, takes this verse as an interpretative practice
(Netti 5, 9). It is interesting that similar claims about texts can be found outside of Buddhist
texts, as for example, the second-​century grammarian Patañjali’s statement that śāstras can have
“auspicious beginnings, middles, and ends” (maṅgalādīni maṅgalamadhyāni maṅgalāntāni hi
śāstrāṇi). As in the Pali tradition, this claim came to be used as an interpretative method (cited
in Minkowski, “Why Should We Read the Maṅgala Verses?,” 22).
21. Kalyāṇaṃ bhaddakaṃ anavajjameva katvā deseti (Sv i.175).
 Introduction 11

count as a unit of text to be analyzed in each part? It depends on the text in front
of one. Buddhaghosa says that every verse (gāthā) is to be taken as beautiful in its
beginning words, its middle words, and its ending words, and every sutta can be
found to be beautiful in its beginning by its introduction (nidāna), in the end by
its conclusion, and in the middle by the rest of its content:

For when the Bhagavan teaches even a single verse it is beautiful in the be-
ginning with its first line of the Dhamma because of being entirely good, it
is beautiful in the middle with its second and third lines, and it is beautiful
in the end with its conclusion. A sutta with a single sequence of meaning is
beautiful in the beginning because of its introduction, beautiful in the end
because of its conclusion, and beautiful in the middle because of the rest.22

And the entire Dispensation (sāsana) is a unit that can be analyzed for its beauty
and goodness in three parts: “the Dhamma of the entire Dispensation is beau-
tiful in the beginning because of the morality that has become one’s own well-​
being, beautiful in the middle because of calming and insight meditation as path
and fruit, and beautiful in the end because of nibbāna.”23 This device allows for
attention to parts and wholes, as well as conceiving of various types of wholes.
Further, Buddhaghosa puts this technique into practice in the early pages of
his own Visuddhimagga, where he explains its structure.24 In this way the three
sections that structure the Visuddhimagga are in fact the exegetical expansion and
enactment of this important feature of the Buddha’s speech.
He goes on. Structuring the Dhamma into sīla, samādhi, and paññā is not the
only way to show that the Dispensation is beautiful in the beginning, middle, and
end: the Dhamma is also said to be “beautiful in the beginning because of the thor-
ough Awakening of the Buddha, beautiful in the middle because the Dhamma is a
thorough teaching, and beautiful in the end because of the thorough attainment

22.  Vism 213 (VII.69):  Yan hi Bhagavā ekagātham pi deseti, sā samantabhadrakattā

dhammassa paṭhamapādena ādikalyāṇā, dutiyatatiyapādehi majjhe kalyāṇā, pacchimapādena
pariyosānakalyāṇā. Ekānusandhikaṃ suttaṃ nidānena ādikalyāṇaṃ, nigamanena
pariyosānakalyāṇaṃ, sesena majjhe kalyāṇaṃ. See Appendix A for the translation of the entire
passage.
23.  Vism 213 (VII.70):  Sakalopi sāsanadhammo attano atthabhūtena sīlena ādikalyāṇo,
samathavipassanāmaggaphalehi majjhe kalyāṇo, nibbānena pariyosānakalyāṇo.
24. Vism 4–​5 (I.10). This structure is given rather flexibly with certain variants in Vism 213–​14
(VII.70): good in the beginning according to sīla, in the middle according to calm and insight
(samathavipassanā) and path and fruit, and in the end according to nibbāna; alternatively,
in the beginning according to sīla and samādhi, in the middle, according to the path that is
vipassanā, and in the end, with the nibbāna that is the fruit.
12 In t rodu ct ion

of the Community.”25 This maps the triple gem, the Buddha, Dhamma, and
Saṅgha, on to the three places of beauty. Or, if you prefer, it may be analyzed in
terms of the reception of the words for those who hear and practice them:

It is beautiful in the beginning since one hearing it brings about only what
is beautiful just by listening, due to its eliminating the hindrances; it is
beautiful in the middle since one practicing it brings about what is beau-
tiful just by practicing it, due to its bringing about the happiness of calm
and insight meditation; and likewise it is beautiful in the end since one
obtaining it brings about what is beautiful just by the fruit of practice,
due to its bringing about the state similar [to that of the buddhas] when
completed.26

How the words make an immediate or subsequent impact on the audience is a

way they can be construed as beautiful in beginning, middle, and end.
Note that these alternative interpretive possibilities illustrate inclusivity, mul-
tiplicity, and perhaps even immeasurability in expanding meaning—​teachings
are variously presented and variously received. In fact, the Netti presents this very
tripartite tool of beautiful in beginning, middle, and end in just this way:  the
Buddha’s teaching is beautiful in the beginning in being concise, beautiful in the
middle in its diffuse or expansive nature, and beautiful in the end for how it can
be known in detail.27 What is concise in the beginning can become expanded and
extensive further out (and still be included in the whole).
What should be evident so far is that Buddhaghosa urges the reader to
attend to beauty in every unit of text (and that texts may be divided up variously
for this kind of analysis), because of its content, its maker, its reception, and
its effects. We might also note that for readers concerned that the immeasura-
bility of the Buddha’s words entails that any interpretation is possible (though
it is notable that Buddhaghosa himself is not worried about such interpretative
freedom, at least not as a general problem), we have here, and in other quali-
ties of the Buddha’s words, some constraint on meaning: any interpretation that

25.  Vism 214 (VII. 70):  Buddhasubodhitāya vā ādikalyāṇo, dhammasudhammatāya

majjhekalyāṇo, saṅghasuppaṭipattiyā pariyosānakalyāṇo.
26.  Vism 213–​ 214 (VII.70):  Suyyamāno cesa nīvaraṇavikkhambhanato savanenapi
kalyāṇameva āvahatīti ādikalyāṇo, paṭipajjiyamāno samathavipassanāsukhāvahanato
paṭipattiyāpi kalyāṇameva āvahatīti majjhekalyāṇo, tathā paṭipanno ca paṭipattiphale niṭṭhite
tādibhāvāvahanato paṭipattiphalenapi kalyāṇameva āvahatīti pariyosānakalyāṇo.
27. Netti 9; see Ñāṇamoli, The Guide, 19, for his translation.
 Introduction 13

yields a reading of the Buddha saying something ugly cannot have issued from
the Buddha’s words.

Attend to the Dhamma’s Immediate Impact

Our original formula—​that what the Buddha taught is “well-​spoken, visible here
and now, timeless, inviting one to come and see, leading forward, and to be ex-
perienced by the wise for themselves”—​receives further exegetical expansion. We
can listen in further to find some of what Buddhaghosa does with the idea that
the Buddha’s teaching is “visible here and now” and “timeless.” “Visible here and
now” suggests to him that something can be seen immediately by one who puts
the Buddha’s words into practice: “here the Noble Path is indeed visible here and
now in that it can be seen for oneself by a noble person not doing lustful things,
et cetera in his own continuing experience.”28 As he goes on to explain, such a
person ceases problematic motivations and intentions and so can see for him-
self that painful and unhappy experience that stems from them ceases too (for
the whole translation of this passage, see Appendix A). He continues this line of
thinking in his gloss on how the Dhamma is “timeless”: “timeless means that there
is no time between it and the bearing of its fruit.”29 The Buddha’s words about the
Path have an immediate impact.
In a different reading of the line, Buddhaghosa picks up visuality in an-
other way, where “visible here and now” concerns what is seen in the sense
of the person who is seeing: “alternatively, ‘what is seen’ is called seeing, and
what is visible is just what is seen, which has the meaning of seeing. . . . For
the transcendent Dhamma is just that very seeing which itself leads one away
from the fear of saṃsāra by means of the comprehension that is meditative
cultivation and the comprehension that is direct experience.”30 In this inter-
pretation, the Dhamma that the Buddha taught should be understood as a
way of seeing that transforms a person so that fear of saṃsāra is allayed. The
words of the Buddha can change the way a person sees or looks at the world
in both meditation and direct experience, altering what becomes visible in
the first place.

28.  Vism 215 (VII.76):  ettha pana ariyamaggo tāva attano santāne rāgādīnaṃ abhāvaṃ
karontena ariyapuggalena sāmaṃ daṭṭhabboti sandiṭṭhiko.
29. Vism 216 VII.80: Attano phaladānaṃ sandhāya nāssa kāloti akālo.
30. Vism 216 (VII.78): Atha vā diṭṭhanti dassanaṃ vuccati. Diṭṭhameva sandiṭṭhaṃ, dassananti
attho  .  .  . Lokuttaradhammo hi bhāvanābhisamayavasena sacchikiriyābhisamayavasena ca
dissamānoyeva vaṭṭabhayaṃ nivatteti.
14 In t rodu ct ion

These features of scripture’s immediacy and transformative impact become

vital to Buddhaghosa’s interpretative practice. He is, as we come to observe,
nearly always concerned with the original narrative context in which the Buddha’s
words were given in the suttas and the Vinaya, and how the words had an imme-
diate and transformative impact on that audience. When interpreting both suttas
and Vinaya rules, the original narrative of the teaching is important for under-
standing the meaning of the teaching or the rule itself, no matter how abstract or
universal it may be. At the same time, the Dhamma is said to be transcendent, and
the teachings universalizable. They can speak far outside of their original context,
but are often understood to do so precisely for how they speak to that context.
This is in part because the dialogical narrative helps us to see the transformative
impact of the words on the audience and the shift they allow in seeing and un-
derstanding. Knowing how a text was initially received is instructive for knowing
how it may be subsequently received, even while what a text can come to mean is
not limited to its original context.
But some of this begins to get ahead of ourselves, for these themes—​the
immeasurability, the beauty, the immediacy, the transformative impact of
the spoken Dhamma—​are explored throughout the book, and are only here
mentioned as examples of the kinds of things that Buddhaghosa drew out of
formulaic praises of the Dhamma as he developed them into interpretative
practices.

The Immeasurable Teachings of

an Omniscient Teacher
Perhaps most intriguing among the many claims about the Buddha’s words is the
idea that they were said to be immeasurable. As we saw earlier, the idea that the
Buddha’s words are immeasurable and unlimited is related to the idea that the
Tathāgata is himself unlimited, an idea that found development in treating his
knowledge as unlimited. We find assertions of the Buddha’s omniscience every-
where in the commentaries. Tentative statements about the Buddha’s “knowing
all”31 in the canon came to be amplified into full-​blown claims of his omniscience
in the commentaries. There are several ways to think about the commentarial
development of this idea. One would be that the commentaries articulated
a buddhology that had developed in relation to the broader Indian context in

31. M.i.171, ii.93: in this sutta, Buddha claims to be a “knower of all” (sabbavidūham asmi) and
hesitates to teach since his knowledge may not find a recipient who can understand it; in other
places he disavows omniscience. We take up the development of claims about the Buddha’s
omniscience in ­chapter 1.
 Introduction 15

which the Jain teacher Mahāvīra was an omniscient competitor; theological

developments in other Indian systems emphasized the omniscience of various
conceptions of Īśvara; purāṇic and devotional Hinduism made similar claims
regarding Viṣṇu, Śiva, and the Devī; and the Buddha was seen in increasingly
extravagant terms in Mahāyāna Buddhist movements. Even without going out-
side of the Pali sources, a robust devotional strand within the Pali tradition that
extolled the Buddha and his knowledge of the Dhamma in the highest possible
terms had been present from the very beginning.
Not exclusive to any of these likely factors is a suggestion that works from a
slightly different angle, and one that I want to explore in this book. It is possible
that the commentarial project itself facilitated the idea that the Buddha was om-
niscient, and that Buddhaghosa came to emphasize omniscience because it gave a
name to features of the textual corpus that he found immeasurably and infinitely
generative. This is to say, perhaps the Buddha’s omniscience was something that
commentarial work with the texts discovers, rather than (only) presumes. For ex-
ample, the modular nature of texts, passages, and categories that Buddhaghosa
encounters and enacts can be seen as one way in which texts generate new
meaning and continue to speak to multiple (or perhaps infinite) contexts. What
we might call rampant intertextuality was for him the unlimited application
of the teachings of an omniscient being. And the temporal aspects of reading
scripture that we have begun to notice raise intriguing possibilities for how texts
might speak to their own contexts as well as to present and future contexts. The
transcendent Dhamma can become immanent as it speaks to the “here and now,”
but is of course always transcendent, speaking well beyond it. The Abhidhamma
methods are the transcendent Dhamma lifted out context and speaking in per-
haps the Buddha’s most universalist register.
We come to see that the commentarial conception of the Buddha’s omnis-
cience is quite specific: the Buddha can know without obstacle or limit any subject
when he turns his mind and attends to it. I discuss this at length in c­ hapter 1, only
noting here that this conception is particularly appropriate to the commentarial
methods that Buddhaghosa advances, where meaning is developed by analytic
methods that when applied to particular contexts need never, in principle at least,
stop. Scripture is, for Buddhaghosa, “an ocean of methods,” a bottomless sea of
methods that find ever deeper application. And if scripture is oceanic, then so
too is the “ocean of the Buddha’s knowledge” from which it issues.32 The oceanic

32. There are four oceans: “the ocean of saṃsāra, the ocean of water, the ocean of methods, and
the ocean of knowledge (saṃsārasāgaro, jalasāgaro, nayasāgaro, ñāṇasāgaroti, As 10), where the
ocean of methods is “the tipiṭaka that is the Buddha’s words” (Katamo nayasāgaro? Tepiṭakaṃ
buddhavacanaṃ, As 11). See ­chapter 1 and Appendix C.
16 In t rodu ct ion

enormity is to be felt particularly on the side of the one beholding these two
oceans who will find them even more “incalculable and immeasurable”33 than the
great briny sea itself.
Such observations about textual infinity are hardly unique to Buddhaghosa.
Other interpreters have seen how the written text may be read and reread in an
infinite number of ways as it encounters ever new readers and opens up new lines
of thought for them. As Paul Ricoeur sees it, a world of meaning lies not so much
behind a text in the author’s intention, but in front of it as the text encounters
each of its readers. The ever-​changing contexts in which a text is read expand
its meaning, an infinity of the written text he calls its “surplus of meaning,” and
which Jacques Derrida calls its “plenitude.”34 But Buddhaghosa would find him-
self more at home with other medieval thinkers who knew this (rather than
modern deconstructionists like Derrida). For it is not just because a text has an
infinite number of potential readers that its meaning may develop infinitely, but
also, at least in the case of scripture, because of the omniscience of its author. As
Ian Almond suggests, medieval thinkers like the thirteenth-​century Sufi exegete
Ibn ‘Arabi linked scriptural infinity to an omniscient author. For Ibn ‘Arabi, the
Quran, as the “inexhaustible words” of God, “has no single message but, rather,
a variety of messages, each one gauged to the competence and situation of its
reader.”35 To interpret the Quran is “to participate in its expansion,” as under-
standing the text draws out the meaning anticipated by God. In Ibn ‘Arabi’s
own words: “on God’s part there are perpetual turnings of attentiveness and in-
exhaustible words”; thus “the situation is new forever.”36 Buddhaghosa and Ibn
‘Arabi would find much to discuss with the Kabbalists. The Torah was said to
possess infinite meaning; one medieval text asserts that “many lights shine forth
from each word and each letter.”37 Gershom Scholem notes the widespread belief
“that the number of possible readings of the Torah was equal to the number of
the 600,000 children of Israel who were present at Mount Sinai—​in other words,

33. As 11: asaṅkhyeyyo appameyyo. This metaphor gets extensive development in the Atthasālinī,
discussed in ­chapter 1.
34.  On Derrida, see Almond, “The Meaning of Infinity in Sufi and Deconstructive
Hermeneutics:  When Is an Empty Text an Infinite One?,” 104. Ricoeur, Interpretation
Theory: Discourse and the Surplus of Meaning, 92.
35. Almond, “The Meaning of Infinity in Sufi and Deconstructive Hermeneutics,” 103.
36. Almond, “The Meaning of Infinity in Sufi and Deconstructive Hermeneutics,” 108.
37. Scholem, Kabbalah, 172.
 Introduction 17

that each single Jew approached the Torah by a path that he alone could follow.”38
We can continue leaping across medieval traditions to note similar ideas in the
work of Meister Eckhart, who shares with Buddhaghosa and Ibn ‘Arabi the oce-
anic imagery of the “shoreless sea” that is scripture, where “there is none so wise
that when he tries to fathom it, he will not find it deeper yet and discover more
in it.”39
As intriguing as this comparative potential may be, we have our hands full
dealing with Buddhaghosa, for once the omniscient Buddha and his inexhaust-
ible words become central to the commentarial project, questions about what
this means for the interpreter only deepen. What is it to interpret a text com-
posed by an omniscient author? How can texts convey his omniscient and im-
measurable knowledge? How can words—​finite utterances, recitations, suttas,
books—​deliver this unlimited, immeasurable, and omniscient knowledge? And
how might we, with our limited understandings, receive and grasp it?
Other scholars have also noted this distinctive interpretative challenge
facing early Buddhist interpreters. George Bond writes that for the Theravada
“the uniqueness of the Buddha’s teachings constitutes the crux of the her-
meneutical problem. The wisdom of the Buddha is believed to have been as
immense as the ocean. For all practical purposes, he had what can be called om-
niscience.”40 Bond argues that the Pali commentators frame their authority and
significance in terms of this project: interpreting the Dhamma requires one to
grapple with the Buddha’s omniscience. Kate Crosby has written about how the
Abhidhamma tradition in particular sought “to encapsulate and represent as
far as possible the Buddha’s omniscience.”41 Writing more broadly about Indian
Buddhist traditions, Ronald Davidson observes a kind of circular relationship
and permeability “between the Buddha, his dharma, the reality discovered by
him, and the cognition of that reality.”42 Coming to understand one leads to the

38.  Scholem, Kabbalah, 172, and cited in Almond, “The Meaning of Infinity in Sufi and
Deconstructive Hermeneutics,” 103.
39. Almond, “The Meaning of Infinity in Sufi and Deconstructive Hermeneutics,” 106, citing
Walshe, Meister Eckhart: German Sermons and Treatises, 250. Griffiths also notes oceanic im-
agery in “religious readings” of scripture (Religious Reading, 41).
40. Bond, “Theravada Buddhism and the Aims of Buddhist Studies,” 59. He goes on to argue
that the “Pāli Commentaries receive their authority and significance for the Theravādins be-
cause they provide the only avenue of approach to the meaning the Dhamma and represent the
only solution to this hermeneutical problem.”
41. Crosby, Traditional Theravada Meditation and Its Modern Era Suppression, 76.
42. Davidson, “Standards of Scriptural Authority,” 294–​95.
18 In t rodu ct ion

others.43 Davidson’s observation is particularly useful for helping us see the ways
that the spoken Dhamma is so often understandable principally (as I show for
Buddhaghosa) in reference to the Buddha. And of course the observation it-
self may be seen as a distant echo of the Buddha’s own words: “whoever sees the
Dhamma, sees me, and whoever sees me, sees the Dhamma.”44

Buddhaghosa
Despite being one of the giants of the Buddhist tradition, Buddhaghosa and
his scholarly work have attracted only a handful of monographs. He has also re-
ceived, unjustly, I  believe, occasional scholarly disdain.45 Much of the modern
attention that he has received has been concerned with text-​critical questions,
such as which texts traditionally ascribed to him were likely to have actually been
his work and what can be discerned about the history of the texts he inherited,
edited, and translated. These are questions complicated by the fact that he is
the editor and translator (into Pali) of a very large body of received tradition
(in ancient Sinhala); “his” commentaries are actually, in his own representation
of the matter, the transmission of this older, but now lost, corpus of material.
This material, the aṭṭhakathā, was alleged to have been taught by teachers dating
to the Buddha’s time, and, while not considered “Buddha’s words,” is reported by
Buddhaghosa to have been recited at the First Council.46 The Visuddhimagga too
seems to have been in part based on an earlier text.47

43.  Davidson, “Standards of Scriptural Authority,” 296. He cites the Abhidharmadīpa, a

Vaibhāṣika text, which connects the spoken dharma to the Tathāgatha’s omniscience in
these terms.
44. S iii.120: Yo kho, vakkali, dhammaṃ passati so maṃ passati; yo maṃ passati so dhammaṃ
passati. Buddha tells this to Vakkali, a critically ill monk who had felt remorse and regret at not
having been able in his ailing state to see the Buddha.
45.  For example, Keown finds it appropriate to say that one “can move at some speed”
through the chapter on morality of the Visuddhimagga, since despite the detail developed by
Buddhaghosa, “the harvest in terms of a deeper understanding of sīla is disappointingly sparse”;
Buddhaghosa provides only “dry and disconnected classifications” on questions of morality,
and while Keown finds Buddhaghosa’s discussion of the subject “difficult to penetrate,” he is
assured that “much can be passed over without comment” (“Morality in the Visuddhimagga,”
73, 61).
46.  In the colophons of his commentaries to each of the four nikāyas, Buddhaghosa says
the aṭṭhakathā was recited at the community of five hundred monks in order to explain the
meaning of the canonical books (as, for example, Sv i.1: atthappakāsanatthaṃ, aṭṭhakathā ādito
vasisatehi; Pañcahi yā saṅgītā, anusaṅgītā ca pacchāpi.)
47. The Visuddhimagga is thought to have a precursor in the Vimuttimagga, though the degree
to which the two texts differ is considerable. See Bapat, Vimuttimagga and Visuddhimagga: A
 Introduction 19

While the Pali tradition attributes a very large body of commentarial material
to him, modern scholars have been more circumspect about which commentaries
they see as his and indeed what authorship even meant in this context.48 Authorship
conceived of as a single historical person writing something new and entirely
original is elusive in this context. My approach to questions of his authorship is
to consider seriously the possibility that the Samantapāsādikā, the aṭṭhakathās
on the four nikāyas, the Paramatthajotikā, and the three commentaries on the
books of the Abhidhamma, were handled by the same author (or lead author, if,
as most likely, he was working with a team) as the Visuddhimagga (though the
degree of originality of the commentarial material, given that the original Sinhala
versions are no longer extant, is impossible to establish49). I see in this corpus of
texts a very cohesive and systematic program produced by the “school” that his
name has come to represent and is how I use it here; the contours of this school
emerge in the chapters that follow. Even allowing for the prevalence of whole-
sale borrowing and recycling of textual material by authors of different texts in
premodern India, we have in the texts listed previously very substantial bodies of
very similar textual passages and, for the most part, a singular voice and approach.
Indeed, many of the main passages in the nidānas of all three piṭakas on which
I focus are identical, and many passages in the commentaries are identical to parts
of the Visuddhimagga.
I am particularly interested in the modular nature of such identical passages as
they traverse texts and genres, and how they are used in one context for a partic-
ular purpose and then deployed in another context for a completely different pur-
pose. Modularity—​the degree to which a system’s components may be separated
and recombined to new effect—​is dear to the heart of Buddhaghosa’s thinking

Comparative Study, and, for a translation of the Vimuttimagga, see Ehara, Soma Thera, and
Kheminda Thera, The Path of Freedom.
48.  For recent work on questions of which commentaries are connected to Buddhaghosa,
see Cousins, “The Case of the Abhidhamma Commentary,” and von Hinüber, “Building the
Theravāda Commentaries: Buddhaghosa and Dhammapāla as Authors, Compilers, Redactors,
Editors, and Critics.” While perhaps not all of the commentaries attributed to the editorial, au-
thorial, and translation work of Buddhaghosa could be his, we can usefully speak of a “school”
called Buddhaghosa, as Cousins suggests (390). Von Hinüber doubts that Buddhaghosa was
the author of the Abhidhamma and Vinaya commentaries, but considers it likely that he was
connected with them as “the overall organizer” or “head” (262–​64). Also important are Endo,
Studies in Pāli Commentarial Literature; Norman, Pāli Literature, 120–​30; Collins, “Remarks
on the Visuddhimagga,” 50; and Malalasekera, The Pāli Literature of Ceylon, 94.
49.  On the specific question of the relationship of the Pali aṭṭhakathās with the Sinhala
commentaries on which they are based, see, in addition to the sources listed in the pre-
vious footnote, Palihawadana “Dhammapada 1 and 2 and their Commentaries,” and Pind,
“Buddhaghosa: His Words and Scholarly Background,” 135–​56.
20 In t rodu ct ion

on the Dhamma, and vital for understanding the systematicity in the work the
tradition attributed to him. It becomes a technique for generating new meaning
and purpose. For example, our passage on the beautiful nature of the Buddha’s
words is used in the Visuddhimagga as a contemplative exercise designed to calm
and focus the mind (Vism VII.69–​70). But the same passage is also used in the
introduction to the commentary on the Vinaya (Sp i.126), where such reflections
about the Dhamma function as stage-​setting for explaining the greatness of the
Buddha and how he came to know and teach the monastic rules. Parts of the
same passage are again used in the beginning of the Visuddhimagga, to explain
its structure (Vism I.10). This small but representative example suggests both the
systematicity of the aṭṭhakathā layer (which includes and makes constant refer-
ence to the Visuddhimagga) taken as a whole and also the need to be sensitive
to context and purpose in the interpretation of any particular passage, since the
work that it might be doing varies in these different contexts. Notable also in
this example is that the passage doing exegetical work explaining the Buddha’s
words is at one and the same time used as a contemplative exercise in a training
on meditation, suggesting a conflation or a dual purpose of textual analysis and
contemplative analysis (a theme we return to many times in the book). This may
be a case where modern assumptions that would sharply delineate textual analysis
and contemplative practice need to be reexamined.
I aim to show that the Visuddhimagga and the commentaries attributed to
Buddhaghosa constitute a body of material reflective about and guided by a
programmatic system. While commentaries must follow their root texts in the
ordering of their contributions and thus can sometimes obscure how they fit
into the whole of a commentator’s own intellectual framework, Buddhaghosa
provides substantial reflection on how he sees the “whole” that he is endeavoring
to construct; indeed, demonstrating some of how he thought about that
“whole” (that which concerns interpreting the Dhamma textually) is an aim
of this book. One obvious whole in question is, of course, the Visuddhimagga,
which is a very systematic presentation of the Dhamma. Buddhaghosa sees the
Visuddhimagga as the hub of his commentarial writings. He depicts a system with
“the Visuddhimagga at the center of the commentaries on the four āgamas,”50
and the commentarial tradition following him took this to be a claim about a
single commentarial system of shared material that also included commentaries
on the Abhidhamma and Vinaya.51 Of course, the Visuddhimagga is just “part”
of the much greater whole that is the Dhamma itself as it reflects or conveys

50. Sv i.2: majjhe visuddhimaggo esa catunnampi āgamānañhi.

51. Dīghanikāya ṭīkā i.22, for example.
 Introduction 21

the immeasurable “whole” of the Buddha’s awakened knowledge. Exploring

“wholes” and “parts,” and the immeasurability of the ultimate “whole,” is also a
recurrent theme.
It is also important to understand how Buddhaghosa describes his labors
within what we can glean of the intellectual conditions bearing on them. The
praise verses introducing his works mention his efforts to clarify and trans-
late the ancient aṭṭhakathā, and his regard for the ancient commentators—​the
elders of old (porāṇas) and earlier masters (pubbācariyas)—​on whom he bases
his work. Several verses that open the Paramatthajotikā (the commentary on the
Khuddakapāṭha) are quite striking for the humility they suggest and his aware-
ness of standing on the shoulders of giants, as it were. After praising the three
jewels—​the Buddha, the Dhamma, and the Saṅgha—​he says:

Because of the depth of the Khuddaka certain commentaries are very diffi-
cult for someone like me who is not awakened to the Dispensation.
But since the certainty of the early masters is unbroken even today, the
ninefold Dispensation of the Teacher is still established.
Therefore, I want to make this commentary on the meaning, relying
on the Dispensation and the ancient deliberations, with veneration of the
Good Dhamma, not from a desire for self-​praise or for the purpose of
despising others. So listen attentively!52

This full acknowledgment of his condition—​of being “someone like me” who is
unawakened—​appears to require his commitment to the hoary and unbroken tra-
dition. His humility in this endeavor appears from time to time elsewhere in his
work; it might just indicate a trope of modesty and deference to the authorities,
but it can also be interpreted in a way that is in keeping with what he takes to be
Buddhist practice, which, as I  show, involves being personally oriented to and

52. Pj I.1: Khuddakānaṃ gambhīrattā, kiñcāpi atidukkarā/​Vaṇṇanā mādisenesā, abodhantena

sāsanaṃ. Ajjāpi tu abbocchinno, pubbācariyanicchayo/​Tatheva ca ṭhitaṃ yasmā, navaṅgaṃ
satthusāsanaṃ. Tasmāhaṃ kātumicchāmi, atthasaṃvaṇṇanaṃ imaṃ;/​Sāsanañceva nissāya,
porāṇañca vinicchayaṃ. Saddhammabahumānena, nāttukkaṃsanakamyatā; Nāññesaṃ
vambhanatthāya, taṃ suṇātha samāhitāti. One could alternatively translate the last two verses
by separating them: “Therefore, I want to make this commentary on the meaning, relying on
the Dispensation and the ancient deliberations. So listen attentively, with veneration of the
Good Dhamma and not from a desire for self-​praise or for the purpose of despising others.”
It should be noted that Adikaram, whose work on the porāṇas is indispensable to the study
of the Pali commentaries, has a very different reading of these verses, arguing that this hu-
mility suggests that the Paramatthajotikā is unlikely to be Buddhaghosa’s (Early History of
Buddhism in Ceylon, 7–​8). I am grateful to Charles Hallisey for reading and discussing this
passage with me.
22 In t rodu ct ion

profoundly awed by the greatness of the Buddha and his Dispensation (as they
are remembered from the past).
As we have seen, legends about Buddhaghosa’s life describe him as arriving in
Lanka as an outsider and having to prove his mettle to the monastic authorities
by producing the Visuddhimagga repeatedly and consistently, and only then is
he invited to translate the commentaries from the local language of old Sinhala
to the translocal language of Pali. Whatever we might make of the historical ve-
racity of these legends, they do suggest a milieu in which the production of the
Visuddhimagga and his involvement with the aṭṭhakathā were carried out under
the watchful supervision of authorities committed to certain representations
of a tradition about which there were differing views and strong investments.
Buddhaghosa avers that he follows the Sīhala-​aṭṭhakathā, “not contradicting the
understanding of the luminaries of the lineage of Elders, those residing in the
Mahāvihāra.”53 At the same time, he often offers alternative readings and allows
multiple readings to stand side-​by-​side in his commentaries, sometimes suggesting
a preferred reading but without insisting on it as the final or only interpretation.
Perhaps this is a way that he worked to make possible readings he favored without
omitting or refuting those of the “luminaries” who may have overseen his work.
Also important to note as we assess the nature of his project is what he
does not say about his work. One way to track this is to consider the opening
praise verses of his works. We can compare his opener to his commentary on the
Khuddaka that we just saw to those of the slightly later fifth-​century Buddhist
thinker in India, Diṅnāga. Diṅnāga has often been credited with initiating
Buddhist thinkers’ epistemological turn and their development of rational argu-
mentation that requires entering into polemics, defending one’s own argument,
and refuting that of others. As Piotr Balcerowicz shows, Diṅnāga marks a change
in opening verses in this period that reflects a context of debate with rival schools
of philosophy reading and rebutting one another, and the shift is demonstrated
by subsequent Jain and Buddhist accounts of the purposes of their work. Later
texts’ opening verses indicate that they intend to argue for a position rationally
while dismantling rivals’ views; Balcerowicz contrasts these elements with those
of earlier texts aimed at an in-​house audience and concerned with “laudation.”54
For example, Diṅnāga’s verses opening his Pramāṇa-​samuccaya describe his
efforts to establish the cognitive criterion (pramāṇa) that supports his positions
“with the purpose of refuting rival [theories concerning] the cognitive criterion

53. Sv i.21, as translated by von Hinüber, “Building the Theravāda Commentaries,” 355.

54.  Balcerowicz, “Some Remarks on the Opening Sections in Jaina Epistemological
Treatises,” 57–​58.
 Introduction 23

and with the purpose of indicating [superior] qualities of my own [theory].”55

Buddhaghosa never talks about his work this way.
As important as this historical shift toward the epistemologically argued
philosophical treatise may be, it would be restrictive to limit our conception
of philosophy to it. The Indian tradition is full of diverse genres that scholars
have found philosophically significant, including the speculative and dialogic
discussions of the Upaniṣads, various forms of śāstra, literary traditions, and
so on. Buddhaghosa’s contribution may not shine brightly if it is to be studied
only within the terms of modern analytic philosophy as the received philo-
sophical practice used to explore much Mahāyāna philosophy today. Rather he
represents the culmination of a fundamentally different Buddhist approach to
the teachings. Buddhaghosa’s philosophical significance can become better un-
derstood as we allow ourselves to be guided by how he describes his methods and
purpose. For now, I can tip my hand to mention at least two important philo-
sophical contributions of his work that I elaborate in the chapters that follow: his
practices of phenomenological analysis, as carried out in his commentary on
the Abhidhamma, and his dialogical philosophy, where richly drawn narrative
context and literary experience in both the Suttanta and Vinaya commentaries
prompt therapeutic and soteriological transformation.
Buddhaghosa’s doctrinal role within Theravada Buddhism has been histor-
ically highly significant, and his work, particularly his systematization of the
contemplative practices in the Visuddhimagga, continues to be influential for
many Buddhist communities. The medieval scholar Sāriputta, for example, wrote
handbooks and commentaries on Buddhaghosa’s work that made it more acces-
sible and helped establish its centrality to the tradition,56 and Buddhaghosa’s work
influenced vernacular traditions, in some cases heavily.57 Evidence from monastic
libraries in Southeast Asia indicates that the Visuddhimagga could often be found
among a range of other textual sources available on the Dhamma (and where the
tipiṭaka itself was not widely available in its entirety, at least until recently).58
In the modern period, Buddhaghosa’s meditation instructions as well as his

55. As translated and discussed by Balcerowicz, “Some Remarks on the Opening Sections in
Jaina Epistemological Treatises,” 61.
56. Crosby, “Sāriputta’s Three Works on the Samantapāsādikā.” Blackburn, Buddhist Learning
and Textual Practice in Eighteenth-​Century Lankan Monastic Culture, 80, notes the impor-
tance of Buddhaghosa’s work in this later period.
57. Hallisey, Devotion in the Medieval Literature of Sri Lanka, 222–​29.
58. Crosby, Theravada Buddhism: Continuity, Diversity, and Identity, 89, citing Keyes, “Merit-​
Transference in the Kammic Theory of Popular Theravāda Buddhism,” 272.
24 In t rodu ct ion

doctrinal formulations have attracted both fierce adherents and detractors, with
many committed views on all sides regarding what it is that he taught. In keeping
with the central role of Buddhaghosa’s work in Burma, the nineteenth-​century
reformer Ledi Sayadaw relied on the Visuddhimagga even while he offered a sig-
nificant alternative to it as he revived the Abhidhamma tradition and fashioned
a new lay-​oriented practice in that country that helped generate the modernist
mindfulness movement internationally.59 More stridently, the Thai reformist
monk Bhikkhu Buddhadāsa situated his own system of doctrine and practice in
opposition to that of Buddhaghosa (as he understood him).60 Nevertheless, John
Strong echoes a widely held view among scholars and practicing Buddhists that
the Visuddhimagga “remains the greatest compendium of Theravāda thought
ever written” and that Buddhaghosa’s exegeses on the canonical texts have shaped
the way the Buddhist scriptures have been read ever since.61

This Book
In keeping with textual practices I have come to admire in Buddhaghosa, I should
say something about what prompted this book, its context, and how I hope it
might be received. When I first began to study Abhidhamma and tried to figure
out what it is, I found myself returning time and again to Buddhaghosa’s introduc-
tion to his commentary on its first book, the Atthasālinī. As mentioned earlier,
this introductory matter is also quoted in large measure in his introductions to
the other piṭakas, and came to constitute the theoretical and practical frame-
work he used to interpret the types of Buddhist knowledge. The more I read his
commentaries on all genres the more I  found that this framework for treating
Buddhist knowledge informed and shaped Buddhaghosa’s actual commentarial
practice. Learning Buddhaghosa’s views of how Buddhist texts work has helped
me follow him more carefully as he moves around in the vast literature of the
Pali canon. It has also helped me grasp the very systematic architectonic of the
Dhamma that his work helped construct.
It also changed my encounter with canonical sources. Texts I thought I under-
stood became, in effect, new again for me when I read them with Buddhaghosa.
Guided by Buddhaghosa’s theories of text and his distinctive reading practices
made my reading of canonical sources much richer and often sources of great

59. Braun, The Birth of Insight: Meditation, Modern Buddhism, and the Burmese Monk Ledi
Sayadaw.
60. Swearer, ed., Me and Mine: Selected Essays of Bhikkhu Buddhadasa, 121–​23.
61. Strong, “Buddhaghosa,” 75.
 Introduction 25

delight. The delight is something Buddhaghosa predicted:  “Endless joy and

happiness arise for sons of a good family, possessing faith, abounding in serene
clarity, and whose knowledge is excellent, who reflect on two of the scriptures.
Which two? The Vinaya and Abhidhamma.”62 (I may not meet  all of these
criteria, but I  remain intrigued that these texts can yield joy and happiness.)
Further, I began to notice, repeatedly, how Buddhist knowledge is nearly always
mediated through the Buddha’s qualities and story. In particular, I  was struck
by how frequently I  was informed of the Buddha’s omniscience, which was
taken to be demonstrated by whatever passage was before me. I also became—​
again in following his advice for reading texts—​profoundly interested in con-
text and the ways that knowledge is situated in contexts that define what it is.
Buddhaghosa is concerned always with how knowledge is produced, discussed,
transmitted, and received narratively. I began to see the possibilities of a literary
reading of the Suttanta and Vinaya that can occur when we read them within
their contextual frames and avoid (at least sometimes) a fractured reading of their
doctrines or rules abstracted from their “original” narratives. I began to think of
Buddhist philosophical issues differently; ethical, philosophical, and soteriolog-
ical teachings are fashioned and shaped by narrative and dialogical contexts that
the canonical texts and Buddhaghosa construct for them. Often Buddhaghosa’s
reading practices are unexpected, sometimes even, as Jonathan Walters has noted,
“startling.”63 Moments of being startled draw me in because they give access to
knowledge practices that are, from the ground up, different from the categories
and practices of the modern west. They pave the way for a fresh understanding
of what is going on conceptually and philosophically in ancient texts. But even
as Buddhaghosa’s philology and hermeneutics have helped me read the scrip-
tural tradition differently, my focus remains, first and foremost, on exploring the
contours and significance of Buddhaghosa’s intellectual project.
The focus on theories of texts and hermeneutics has been understudied in
Buddhism, which is surprising in the context at least of the Pali tradition, where
so much textual reflection and guidance on language, interpretative methods,
texts, canon, and reading practice are actually available in sources. To be sure,
there is very good work on the Nettippakaraṇa and Peṭakopadesa and on the
formal services commentaries offer, and several valuable studies of Buddhist

62.  As 11:  Dvepi hi tantiyo paccavekkhantānaṃ saddhāsampannānaṃ pasādabahulānaṃ

ñāṇuttarānaṃ kulaputtānaṃ anantaṃ pītisomanassaṃ uppajjati. Katamā dve? Vinayañca
abhidhammañca (As 11). This is in the context of the “ocean of methods” (nayasāgara) com-
posed of the three piṭakas.
63. Walters, “Suttas as History,” 280.
26 In t rodu ct ion

scholasticism and hermeneutics.64 But only a handful of scholars have written

about Buddhaghosa with an interest in the sort of literary and interpretative
questions I am suggesting. First among them is Charles Hallisey, without whose
courses, conference papers, published articles, and conversations with me this
project could never have been even conceived. His work is indispensible to my
own approach. B. C. Law’s monograph on Buddhaghosa considers some aspects
of interpretative practice in the aṭṭhakathās and the Visuddhimagga, and George
Bond’s excellent The Word of the Buddha has a chapter on Buddhaghosa’s styles of
interpretation; this project follows up on many of his suggestions. Steven Collins’
attention to narrative and systematic styles of Buddhist thinking laid important
groundwork for considering the literary aspects of Pali material in general and the
Visuddhimagga in particular.65 Drawing on Collins, Jonathan Walters has written
with insight about Buddhaghosa’s literary sense and how radically different his in-
terpretative approach is when contrasted with modern assumptions about texts.66
Toshiichi Endo has described in detail canonical and commentarial approaches
to buddhology and important features of Pali commentarial literature.67 But for
the most part, many of the ideas about texts, genre, discourse, and meaning that
I have begun to learn about from Buddhaghosa—​and their broader theoretical
and philosophical significance—​are far from exhausted by the existing scholarly
literature.
My method might need additional comment because certain features of it
may be unusual in the fields to which I  hope this book makes a contribution.
As suggested earlier, my method is to follow Buddhaghosa around his texts and
try to understand—​and then describe in terms intelligible to my own intel-
lectual milieu—​what he is up to. I think that authors often tell us how to read
them, and that Buddhaghosa is a particularly astute instructor in this respect.
This book is, above all, an attempt to read Buddhaghosa in the manner of an
apprentice, according to the exegetical cues and practices he regularly advocates,

64.  On the Pali hermeneutical guides and protocols, see Ñāṇamoli, The Guide, and Bond,
The Word of the Buddha. On Buddhist hermeneutics from a range of traditions, see Lopez,
Buddhist Hermeneutics, and “Buddhist Books and Texts:  Exegesis and Hermeneutics”;
Cabezon, Buddhism and Language:  A Study of Indo-​Tibetan Scholasticism; Nance, Speaking
for Buddhas: Scriptural Commentary in Indian Buddhism, and “Buddhist Hermeneutics”; and
Skilling, “Vasubandhu and the Vyākhyāyukti Literature.”
65. Collins, Nirvana and Other Buddhist Felicities, and “Remarks on the Visuddhimagga, and
on its Treatment of the Memory of Former Dwelling(s) (pubbenivāsānussatiñāṇa).”
66.  Walters, “Suttas as History:  Four Approaches to the Sermon on the Noble Quest
(Ariyapariyesanasutta),” particularly 266–​72.
67. Endo, Buddha in Theravada Buddhism, and Studies in Pāli Commentarial Literature.
 Introduction 27

and to center on the issues he thinks are important. It is by no means an exhaus-

tive account (whatever that could mean in a textual encounter that he himself
insists is immeasurable). I see this book as an effort to practice a style of textual
studies that yields to the philological theories and apparatus of the ancient author
under study, rather than imposing a reading style shaped principally by modern
western hermeneutical, philological, and philosophical theoretical assumptions
and practices (though I  cannot avoid these entirely, of course). It is not that
I think that studies that bring modern theoretical paradigms to ancient texts are
illegitimate or not valuable—​indeed, to hold that view would be to reject most
modern work on texts. Rather I want to try something different here in an effort
to, wherever possible, explore Buddhaghosa’s writings according the theoretical
paradigms he describes.
There are many things this book is not. I  do not systematically outline all
formal principles of Pali exegesis (many of these have been ably discussed by
others68), though I do offer an account of many of the exegetical protocols that
Buddhaghosa discussed and deployed to make sense of scripture. My efforts
are not historical: I do not attempt to uncover the dates and chronology of the
early Pali literature; explore the histories of oral and written transmission; en-
gage questions of scriptural authenticity, authority, and authorship; investigate
the chronology or development of hermeneutical methods in the tradition; or
take up many of the other important tasks to which other textual scholars have
contributed much good work.69 While noting that the Mahāvihāra tradition
(what we now call the Theravada) was forged in a complex historical environ-
ment in rivalry with other schools, I am not here concerned to locate its concep-
tion of texts among the varied competing Buddhist schools and movements, or
to discern the contours of early Buddhism’s overall approach to scripture. Nor
am I  focused on the Mahāvihāra tradition’s own historiography or historical
narrative, nor on Buddhaghosa’s successors and their reception of his texts. I do
not explore the layers of subcommentary in Pali and vernacular languages that
discussed and elaborated Buddhaghosa’s commentaries, as important as such a
project would be. My ambitions have been to focus in a single-​minded way on

68. See Ñāṇamoli, The Guide, and Bond, In the Buddha’s Words.

69. For surveys and histories of Pali texts, see Adikaram, Early History of Buddhism in Ceylon;
Norman, Pāli Literature; Lamotte, History of Indian Buddhism, 149–​90; Warder, Indian
Buddhism; von Hinüber, A Handbook of Pāli Literature. On Pali oral literature, see Cousins,
“Pali Oral Literature.” On questions of the authority of Buddha’s words, see Bond, The Word
of the Buddha, ch. 1; Lamotte, “Assessment of Textual Interpretation in Buddhism”; and
Davidson, “Appendix: An Introduction to the Standards of Scriptural Authenticity in Indian
Buddhism.” I discuss previous work on hermeneutics and scholasticism later.
28 In t rodu ct ion

this one thinker—​albeit one whose work and significance are so monumental
and my efforts to understand them are still so tentative that I have yet to over-
come the feeling that I am but a tiny hare tossed into a vast ocean scrambling for
a foothold (to borrow an image from Buddhaghosa). I center on what might be
called the conceptual, literary, and rhetorical dimensions of what Buddhaghosa
thought could happen in an encounter with Buddhist scripture, and how these
might in some instances have philosophical significance. I am interested in how
Buddhaghosa conceived of both scripture and commentary, what his definitions
of genre were, and how he thought scriptural language works and texts make
meaning. My method is to read with him to discover the methods he used and to
encounter the canon he saw.
Because of this, my arguments about what Buddhaghosa is doing develop cu-
mulatively and unfold over the whole of the book; a much larger edifice—​a larger
whole—​is slowly being built, and parts need to be explored in relationship to
this whole. Moreover, as Wittgenstein warned, “light dawns gradually over the
whole,” perhaps especially when taking up a polymathic thinker credited with
helping to construct a vast corpus. Part One lays out the building blocks of the
whole enterprise (­chapter  1 exploring claims of the Buddha’s omniscience and
the immeasurability of his words, and ­chapter 2 describing the main interpreta-
tive distinctions that later bear some of the weight of certain arguments about
Buddhist knowledge articulated in the piṭakas). Part Two (­chapters 3–​5) considers
Buddhaghosa’s treatment of each of the three piṭakas to build on and contrast
with one another points of emphasis and styles of thought. In addition to the
additive and layered nature of this exploration, I am also committed to tracing
the modular nature of Buddhaghosa’s treatment of the Dhamma, which entails
attention to context and purpose across the larger edifice. Buddhaghosa is an au-
thor who may particularly suffer from the tendencies of busy modern scholars
to drop in on specific discussions and lift out bits of doctrine shorn of the very
contexts in which they were at work; scholars have often used the Visuddhimagga
principally as a reference tool rather than reading it for the intellectual edifice it
is attempting to construct.70 Dipping into the text to look up definitions, lists, or
points of doctrine risks obscuring the work these are doing within a larger net-
work of practices. One might assume that an abstracted passage is speaking in
generalist terms or making metaphysical claims, when in fact it could be working
within a particular context as a method for a particular purpose.

70.  Norman describes the Visuddhimagga as “not so much a commentary as a sort of

encyclopaedia, a compendium of Buddhist doctrine and metaphysics” (Pāli Literature,
120). Geiger, too, sees the Visuddhimagga as an “encyclopaedia” of Buddhist doctrines (Pāli
Language and Literature, 29–​30).
 Introduction 29

Let me state in brief here what we investigate in detail in the chapters that
follow. In c­hapter  1, I  explore Buddhaghosa’s introduction to buddhavacana,
which centers on the three piṭakas, conceived as methods and competencies in
learning that give glimpses of the immeasurability of the Buddha’s omniscient
ken. I show how the commentators developed their claim of this omniscience,
noting that they did not do so via a formal philosophical argument, but rather
through trying to discover how buddhavacana is, in principle, endless. Chapter 2
considers the genres (the three piṭakas and the genre of aṭṭhakathā) and the var-
ious discourses, registers of speech, and modes of intellection and pedagogy the
commentators identified in buddhavacana, and then deployed as distinctions
useful for interpretation. These include the ways the teaching can be taught in
brief and in detail (saṅkittena/​vitthārena), with meaning and with phrasing
(attha/​byañjana), in conventional uses of language and uses that get at furthest
meaning (sammuti/​paramattha), definitively and in ways requiring further in-
terpretation (nītattha/​neyyattha), and contextually and categorically (pariyāya/​
nippariyāya). While all of these are important for his exegetical work and are
returned to throughout the book, Buddhaghosa developed this last distinction
between contextual and categorical modes of discourse as vital for interpreting
the Suttanta and Abhidhamma, respectively. To trace how this distinction works
in practice, ­chapter  3 takes up the highly contextual form of knowledge the
commentators saw at work in Suttanta knowledge by focusing on the nidāna, the
introductory framing of a sutta. This chapter explores how Buddhaghosa drew
out the narrative, contextual, and particularist conception of buddhavacana and
its expression of the Buddha’s omniscience. Chapter 4 considers his view of the
abstract modes of analysis that constitute the Abhidhamma, and how he saw its
methods as enacting the infinite nature of the Buddha’s omniscience in a different
form. Finally, c­ hapter 5 takes up the commentarial reading of the Vinaya Piṭaka
in how it explores the way that the monastic rules demonstrate the Buddha’s in-
finite knowledge, in particular, of past, present, and future. I have included three
appendices that give translations of substantial selections of commentaries on
which I draw heavily in the book, and which the reader may want to read to get a
sense of Buddhaghosa’s range and commentarial style.
 PART I

Building Blocks for

an Interpretative Program
 1

The Buddha’s Omniscience and

the Immeasurability of Scripture

Philological disciplines of making sense of texts are not the same

everywhere. While, in our own time, philological practice has centered on literary
and historical criticism, including producing critical editions and studying the
authorship, dating, and historical provenance of texts, Buddhaghosa’s concerns
were substantially different. To be sure, he was interested in the authority and
authenticity of canonical sources, and his editorial practices may have been crit-
ical,1 but he did not share modern concerns for the dating and chronology of
texts or the same notions of history that underlie modern philology. Rather, the
exegetical tradition he inherited and helped to shape, as the pages that follow
describe, was disciplined by certain claims about texts, language, genre, and in-
tellectual practice that were informed by distinctive ideas concerning the quality
and nature of the Buddha’s words (buddhavacana).
This chapter explores these claims and ideas to discern Buddhaghosa’s
theory of texts. I see this project as twofold: First, what did he think were the
conditions for the scriptural texts he called buddhavacana? And second, what are
the conditions for understanding buddhavacana? To the first question, I argue
that to understand Buddhaghosa’s theory of scripture we need to see how his
conception of buddhavacana rested on and developed a buddhology that made
exalted claims about the Buddha, above all, that he was omniscient in the sense
of possessing unobstructed and immeasurable knowledge. These qualities are
said to be evident in his “well-​spoken words.” To the second question, I demon-
strate how Buddhaghosa describes the reception of certain notable texts both in

1. See Endo, Studies in Pāli Commentarial Literature, for analysis of Buddhaghosa’s editorial
work and other modern philological discussions of his authorship and dating of particular texts.
34 Build ing Blo cks for an In t er pretat ive Pro gr a m

the Buddha’s own day and in the transmission of those texts to the present and
future. What does it mean to become learned in buddhavacana? While the next
chapter discusses specific disciplinary strategies for making sense of the varieties
of discourse that constitute scripture by focusing on his actual hermeneutical
protocols and distinctions, this chapter explores at a more general level the
intellectual conditions Buddhaghosa anticipated for grasping buddhavacana.
I see the chief conceptual challenges Buddhaghosa discerned in buddhavacana
and its interpretability as centered on how textual knowledge stands in relation-
ship to Buddha’s omniscient knowledge. How can texts convey the omniscient
ken of the Buddha, which in turn grasps the immeasurable Dhamma? And how
can these be grasped by those who would seek to understand it?

Buddhavacana
While the term buddhavacana is seldom mentioned in the canonical sources
as such—​they speak instead of the “Dhamma”—​buddhavacana is a category
Buddhaghosa inherited, helped to fashion, and made central to his under-
standing of scripture. In the tradition he received, buddhavacana was a quantifi-
able body of material, subject to various kinds of description. It was, as he himself
recounts, the material recited at the First Council, when under the leadership
of Mahākassapa, Upāli recited the Vinaya and Ᾱnanda recited the Dhamma.2
Having described this first recitation and codification of the scriptures, he defines
the First Council’s collected material thus:

All this is understood as the Buddha’s word (buddhavacana), which is

single in taste, twofold in the Dhamma and Vinaya, threefold in beginning,

2. His account of the First Council occurs in Sv i.2.15, Sp i.4–​16, and Pj I 89–​98, a narrative
based on the Vinaya’s account (Vin ii.285–​290), and which also borrows heavily from the
chronicles. Buddhaghosa’s passage has been quite well worked over in modern scholarship (see
Jayawickrama, trans., The Inception of the Discipline, for a translation of the Samantapāsādikā’s
treatment of the first three councils of the Mahāvihāra tradition, and Jayawickrama’s discussion
of them). As Lamotte has said, “few historical problems have caused so much ink to flow as that
of the Buddhist councils,” though his own discussion is very much worth looking at (History of
Indian Buddhism, 124). For helpful treatments of the substantial scholarship on the councils,
see Collins, “On the Very Idea of the Pāli Canon,” and Hallisey, “Councils as Ideas and Events
in the Theravāda.” It is of course important to note that the Mahāvihāran historical narrative
of the canon, the First Council, and its own unbroken and uniquely authentic inheritance of
these as constructed in its chronicles and commentaries was produced in a context of sectarian
rivalry and, despite its ultimate triumph, likely represents a minority view in its own time.
On this point, see Walters, “Mahāyāna Theravāda and the Origins of the Mahāvihāra,” and
“Buddhist History: The Sri Lankan Pāli Vaṃsas and Their Commentary,” and Skilling et al.,
How Theravada Was the Theravada?
 Omniscience and Immeasurability 35

middle, and end, and [threefold] likewise in terms of the piṭakas, fivefold
in terms of the nikāyas, ninefold in terms of its parts, and composed of
84,000 units of Dhamma.3

This definition of the corpus, which he says was arranged in this manner at the First
Council itself,4 is the framework in which he considers the ways buddhavacana
may be classified. It is instructive to see how he parses these different divisions in
this definition.
First, buddhavacana has a “single taste”:  “everything said by the Bhagavan,
whether by contemplation or by teaching to deities, humans, nāgas, or yakkhas, et
cetera, in the forty-​five-​years between having attained highest perfect Awakening
up to the final nibbāna without remainder, has one taste—​the taste of freedom.”5
This claim about the “whole” unites the diverse teachings around both a single
teleological aim and a single reception—​the same message of freedom went out
to everyone and its taste was everywhere the same.
The twofold distinction of Dhamma and Vinaya can allow for the inclusion
of both Abhidhamma and Suttanta as part of the “Dhamma.” The first threefold
distinction is the tripartite beginning, middle, and end in the words spoken in
his teaching career of expounding buddhavacana flanked by two great trees: that
is, it started with an utterance made under the Bo Tree and ended with his final
words at his death under a Sal tree.6 (Trees are everywhere marking important
achievements and occasions.) Buddhavacana is also threefold defined according to
the three piṭakas, and here Buddhaghosa notes that this includes material “both

3.  Sv i.15; Sp i.16 (cf. As 18):  Tadetaṃ sabbampi buddhavacanaṃ rasavasena ekavidhaṃ,
dhammavinayavasena duvidhaṃ, paṭhamamajjhimapacchimavasena tividhaṃ; tathā
piṭakavasena, nikāyavasena pañcavidhaṃ, aṅgavasena navavidhaṃ, dhammakkhandhavasena
caturāsītisahassavidhanti veditabbaṃ.
4. Sv i.25; Sp i.29; As 27.
5.  Yañhi bhagavatā anuttaraṃ sammāsambodhiṃ abhisambujjhitvā yāva anupādisesāya
nibbānadhātuyā parinibbāyati, etthantare pañcacattālīsavassāni devamanussanāgayakkhādayo
anusāsantena paccavekkhantena vā vuttaṃ, sabbaṃ taṃ ekarasaṃ vimuttirasameva hoti (Sv i.15;
Sp i.16).
6.  Sv.i.16; Sp i.17; As 18. Buddhaghosa notes some dispute about exactly which of two
possibilities the first utterance of buddhavacana consisted in: according to the Dhammapada
reciters it is Dhammapada 153–​54, but Buddhaghosa also says that the first utterance is a verse
from Udāna 1. He resolves this in Pj I.13 by saying that the Dhammapada verse was first uttered,
but in the mind (manasāva vuttavasena), and the other was actually vocalized. This section in
its Atthasālinī version is translated in full in Appendix C.
36 Build ing Blo cks for an In t er pretat ive Pro gr a m

recited and not recited at the First Council” headed by Mahākassapa.7 (We will
examine below some of what he might have had in mind about the material not
recited, and yet included in the resulting collection.) He lists here all the suttas
that make up the four nikāyas (Dīgha, Majjhima, Saṃyutta, and Aṅguttara) to-
gether with the fifteen books that constitute the Khuddaka Nikāya; the seven
books of the Abhidhamma; and the sections that constitute the Vinaya (two
Pātimokkhas, two Vibhaṅgas, twenty-​two Khandhakas, and sixteen Parivāras)—​
a collection of material that modern scholars call “the Pali canon.” This division is
the classification that he is most interested in and which he spends the most time
on, and we shall examine at length his treatment of the genres of knowledge the
piṭakas convey below.
The piṭaka classification can be swallowed by one of its own internal categories
when buddhavacana is said to be fivefold. This means that the whole buddhavacana
can be divided among the four nikāyas (Dīgha, Majjhima, Saṃyutta, and
Aṅguttara) and the fifth, “Miscellaneous Collection” (Khuddaka Nikāya). In this
division, the Vinaya and Abhidhamma piṭakas can be included in the Khuddaka
in addition to its fifteen books: “Which is the Khuddaka Nikāya? The Buddha’s
words except the four nikāyas are established as the entire Vinaya, Abhidhamma,
and the fifteen books starting with the aforementioned Khuddakapāṭha.”8 Here a
category for miscellaneous items is useful to sweep up and include much sundry
material, not least the two other piṭakas, though of course elsewhere we find the
more familiar division of the three piṭakas.
Buddhaghosa goes on to say that buddhavacana may also be divided among
an ancient ninefold division of the Dhamma, which was a stock canonical
account of genre: teachings, recitations, expositions, verses, inspired utterances,
quotations, birth stories, teachings of wonders, and questions-​and-​answers.9
While mentioned in the four nikāyas as a classification of the Dhamma, it is only

7. Sv i.17; Sp i.17: Tattha paṭhamasaṅgītiyaṃ saṅgītañca asaṅgītañca sabbampi samodhānetvā;

note, though, that elsewhere he states that all of buddhavacana was recited at the First
Council:  evametaṃ sabbampi buddhavacanaṃ pañcasatikasaṅgītikāle saṅgāyantena
mahākassapappamukhena (As 27).
8.  Sv i.23; Sp i.27; As 26:  Katamo khuddakanikāyo? Sakalaṃ vinayapiṭakaṃ
abhidhammapiṭakaṃ khuddakapāṭhādayo ca pubbe nidassitā pannarasabhedā ṭhapetvā cattāro
nikāye avasesaṃ buddhavacananti. His list of Khuddaka books includes:  Khuddakapāṭha,
Dhammapada, Udāna, Itivuttaka, Suttanipāta, Vimānavatthu, Petavatthu, Theragāthā,
Therīgāthā, Jātaka, Niddesa, Paṭisambhidā, Apadāna, Buddhavaṃsa, and Cariyāpiṭaka (Sp
i.18; As 18; Sv 1.17; KhpA 11); it does not include, notably for our purposes, the hermeneutical
texts the Nettippakaraṇa and the Peṭakopadesa (though the Burmese tradition adds these and
the Milindapañha to the Khuddaka texts).
9. Idhekaccassa buddhavacanaṃ ariyāputaṃ hoti suttaṃ geyyaṃ veyyākaraṇaṃ gāthā udānaṃ
itivuttakaṃ jātakaṃ abbhutadhammaṃ vedallaṃ (Nidd II, 118, 144; Mil 341, 345, et cetera.).
 Omniscience and Immeasurability 37

in a few of the later Khuddaka texts that the ninefold division is equated with
buddhavacana, as such. This classification is nowhere explained in the canon
itself, but its frequency in a stock phrasing describing the Dhamma may have
obliged Buddhaghosa to at least mention it here. Scholars have suggested that
though Buddhaghosa attempts to classify particular parts of the Suttanta into
these divisions it is clear that the full significance and utility of this classification
is lost on him.10
The notion that buddhavacana contains 84,000 units (khandha) of dhamma
is an echo of Ᾱnanda’s words, quoted not from the First Council as we might
expect, but rather from Ānanda’s biographical poem recorded in the Theragāthā,
where he stated, “I learned 82,000 from the Buddha, and 2,000 from the monks,
thus mastering 84,000 [units of ] the Dhamma.”11 Buddhaghosa elaborates that
units in the Suttanta are divided per connection or sequence (anusandhika) or,
in suttas with a question-​answer format, where the question constitutes one unit
and the answer a second; in the Abhidhamma, each analysis of a dyad or triad,
or each analysis of a thought-​process, constitutes a unit; and in the Vinaya, units
are the listing of subjects, the analysis of terms, offenses, determinations of inno-
cence, offenses in the supplementary section, and in the threefold classification
of offenses.12 Perhaps most important to note about Ᾱnanda’s claim is that the
words of others are also included in buddhavacana. As we will see, words inspired
by the Buddha yet spoken by his disciples and subsequent elders have, from the
time of canonical sources themselves, been regarded as authentic teachings of the
Buddha.
As mentioned earlier, Buddhaghosa asserts that all of these classifications
were arranged and recited in the full seven-​month period of the First Council.
This manner of defining buddhavacana emphasizes a very systematic collection
of textual material that benefits from various ways of classifying it; through
such classifications, each category can be interpreted variously (as for example,
Abhidhamma is sometimes best conceived as a piṭaka, sometimes as part of

References to the ninefold distinction are common in the four nikāyas, including A ii.6–​7; M
i.133, et cetera.
10. See Jayawickrama, The Inception of the Discipline, 25–​26, 101–​2, Norman, Pāli Literature,
15–​17, and Lamotte, History of Indian Buddhism, 143–​45. Buddhaghosa’s treatment of it can be
found in several places: Sp i.28; Sv i.23; As 26. The ninefold Dhamma is found many places in
the canonical sources, as for example, M.ii.105–​6. It is not, however, an important hermeneu-
tical distinction for the Netti or Peṭ, and the Dīpavaṃsa records how it came to be replaced at
the First Council by the piṭaka classification (Norman, Pāli Literature, 16).
11. Sp.i.29; Sv.i.24; As 27, quoting Th 92.
12. Sp i.29; As 27; Sv i.24.
38 Build ing Blo cks for an In t er pretat ive Pro gr a m

the Khuddaka Nikāya, sometimes as exposition [veyyākaraṇa], and sometimes

according its number of units—​all of which, Buddhaghosa suggests, support its
authenticity as buddhavacana and help us understand it).13 This style of modular
classification, where an item can be understood variously by placing it in different
categories, is pervasive, and Buddhaghosa often deploys it. In his own analysis,
he finds the threefold piṭaka classification most useful as he considers how the
piṭakas work as genres and methods of study.
As much as this account of buddhavacana indicates a closed body of scripture,
Buddhaghosa also thought that what was laid down as buddhavacana at the First
Council is, in both principle and practice (as I  demonstrate extensively later),
immeasurable and unbounded, and thus something that resists containment.
Buddhaghosa develops a further hermeneutic stance about what Buddhist scrip-
ture is: the texts are at once precise collections while yet also giving unbounded
expression to Buddha’s omniscient ken. The immeasurability of the Buddha’s
words meant that they can be, under certain circumstances, expanded. For imme-
diate and practical purposes, this notion of expansion became a useful idea for in-
cluding in buddhavacana material that clearly was not in any obvious way recited
at the First Council, as we shall consider shortly. Further, the idea of expanding
the immeasurable nature of the Buddha’s words becomes an important idea for
interpreting the nature of commentary, pedagogy, and learning—​that is, the
practices of understanding that can begin to plumb that immeasurable ken.

The Immeasurability of the Teaching

There is considerable canonical precedent for the suggestion that the Buddha’s
teaching is in some important sense immeasurable and unbounded. First, the
linguistic forms or phrasings (byañjana) of the teaching are said to be immeas-
urable. We have already considered the “immeasurable words and immeasurable
phrasings” in the Tathāgata’s immeasurable teachings of the Dhamma.14 And the
noble truth of suffering is said to have “immeasurable expressions, immeasurable
phrasings, and immeasurable illustrations.”15 Buddhaghosa takes this latter claim
to mean that there are endless ways—​in terminology, wording of expression,

13. As 27–​28.
14.  Tattha aparimāṇā padā aparimāṇā byañjanā aparimāṇā tathāgatassa dhammadesanā
(A ii.182) and tathāgatassa dhammadesanā, apariyādinnaṃyevassa tathāgatassa
dhammapadabyañjanaṃ (M i.83). See “Introduction.”
15.  S v.430:  Tattha aparimāṇā vaṇṇā aparimāṇā byañjanā aparimāṇā saṅkāsanā—​‘itipidaṃ
dukkhaṃ ariyasacca’nti.
 Omniscience and Immeasurability 39

and illustration—​that the Buddha could preach the truth of suffering:  “when
this truth is extended in detail by every mode, one by one, there is no end to the
expressions.”16 The Buddha’s facility with language, it seems, allowed him to ex-
press ideas in endless linguistic forms, and suffering in particular is immeasurably
expressible.
As much as the linguistic form—​the phrasing (byañjana)—​of the teaching
may be expanded immeasurably, so too, its meaning (attha). When the Buddha
spoke he could not exhaust either the meaning or the phrasing of the Dhamma.
In fact, when it comes to analytical understanding, “it is implausible and impos-
sible that one who possesses the four kinds of analysis (paṭisambhidā) would
come to the end of either meaning or phrasing.”17 Paṭisambhidā is analysis,
making distinctions. In Buddhaghosa’s understanding, the four kinds of anal-
ysis are the analysis of “things” (or “meaning” or “purpose,” attha), the analysis
of phenomena (dhamma) [or, in some places, the spoken Dhamma], the analysis
of language (nirutti), and the analysis of knowledge (paṭibhāna). Very briefly and
generally, the analysis of things is the capacity to discern phenomena that have
been caused or conditioned: to analyze the “effect of a cause” (hetuphala). The
analysis of dhammas is discerning conditions (paccaya). The analysis of language is
the capacity for discriminating how language is being used. And analysis of knowl­
edge is a reflexive or meta-​analytic capacity to review the previous three.18 These

16. Here is the full translation of the relevant commentary on S v.430: “ ‘Expressions are immeas-
urable’ means that ‘the syllables’ [of the text] are immeasurable. ‘Phrasing’ is just a synonym of
that, or phrasing is one part of an expression. ‘Illustration’ is a variety [of it]. When this truth
[of suffering] is extended in detail by every mode, one by one, there is no end to the expressions,
et cetera.” (aparimāṇā vaṇṇāti appamāṇāni akkharāni. Byañjanāti tesaṃyeva vevacanaṃ,
vaṇṇānaṃ vā ekadesā yadidaṃ byañjanā nāma. Saṅkāsanāti vibhattiyo. Ekamekasmiñhi sacce
sabbākārena vitthāriyamāne vaṇṇādīnaṃ anto nāma natthi [Spk iii.298]). A  similar idea is
conveyed in Mp iii.166 on A ii.182.
17. Aṭṭhānametaṃ, bhikkhave, anavakāso yaṃ catūhi paṭisambhidāhi samannāgato atthato vā
byañjanato vā pariyādānaṃ gaccheyyā’’ti (A ii.139). Lamotte suggests that as a general prin-
ciple “the meaning is single and invariable, while the letter is multiple and infinitely variable”
(Lamotte, “The Assessment of Textual Interpretation in Buddhism,” 13). But the two passages
(S iv.281, 297) he cites in support of this claim need not be taken as general or universalizable
claims about meaning and phrasing, but rather as a preference for a method in which partic-
ular words and phrasings are taken this way in these two cases. I discuss this matter in the next
chapter.
18. See Vism 441 (XIV.22–​3) and Vibh-​a 386–​88; “dhamma” can also mean the spoken, scrip-
tural teachings (Vism 441, XIV.24). See also the discussion of the four paṭisambhidā, which
were variously understood in the textual sources, in Aung and Rhys Davids, trans., Points of
Controversy, 377–​82. Ronkin (Early Buddhist Metaphysics, 87–​88) discusses the differences in
how these terms are understood in the Paṭisambhidāmagga and the Vibhaṅga. Nance gives sub-
stantial treatment of these four analyses (or “discriminations”) in Sanskrit and Tibetan sources
(Speaking for Buddhas, 55–​65), which differ from how they are interpreted in the Pali sources,
40 Build ing Blo cks for an In t er pretat ive Pro gr a m

forms of analysis are important for our story because they indicate the idea that
analysis—​of things, of phenomena, of language, and of reflexive processes—​is
the way that the Buddha’s teachings are expanded. In the case of the first two,
analyzing the phenomenal world and its causal processes becomes the potentially
endless application of the teachings; and in the case of the second two, analyzing
language and reflexive processes becomes the endless work of study, exegesis, and
thought. We consider these types of analysis further in the next chapter.
But we must return to claims about the immeasurability of the Buddha’s
teachings and how and where these are made. At the commentarial level,
Buddhaghosa is particularly interested in how the Buddha, when thinking
through the Dhamma in his mind, cannot find an end of it. In one particu-
larly vivid and extended illustration of this idea developed in the Atthasālinī,
Buddhaghosa describes the fourth week after the Buddha’s first night of awak-
ening, as he sat and explored with his omniscience the content of the seven books
that became the Abhidhamma Piṭaka (see Appendix C). The Buddha wonders
“how large is the Dhamma [that he] contemplated for seven days and nights?”
and answers, “the Dhamma [contemplated in] the mind” is “endless and immeas-
urable.”19 The conundrums apparent in this situation become a source of further
wonder to Buddhaghosa: how can something immeasurable be communicable in
a finite amount of time? But it is not to be said that “the Teacher, having put into
speech the Dhamma thought out by the mind for seven days, teaching even for a
hundred years, a thousand years, or a hundred thousand years, having reached its
summit, is not able to teach it.”20 And later he was able to convey this immeasur-
able content in a three-​month period to his mother in one of the heavens. At that
time, he taught the Abhidhamma at the root of a heavenly coral tree in Tāvatiṃsa
heaven, where “the teaching set forth without interruption for three months was
endless and immeasurable as it flowed rapidly like the heavenly Ganges and like
water rushing from an upside-​down pot.”21

but he notes similarly that the four discriminations facilitate “the inexhaustibility and limitless-
ness of the teaching of the dharma” (57).
19. As 15: satta rattindivāni sammasitadhammo kittako ahosīti? Ananto aparimāṇo ahosi. Ayaṃ
tāva manasādesanā nāma.
20.  As 15:  satthā pana evaṃ sattāhaṃ manasā cintitadhammaṃ vacībhedaṃ katvā desento
vassasatenapi vassasahassenapi vassasatasahassenapi matthakaṃ pāpetvā desetuṃ na sakkotīti na
vattabbaṃ.
21. As 15: Tayo māse nirantaraṃ pavattitadesanā vegena pavattā ākāsagaṅgā viya adhomukha
ṭhapitaudakaghaṭā nikkhantaudakadhārā viya ca hutvā anantā aparimāṇā ahosi. The idea
that the Buddha taught his mother the Abhidhamma in heaven seems to have been held
only by the Mahāvihārans; see Skilling on this tradition (“Dharma, Dhāraṇī, Abhidharma,
Avadāna: What Was Taught in Trayastriṃśa?,” 51–​53).
 Omniscience and Immeasurability 41

From here Buddhaghosa feels compelled to elaborate the qualities of

the Buddha’s speech that it might be executed so quickly. He now shifts to
observations of all buddhas’ speech to suggest that they can communicate a great
deal quite rapidly:

For a teaching of buddhas making a blessing at the time of blessing a meal,

if expanded just a little, is the length of the Dīgha and Majjhima Nikāyas.
And the teaching of those teaching the Dhamma to an assembly gathered
at the end of a meal is the length of the two large nikāyas, the Saṃyutta
and Aṅguttara together. How is this possible? The life spans of buddhas
are fleeting, and [so] their teeth touch well, their mouths flow smoothly,
their tongues are supple, their sounds, sweet, and their words, quick.
Therefore, so much Dhamma can be taught in a moment. However, the
Dhamma taught in the three months is still endless and immeasurable.22

Part of the answer to how so much could be conveyed by the Buddha is the
quickness of his elocution. In light of how quickly teachings comprising the
nikāyas can be conveyed—​the length of a Dhamma talk after a meal—​the three
months it took him to preach the Abhidhamma to his mother seem like a very long
time by contrast, but one appropriate for the endlessness of the Abhidhamma,
which in this discussion is his main theme.
Surely this only raises further questions. How can such a teaching as the
Abhidhamma be grasped, and then conveyed, by others? Here Buddhaghosa
allows that even so erudite a person as Ᾱnanda (“he who heard much”), a master
of all three piṭakas, “who could learn, recite, or teach 1,500 verses and 60,000
words standing there as easily as if he was plucking flowers from vines,” even he
“studying a hundred or a thousand years would not be able to reach the end of
the teaching taught by the Teacher in that way in the three-​month period.”23 Then
how and to whom was it conveyed on earth? It seems that at the time he was
giving the teaching in heaven, it became necessary for the Buddha to take breaks

22.  As 15:  Buddhānañhi bhattānumodanakālepi thokaṃ vaḍḍhetvā anumodentānaṃ desanā

dīghamajjhimanikāyappamāṇā hoti. Pacchābhattaṃ pana sampattaparisāya dhammaṃ
desentānaṃ desanā saṃyuttaaṅguttarikadvemahānikāyappamāṇāva hoti. Kasmā?
Buddhānañhi bhavaṅgaparivāso lahuko dantāvaraṇaṃ suphusitaṃ mukhādānaṃ siliṭṭhaṃ
jivhā mudukā saro madhuro vacanaṃ lahuparivattaṃ. Tasmā taṃ muhuttaṃ desitadhammopi
ettako hoti. Temāsaṃ desitadhammo pana ananto aparimāṇoyeva.
23.  As 15–​16:  Ānandatthero hi bahussuto tipiṭakadharo pañcadasa gāthāsahassāni saṭṭhi
padasahassāni latāpupphāni ākaḍḍhanto viya ṭhitapadeneva ṭhatvā gaṇhāti vā vāceti vā deseti
vā. . . . satthārā temāsaṃ iminā nīhārena desitadesanaṃ vassasataṃ vassasahassaṃ uggaṇhantopi
matthakaṃ pāpetuṃ na sakkoti.
42 Build ing Blo cks for an In t er pretat ive Pro gr a m

for rest and refreshment (leaving a clone in heaven to carry on), and on one such
siesta he was attended by Sāriputta. Sāriputta was foremost among monks skilled
in analysis (paṭisambhidā), and because of this skill the Buddha was able to teach
him the Abhidhamma quite quickly.

Then the Teacher gave him the method (naya). He showed him,
“Sāriputta, so much Dhamma is taught by me.” When the Perfectly
Awakened Buddha was giving the method in this way to the chief disciple
accomplished in analysis (paṭisambhidā), there was the gift of the method
much like one pointing at what is seen by stretching out one’s hand while
standing at the seashore. And to the elder also the Dhamma, taught by the
Bhagavan with a hundred methods, a thousand methods, a hundred thou-
sand methods, became clear.24

This suggests that Sāriputta’s capacity for analysis made it possible for him to
grasp a method—​here configured as a gift—​conveyed through a gesture likened
to pointing at the ocean (a stock figure of endlessness). Sāriputta could himself
expand, through paṭisambhidā analysis, these methods in order to communi-
cate the immeasurability of the Abhidhamma. Abhidhamma is here represented
as methods that can be given briefly to those who know how to do the analysis
that can expand it. (It is noteworthy that mastery of the four types of analysis is
ascribed to Buddhaghosa himself in his early career as a monk, at least as he is
remembered in the later legends of his life).25
These details may seem fanciful, but if we slide over them too quickly we risk
missing something essential about Buddhaghosa’s conception of texts. First, he
approaches scripture with a literary, even poetic, sensibility, alert to the special
qualities of the Buddha’s speech whereby it conveys something infinite within its
limited forms. This literary aesthetic is not always noted in Buddhaghosa’s work,
but is an enduring theme of the interpretation that I develop in c­ hapters 3 and
5. Second, notable about this particular account is the idea that the Abhidhamma
conveys the immeasurable through analytical methods. The nature and

24. As 16: Athassa satthā nayaṃ deti. Sāriputta, ettako dhammo mayā desito’ti ācikkhati. Evaṃ
sammāsambuddhe nayaṃ dente paṭisambhidāppattassa aggasāvakassa velante ṭhatvā hatthaṃ
pasāretvā dassitasamuddasadisaṃ nayadānaṃ hoti. Therassāpi nayasatena nayasahassena
nayasatasahassena bhagavatā desitadhammo upaṭṭhātiyeva. The colophon of the Atthasālinī
also describes the teaching of the Abhidhamma to Sāriputta as giving him a teaching “by way
of method” (As 1: desetvā nayato).
25.  Ñāṇamoli, The Path of Purification, xxxvii, recounting the relevant passage in the
Buddhaghosuppatti.
 Omniscience and Immeasurability 43

implications of this reading of analytical texts are my focus in ­chapter 4. Third,

others can expand what the Buddha gestured to in brief, an idea important for
interpreting both scripture and commentary and on which we can venture a few
remarks here.
The idea that others, like Sāriputta, can expand teachings indicated briefly
by the Buddha goes back to canonical texts. Buddhaghosa mentions canonical
precedent when he comes to argue for the inclusion of the Kathāvatthu in the
buddhavacana recited at the First Council. He notes that certain texts could be
expanded by other people when the Buddha gave short, summary versions of
ideas in list form, rather like a table of contents. According to Buddhaghosa, the
Buddha did this in the case of the Madhupiṇḍika Sutta, which he gave via a brief
outline (uddesa) without going into detail about the meaning. The meaning was
then elaborated by the elder Mahākaccāna, whose expansion the Buddha later
approved.26 This example is helpful for him in accounting for how it is that the
Kathāvatthu could be included among the books of the Abhidhamma. The in-
clusion of this text attracted some skepticism, even on his own account, given
that it was composed 218  years after the Buddha’s parinibbāna, and thus long
after the First Council. In Buddhaghosa’s reckoning, the Buddha laid down the
“matrix” (mātikā) of this work, foreseeing that 218 years from his death the elder
Moggaliputta Tissa would elaborate it into the Kathāvatthu in the presence of
a great company of monks. It was not a matter of Tissa’s “own knowledge” but
rather the elaboration of the matrix given by the Buddha, and thus it qualifies
as buddhavacana.27 Note that his use of the word “matrix,” literally, the “point
of origin,” replaces the outline (uddesa) mentioned in the sutta example that he
marshals as precedent. In any case, this potency of the Dhamma, that it can be
ramified exponentially and endlessly from a single point of origin (mātikā), is
crucial, of course, in interpreting what Buddhaghosa took Abhidhamma to be.28
Even as it helps with the particular case of the Kathāvatthu, the principle of the
expansive potency of buddhavacana goes far beyond its value for substantiating
the authenticity of certain texts. It is also, as we have seen, key to interpreting the
Abhidhamma as a series of algorithmic analytical methods (“algorithmic” in the
sense of involving a small number of procedures that can be applied to recurrent

26. As 4–​6, citing M.i.110 of the Madhupiṇḍika Sutta. Mahākaccāna is considered foremost of
monks able to “expand the meaning in detail of what was said in brief ” (saṅkhittena bhāsitassa
vitthāreṇa atthaṃ vibhajantānaṃ [A i.23]).
27. As 4.
28. This is very much in keeping Gethin’s analysis of mātikā in “The Mātikās: Memorization,
Mindfulness, and the List,” 149–​72.
44 Build ing Blo cks for an In t er pretat ive Pro gr a m

problems). But perhaps most basically the idea that buddhavacana conveyed
something immeasurable and endless made it possible to read all of the texts with
an interpretative style that seeks in them, whether parts or wholes, the immeasur-
able, infinite, omniscient ken of the Buddha, known as the Dhamma itself.

Immeasurability and Omniscience

Embedded in the idea that the Buddha spoke words that were immeasurable
was, at least as the commentarial traditions interpreted it, a full-​blown doctrine
of the Buddha’s omniscience. The Buddha was, according to Buddhaghosa’s
buddhology, “omniscient” or “all-​knowing” (sabbaññū), and any interpretation
of his words or theory of scripture must be understood in relationship to this
claim. But what did omniscience mean and how did immeasurability come to
one of the most salient features of the Buddha’s words?
The idea that the Buddha was omniscient seems sometimes to have been
resisted by the Buddha himself as he is depicted in the canonical material.
The suttas treat with skepticism Jain assertions about their leader Mahāvīra as
“knowing all, seeing all, having complete knowledge and vision.”29 The Buddha
rebuts this as impossible, and ridicules the Jains for promoting it. And when
asked straight out if he possesses such knowledge, the Buddha disavows it on sev-
eral occasions, claiming instead to be a “three-​knowledge man.” That is, he knows
what he attained the night of final awakening when he achieved first, knowledge
of all of his previous births, then the knowledge of the rebirths of all beings, and
thirdly, the liberating knowledge attained from the absence of the taints.30 He
rejects the idea that anyone can know and see everything “all at once,” simultane-
ously.31 And Ᾱnanda argues that any teacher who claims to be omniscient is to be
regarded with suspicion, since omniscience, in the sense of complete “knowledge
and vision continuously and always present” is so easily refuted when that teacher
has to ask for directions to a certain town or explain why he went to a certain
place for alms but came away empty-​handed.32

29. M i.92–​93: sabbaññū sabbadassāvī aparisesaṃ ñāṇadassanaṃ paṭijānāti (and M ii.31–​32).

On early Jain claims of Mahāvīra’s omniscience and a comparison with Buddhist views, see
Jaini, “On the Sarvajñatva (Omniscience) of Mahāvīra and the Buddha.”
30. M i.482–​83; M ii.31–​32.
31.  M ii.127:  natthi so samaṇo vā brāhmaṇo vā yo sakideva sabbaṃ ñassati, sabbaṃ dakkhiti,
netaṃ ṭhānaṃ vijjatī’’’ti (M ii.127).
32.  M i.519:  satataṃ samitaṃ ñāṇadassanaṃ paccupaṭṭhita. This sutta, the Sandaka Sutta, is
discussed by Anālayo, “The Buddha and Omniscience,” 3–​4.
 Omniscience and Immeasurability 45

Yet elsewhere the Buddha claims to be a “knower of all” (sabbavidū), though

presumably not in the sense of knowing everything all at once.33 This particular
claim is made shortly after his awakening and having decided to teach, where he
declares it to the ascetic Upaka, whom he encounters on the road to Sarnath be-
fore giving his first sermon. Several scholars have suggested that such claims must
be understood in terms of what the Buddha meant by “all” here, which is defined
as the Buddha knowing all of his senses and their objects.34
For their part, the commentators were committed to asserting, expanding,
and promoting the omniscience of the Buddha, a doctrinal development that
appears to occur in tandem with the exegetical enterprise itself.35 For example,
we find Buddhaghosa in the same discussion as we have been considering from
the Atthasālinī, asserting that on the seventh day of awakening, seated under the
Bodhi Tree, the Buddha declared, “I have attained the knowledge of omniscience
on this throne.”36 How did Buddhaghosa interpret this?
The most systematic treatment of the Buddha’s omniscient ken in the canon-
ical sources is in the Paṭisambhidāmagga (The Path of Analysis). Buddhaghosa
knew this text well and quoted it frequently. The Paṭisambhidāmagga attempts
to define omniscience by listing the many topics over which the Buddha’s knowl­
edge ranges. It lists the many ways the Buddha’s knowledge can know past, pre-
sent, and future; know all sensory objects including mind and ideas; know all
things formed and unformed; grasp the extent of the meaning of impermanence,
dukkha, no-​self, and the four truths; see beings’ underlying biases and tendencies;
know the worlds, and so on.37 In this formulation, the Buddha’s omniscient
knowledge is described by listing the kinds of knowledge that he possesses, a

33. M i.171; M ii.93: in this sutta, Buddha claims to be a “knower of all” (sabbavidūham asmi)
and hesitates to teach, since his knowledge may not find a recipient who can understand it. The
later text, the Jātaka-​nidānakathā, claims he attained omniscience (sabbaññutañāṇa) on the
night of enlightenment (as cited in Jaini, “On the Sarvajñatva (Omniscience) of Mahāvīra and
the Buddha.” 103). The Kathāvatthu (Kv 228) claims that the Buddha was “all knowing and all
seeing” (sabbaññūsabbadassāvī ).
34.  S iv.15; see Anālayo, “The Buddha and Omniscience,” 9 and Kalupahana, A History of
Buddhist Philosophy, Continuities and Discontinuities, 43.
35. A parallel development occurs in the ways postcanonical Jain treatises come to introduce
themselves, where “the truth that the omniscient one once fathomed remains the corner-
stone of the treatise in question,” according to Balcerowicz (“Some Remarks on the Opening
Sections in Jaina Epistemological Treatises,” 28).
36. As 12: imasmiṃ vata me pallaṅke sabbaññutaññāṇaṃ paṭividdha’nti. See Endo, Buddha in
Theravada Buddhism, 58–​78, for a description of canonical and commentarial claims about the
Buddha’s omniscience.
37. Patiṣ i.131–​34, chs. 72–​73. See Ñāṇamoli, trans., The Path of Discrimination, 131–​34.
46 Build ing Blo cks for an In t er pretat ive Pro gr a m

listing that enumerates the categories of Abhidhamma analysis that are, in effect,
the fundamental points of doctrine and the methods for its elaboration. These he
can know without limitation.
Buddhaghosa (or his team if it was not he who worked directly on the
Paṭisambhidāmagga commentary, as modern scholars doubt) emphasizes in
the commentary that Buddha’s omniscience meant “unobstructed knowledge”
(anāvaraṇañāṇa) into “all things” (sabbadhamma)38:  “he is called omniscient
by acquiring unobstructed knowledge, not by knowing everything all at once.”39
Toshiichi Endo has also demonstrated that the Pali commentators’ interpre-
tation of Buddha’s omniscience was that his knowledge will arise in an “unob-
structed” manner: to whichever aspect of reality he attends, whether something
in the past, present, or future, or the extent of other beings’ biases and tendencies,
he will know and apprehend it without any obstruction or barrier whereby the
knowledge could come up short or stop.40 The Paṭisambhidāmagga commentary
emphasizes that the Buddha “knows everything constructed and not constructed
without remainder, where ‘everything’ here means the taking up completely and
entirely of all things by way of kind.”41 The Buddha is said to be “All-​Seeing”
(samantacakkhu), another way of indicating his omniscience.42 The importance
of seeing all is as critical as knowing all.
While for the most part the Pali commentators eschewed formal epistemo-
logical theorizing about omniscience of the sort that we see develop in the philo-
sophical Mahāyāna traditions,43 Dhammapāla takes up some of the philosophical
challenges presented by claims of omniscience, arguing, for instance, that this
omniscience must be seen as a type of potentiality (samatthatā); it cannot be
knowledge of all dhammas simultaneously, since to apprehend everything at
once is to not apprehend things singly and distinctly (and so it cannot logically
be simultaneous knowledge of parts and wholes). Yet we cannot say the Buddha

38. Paṭis-​a ii.428–​29.
39. Paṭis-​a ii.429: Anāvaraṇañāṇapaṭilābhenapi sabbaññūti vuccati, na ca sakiṃsabbaññūti.
40. Endo, Buddha in Theravada Buddhism, ch. 2. See also Dhammapāla’s commentary on the
Visuddhimagga, as quoted by Ñāṇamoli, trans., The Path of Purification, 772–​77, n. 7; a similar
passage is also in ItiA i.139–​41); see also Jaini, “On the Sarvajñatva (Omniscience) of Mahāvīra
and the Buddha,” 113–​15.
41.  Paṭis-​a ii.429 on Paṭis i.131:  Sabbaṃ saṅkhatamasaṅkhataṃ anavasesaṃ jānātīti ettha
sabbanti jātivasena sabbadhammānaṃ nissesapariyādānam.
42. Paṭis-​a ii.429.
43.  On which, for example, see McClintock, Omniscience and the Rhetoric of Reason, and
Perrett, “Omniscience in Indian Philosophy of Religion.”
 Omniscience and Immeasurability 47

can hold in his mind at once all dhammas in succession either, since knowables
are infinite. Further, one cannot say he knows everything that it is important to
know through direct perception, but then infers the rest, since inferential knowl­
edge is accompanied by doubt, unlike the Buddha’s knowledge.44 But in the end,
he says, these speculations must be set aside—​they are, “irrelevant” (akāraṇam)
and can only lead to “madness and distress.” The “ken of buddhas,” he avers, was
mentioned by the Buddha himself as included among those matters that are “in-
conceivable” and thus “maddening.”45 Dhammapāla sets aside the logical and
epistemological tangles by asserting that what it means that the Buddha was om-
niscient is that “whatever the Buddha wanted to know either partially or wholly,
he came to know it directly with no obstruction.”46 He had merely to attend to
something to perceive it directly and know it fully, and this could apply to all
past, present, and future phenomena.
For his part, I think Buddhaghosa is at his most subtle on the matter of omnis-
cience in his commentary on another place in the Paṭisambhidāmagga, where the
idea of “Buddha” is defined in terms of omniscience. The canonical text begins a
discussion of the term “Buddha” in this way:

He is “Buddha” because he is the Bhagavan, self-​arising, without a teacher,

awakened to the Truths by himself into things not heard before, and
therein, he has attained omniscience and achieved mastery in the powers.
By what meaning is Buddha “Awakened” (buddha)? “Awakened” be-
cause of having discovered the Truths; “Awakened” because of waking
beings up; “Awakened” because of knowing all things; “Awakened” be-
cause of seeing all things; “Awakened” because of not being able to be
guided by another, “Awakened” because of expanding.47

44. As quoted by Ñāṇamoli, trans., The Path of Purification, 772–​77, n. 7; see also Jaini, “On
the Sarvajñatva (Omniscience) of Mahāvīra and the Buddha,” 113–​15, and Endo, Buddha in
Theravada Buddhism, 72–​73.
45. Here he cites A ii.80: “Monks, the range of buddhas is inconceivable, not to be conceived,
and pondering it, one becomes mad or distressed” (bhikkhave, buddhavisayo acinteyyo, na
cintetabbo; cintento ummādassa vighātassa bhāgī assa).
46. ItiA i.141: yaṃkiñci bhagavatā ñātuṃ icchitaṃ sakalamekadeso vā, tattha appaṭihatavuttitāya
paccakkhato ñāṇaṃ pavattati.
47. Paṭis i.174; this is also in Mahāniddesa (ii.457) and the Cūlaniddesa; and Pj I.14: Buddhoti yo
so bhagavā sayambhū anācariyako pubbe ananussutesu dhammesu sāmaṃ saccāni abhisambujjhi,
tattha ca sabbaññutaṃ pāpuṇi, balesu ca vasībhāvaṃ. Buddhoti kenaṭṭhena buddho? Bujjhitā
saccānīti—​ buddho. Bodhetā pajāyāti—​ buddho. Sabbaññutāya buddho. Sabbadassāvitāya
buddho. Anaññaneyyatāya buddho. Vikasitāya buddho.
48 Build ing Blo cks for an In t er pretat ive Pro gr a m

The commentary on this suggests that the first sentence is the sense of “Buddha”
arrived at through meaning—​that these features of the Buddha, including om-
niscience, are denoted by the word “awakened.” The second part is an elabora-
tion arrived at through phrasing or form, where the sounds present in the word
“Buddha” connote additional or implicit meanings. (We explore the exeget-
ical use of the ideas of meaning [attha] and phrasing [byañjana] in the next
chapter.) I wish in particular to draw the reader’s notice to what happens when
Buddhaghosa glosses these lines:

Here, just as it is said in the world that someone has “understood” because of
“understanding,” this is the case too for one who is Awakened (“Buddha”),
who has discovered the Truths. Just as a leaf-​drying wind is called “leaf-​
drying,” so too “awakened” means waking beings up. Awakened because
of knowing all things: “awakened” is said because of a wisdom capable of
discovering all things. Awakened because of seeing all things: “awakened”
is said because of a wisdom capable of knowing all things. Awakened be-
cause of not being able to be guided by another: “awakened” is said because
of being awakened only by himself, not woken up by another. Awakened
because of expanding: “awakened” is said in the sense of expanding like a
lotus blossoming with many qualities.48

In ordinary speech in English too we often speak in the past participle when
we are referring to ongoing activities:  of students who are understanding we
can say that they have understood; a spouse going to the store has gone to the
store. So too, though “Buddha” is the past participle of the verb “awaken,”
this title also refers to an ongoing process as though it were the present par-
ticiple. Conceptually, this is highly significant: the Buddha’s awakening is still
occurring: he is Awakened and Awakening. Furthermore, indicated in the word
“Buddha” is a capacity for awakening others, for knowing all things, and for
seeing all things. His omniscience is a potential to know and see all things; it is
not that he achieved an encyclopedic knowledge of everything there is on the

48. Paṭis-​a. ii. 485: Ettha ca yathā loke avagantā avagatoti vuccati, evaṃ bujjhitā saccānīti buddho.
Yathā paṇṇasosā vātā paṇṇasusāti vuccanti, evaṃ bodhetā pajāyāti buddho. Sabbaññutāya
buddhoti sabbadhammabujjhanasamatthāya buddhiyā buddhoti vuttaṃ hoti. Sabbadassāvitāya
buddhoti sabbadhammabodhanasamatthāya buddhiyā buddhoti vuttaṃ hoti. Anaññaneyyatāya
buddhoti aññena abodhito sayameva buddhattā buddhoti vuttaṃ hoti. Vikasitāya buddhoti
nānāguṇavikasanato padumamiva vikasanaṭṭhena buddhoti vuttaṃ hoti. This exact passage is
also in Pj I.15, where it is a commentary on what “Buddha” means when taking refuge in the
Buddha, the Dhamma, and the Saṅgha. In the Paṭisambhidāmagga it is in a larger discussion
on meditations on the breath.
 Omniscience and Immeasurability 49

night of awakening; rather he has a capacity to know and see that allows him to
discover and teach all things in an ongoing process of awakening. The image of
the lotus blossoming from a tiny bud is quite apt here: the Buddha’s qualities
expand or unfold over time; we can say that his capacity to know and see all
things is ever unfolding.
This understanding of omniscience is vital for Buddhaghosa’s interpretation
of Buddha’s words as the unfolding of the Buddha’s unobstructed knowing and
seeing. The Pali commentators approached the Buddha’s omniscience not so much
through a formal definition of it in epistemological terms, but rather through
interpreting how the Buddha’s words express the unobstructed and expanding
nature of his knowledge. Implicit in Buddhaghosa’s theory of texts and evident
in his actual practices is the idea that interpretive practice is itself an exploration
of the case-​by-​case workings of the Buddha’s omniscience. The precise ways scrip-
ture expresses these workings is a matter for interpretative study of the different
genres, contexts, and purposes in which and to which he spoke. We shall see in the
chapters that follow that Buddhaghosa’s most sustained treatment of the Buddha’s
omniscience and the immeasurability of his words is in his introductory sections
to each of the three piṭakas, where he examines how the Buddha’s words reveal his
omniscient mind. First, the Brahmajāla Sutta (the first sutta of the first book of
the Suttanta and thus the occasion of Buddhaghosa’s introductory remarks), for
example, is interpreted by Buddhaghosa as revealing the extent of his knowledge of
particular beings. Second, the Atthasālinī, which opens with describing the nature
of the Abhidhamma in general, offers an expansive account of the Buddha’s “ocean”
of omniscient knowledge; in Buddhaghosa’s reading, Abhidhamma is, in my terms,
analysis all the way down. And, third, the Vinaya commentary sees the Buddha’s
unfolding knowledge of past, present, and future as requisite to interpreting this
genre. But first we must do further work on how Buddhaghosa generally under-
stood genres—​which for him mostly centered on the idea of piṭaka—​and the kinds
of study they make possible.

An “Ocean of Methods”:
Piṭakas as Practices of Learning
As I have already suggested, Buddhaghosa is explicit about the simultaneity of
the tipiṭaka conceived both as a quantifiable collection of texts and a means of
experiencing the Buddha’s omniscient wisdom. Here we may explore further ways
in which he develops this idea by considering the various ways he describes what
piṭakas are. First, he defines piṭaka as both “basket” of the sort one can use to carry
something, and as “an area of expertise” or “learning” (pariyatti) in which one can
50 Build ing Blo cks for an In t er pretat ive Pro gr a m

achieve competence.49 The three genres—​Suttanta, Vinaya, and Abhidhamma—​

are both at once:  they are baskets of various teachings and competencies in
learning that material. As baskets or containers, they can be described more or
less precisely as a certain number of collected texts. They “carry” a specific body
of material. But as the learning that scholars can take up, master, and deploy, the
piṭakas are not closed bodies of texts but rather practices, areas of specialization,
and methods of study: “pariyatti means mastery of the Buddha’s words.”50 When
piṭaka means learning it becomes an enactment and further instantiation of the
Buddha’s ken, suggesting a more “open” and dynamic idea of canon than scripture
as container. We can suggest some affinities here with the interpretation theory
of Paul Ricoeur, who argues that genres are “generative devices to produce dis-
course.” They can be used to classify texts by critics, but before that they are gen-
erative devices “to produce new entities of language longer than the sentence,
organic wholes irreducible to a mere addition of sentences.”51 Buddhaghosa, in
drawing out how piṭakas are not just compilations of texts but also areas of study,
shows how they generate disciplinary knowledge.
Having defined piṭaka as, in part, areas of study (pariyatti), Buddhaghosa
goes on to explain further what pariyatti entails. While we consider the distinct
methods and features of each genre in the next chapter, we can consider here
what he thinks the learning practices of the piṭakas generally look like. Each
piṭaka involves (1) a kind of training (sikkhā), (2) a kind of avoidance (pahāna),
and (3) four kinds of depth (gambhīra). (See Appendix C for the entire passage.)
Vinaya is training in higher moral precepts, Suttanta is training in higher aware-
ness, and Abhidhamma is training in higher understanding. Vinaya is avoiding and
opposing transgressions (of moral precepts), Suttanta is avoiding and opposing
being overpowered by thoughts (for concentration), and the Abhidhamma is
avoiding and opposing latent biases (for insight).52 Perhaps most interesting is

49. As 20; Sv i.18; Sp i.20: “People learned about the meaning of ‘piṭaka’ refer to piṭaka in the
sense of an area of expertise and basket” (Piṭakaṃ piṭakatthavidū pariyattibbhājanatthato āhu).
Buddhaghosa also refers to piṭakas as modes of instruction (pariyāya), that is, the distinctive
kinds of discourse we find in the jātaka tales, certain of the suttas, and the Abhidhamma (As
63; Sv iii.883).
50. Vism 442 (XIV.28): pariyatti nāma buddhavacanassa pariyāpuṇanam.
51. Ricoeur, Interpretation Theory: Discourse and the Surplus of Meaning, 32.
52. As 21–​22; Sp i.22; Sv i.19: Tathā hi—​vinayapiṭake visesena adhisīlasikkhā vuttā, suttantapiṭake
adhicittasikkhā, abhidhammapiṭake adhipaññāsikkhā. Vinayapiṭake ca vītikkamappahānaṃ,
kilesānaṃ vītikkamapaṭipakkhattā sīlassa. Suttantapiṭake pariyuṭṭhānappahānaṃ, pariyuṭṭh­
ānapaṭipakkhattā samādhissa. Abhidhammapiṭake anusayappahānaṃ, anusayapaṭipakkhattā
paññāya.
 Omniscience and Immeasurability 51

the fourfold depth that it is possible to acquire in each. For each piṭaka a learner
can become deeply immersed in the Dhamma, the meaning (attha), the teaching
(desanā), and the comprehension (paṭivedha). These are all deep because it is
difficult for persons of little intelligence to gain a footing in them “like a hare
(scrambling about) in the great ocean.”53
Buddhaghosa offers two different options for how each of these depths may
be understood, both of which are instructive in their particular ways. In the first
option, dhamma means the scripture (tanti), attha refers to the meaning of that
text, desanā is the teaching of the scripture that one has fixed in the mind, and
paṭivedha means full and correct understanding of the scripture and the meaning
of the scripture.54 In the second option, dhamma is taken to be cause (hetu), in
that “the analysis of a dhamma is knowledge of a cause”;55 to know what causes
a phenomenon is to know what that phenomenon is. The attha is the “effect” of
the cause: “the analysis of the meaning is the knowledge of the effect of a cause .”56
(It should not go unnoticed that these two forms of analysis are two of the four
kinds of analysis [paṭisambhidā] we considered earlier; here piṭaka as depth in
learning is defined as analysis in these two very specific senses.) Desanā in this
second option is verbally making the dhamma known according to its nature, or
telling it in the correct order or in the reverse order, in brief or extensively, and so
on. And paṭivedha, comprehension, refers here to realization in either a worldly
or transcendent sense.57
To sum up, in the first option, practices of depth involve mastery of a text
(reciting it), grasping its meaning, being able to teach it, and being able to under-
stand it. This option is centered on the Dhamma as text, and the ways the textual

53. As 22; Sp i.22; Sv i.19.

54. As 22; Sp i.22; Sv i.19: Tattha dhammoti tanti. Atthoti tassāyeva attho. Desanāti tassā manasā
vavatthāpitāya tantiyā desanā. Paṭivedhoti tantiyā tantiatthassa ca yathābhūtāvabodho.
55. As 22; Sp i.22; Sv i.19: hetumhi ñāṇaṃ dhammapaṭisambhidā ti. Buddhaghosa elsewhere
defines dhamma as scriptural learning, cause, quality, and something lacking life or essence (As
38:  dhammasaddo panāyaṃ pariyattihetuguṇanissattanijjīvatādīsu dissati). His two options
here deploy these first two definitions. The term dhamma is already potentially confusing, not
least in that it can mean of course both the teaching or doctrine (“the Dhamma”) or phenom-
enon or factor (in an Abhidhamma sense).
56. As 22; Sp i.22; Sv i.19: hetuphale ñāṇaṃ atthapaṭisambhidā ti.
57.  As 22; Sp i.22; Sv i.19:  Aparo nayo, dhammoti hetu. Vuttañhetaṃ—​‘‘hetumhi ñāṇaṃ
dhammapaṭisambhidā’’ti. Atthoti hetuphalaṃ, vuttañhetaṃ—​‘‘hetuphale ñāṇaṃ
atthapaṭisambhidā’’ti. Desanāti paññatti, yathā dhammaṃ dhammābhilāpoti adhippāyo.
Anulomapaṭilomasaṅkhepavitthārādivasena vā kathanaṃ. Paṭivedhoti abhisamayo, so ca
lokiyalokuttaro [ . . . ]. Lokiya, for Buddhaghosa, means worldly concerns in the round of re-
birth, while lokuttara refers to passing beyond saṃsāra (As 47, for example).
52 Build ing Blo cks for an In t er pretat ive Pro gr a m

knowledge of the piṭakas is studied and taught. In the second option, depth of
learning involves analysis of cause and effect, teaching these, and the transcendent
comprehension or realization that comes with that. This option lifts off from the
text to describe practices of knowledge and insight that involve the application
of textual knowledge to experience. These two options are not mutually exclusive
possibilities for what learning means, but rather, they are shifts of emphasis from
textual knowledge to its application to experience. All of these practices of anal-
ysis are highly generative, involving disciplined work to develop understanding
and knowledge.
Buddhaghosa also describes three kinds of expertise or learning:  that of a
person catching a snake, that of a person seeking escape (from saṃsāra), and that
of the treasurer.58 The person who catches a snake by the body or tail end and
gets bitten by it is like foolish people who get hold of the Dhamma and learn it
without examining it. And “having learned the Dhamma without investigating
its meaning through understanding the teachings, they cannot gain insight
into a meaning that is not investigated for they have only learned the teaching
for the purpose of their anger and to be free of criticism.”59 The Dhamma may
turn around and bite them back. He praises instead both the person who seeks
Dhamma for escaping saṃsāra, and the “treasurer” who, having attained release,
acquires the Dhamma in order to secure its transmission for future generations.
What is evident from the foregoing is that scriptural texts are not merely books
or recitations, but are generative practices of teaching and learning embedded
in a larger intellectual, disciplinary, and soteriological culture. The piṭakas are
conceived as moral and disciplinary practices (trainings and avoidances); they
involve depth of various sorts concerning the text, meaning, pedagogy, and com-
prehension, focusing on either the text or the application of the text to life. And
normative judgments may be made about how such learning is handled.
Perhaps the most significant description of the idea of the piṭakas is that they
are specific types of “methods” (naya). In the Atthasālinī, Buddhaghosa develops
an elaborate metaphor of the three piṭakas as “an ocean of methods.”60 (See
Appendix C for a translation of the whole section.) The ocean is for Buddhaghosa
a favorite image of vastness and immeasurability, ideas that are, as we have seen,

58.  As 23; Sp i.24; Sv i.21:  Tisso hi pariyattiyo—​

alagaddūpamā, nissaraṇatthā,
bhaṇḍāgārikapariyattīti.
59. As 23; Sp i.24; Sv i.21: te taṃ dhammaṃ pariyāpuṇitvā tesaṃ dhammānaṃ paññāya atthaṃ
na upaparikkhanti, tesaṃ te dhammā paññāya atthaṃ anupaparikkhataṃ na nijjhānaṃ
khamanti, te upārambhānisaṃsā ceva dhammaṃ pariyāpuṇanti itivādappamokkhānisaṃsā ca.
60. As 11: Katamo nayasāgaro? Tepiṭakaṃ buddhavacanaṃ.
 Omniscience and Immeasurability 53

closely linked to both buddhavacana and the Buddha’s omniscient ken.61 His
claim about the tipiṭaka as “an ocean of methods” is located in an extended meta-
phor that posits four oceans: the ocean of saṃsāra, the ocean of water, the ocean
of methods, and the ocean of knowledge.62 Births in saṃsāra are beginningless,
having no known starting place, and are in this sense oceanic. And the “waters
of the great ocean having a depth of 84,000 yojanas, cannot be measured with a
hundred, a thousand, a hundred thousand measuring cups” and so are, in effect,
“incalculable and immeasurable.”63 The “ocean of method”—​the tipiṭaka, the
buddhavacana—​is an ocean in that “endless joy and happiness arise for sons of
good families possessing faith, abounding in serene clarity, and whose knowledge
is excellent, who reflect on two of the scriptures.”64 And the “ocean of knowledge”
is of course the Buddha’s omniscience.
In each of these cases, the ocean metaphor works to convey the experience
of the one beholding it: the great ocean is in fact not infinite nor is it (at least in
theory) immeasurable; in fact, the ocean is thought to be 84,000 yojanas deep.
But as one stands before it holding a measuring cup, one finds it “incalculable
and immeasurable.” So too is an understanding of our experience in saṃsāra: as
we survey it, its beginning is entirely unknown to us. Similarly, the ocean of the
piṭaka methods when pondered by believing and knowledgeable students of
buddhavacana will evoke an experience of their endlessness. And the Buddha’s
omniscience was realized by him after sitting on the throne of awakening for
seven days, whereupon “he rose from it and stood gazing at the throne with un-
blinking eyes for [another] seven days, thinking, ‘on this throne I penetrated the
knowledge of omniscience.’ ”65
Buddhaghosa is chiefly interested here in the ocean of methods, since the point
of this discussion is to introduce the piṭakas, and specifically the Abhidhamma (as

61. The trope comparing a text or body of knowledge to an ocean for its vast endlessness is
found widely in Indic śāstric and literary traditions, as for example, the Nāṭyaśāstra VI.7, which
likens the bodies of knowledge about drama to oceans. Medieval texts are often called sāgaras,
oceans, such as Kathāsaritsāgara and the Dānasāgara. I have already mentioned the similarities
with this imagery for infinite textuality, outside of India, in Meister Eckhart and Ibn ‘Arabi.
62. As 10: saṃsārasāgaro, jalasāgaro, nayasāgaro, ñāṇasāgaroti.
63. As 11: So caturāsītiyojanasahassagambhīro. Tattha udakassa āḷhakasatehi vā āḷhakasahassehi
vā āḷhakasatasahassehi vā pamāṇaṃ nāma natthi. Atha kho asaṅkhyeyyo appameyyo
mahāudakakkhandhotveva saṅkhyaṃ gacchati.
64.  As 11:  dvepi hi tantiyo paccavekkhantānaṃ saddhāsampannānaṃ pasādabahulānaṃ
ñāṇuttarānaṃ kulaputtānaṃ anantaṃ pītisomanassaṃ uppajjati.
65. As 12–​13: Tato tamhā pallaṅkā vuṭṭhāya ‘imasmiṃ vata me pallaṅke sabbaññutaññāṇaṃ
paṭividdha’nti animisehi cakkhūhi sattāhaṃ pallaṅkaṃ olokento aṭṭhāsi.
54 Build ing Blo cks for an In t er pretat ive Pro gr a m

this passage is the introductory section of the Abhidhamma in the Atthasālinī).

We have just seen that the tipiṭaka is immeasurable because of the “endless joy
and happiness” it gives to those adept in two of its genres. But which two piṭakas
give rise to endless joy and happiness? The two scriptures are the Vinaya and the
Abhidhamma; unfortunately, Buddhaghosa nowhere explains why the Suttanta
methods do not do so. The Vinaya gives rise to such endless joy because “when
monks who are Vinaya experts are contemplating the Vinaya text, that is, the
declaring of the rules according to faults, they realize that it is the ken of buddhas
alone, not the scope of others [to know] ‘in this fault, in this transgression there is
the declaring of a rule.’ ”66 As in the other cases, the “oceanic” is experienced in the
beholder, in this case, the trained expert in this genre. And here, as we have seen
elsewhere with buddhavacana, the scripture is interpreted in relation to its author
and linked to his extraordinary ken. The Vinaya is an artifact of the Buddha’s
genius in that he could see in people’s wrongdoing the need to make rules to
govern the community in the future. Experts in this field can appreciate how apt,
appropriate, and durable this knowledge is.
Buddhaghosa is most interested here in how endless joy and happiness
occur when one comes to experience properly the methods of the Abhidhamma
Piṭaka. He says that monks trained in it experience endless joy and happiness
because they realize precisely how the Buddha arranged the phenomena of
form and formless experience into groupings as though gazing into the night
sky and seeing the constellations of the stars.67 He perceived that the vast
array of phenomena of experiential life can be grouped into the classificatory
schemes of the Abhidhamma lists, just as the vast array of stars in the heavens
can be grouped into constellations. He then tells a story that returns to the
ocean as a way to show how the Abhidhamma, especially the Paṭṭhāna, its
final book, somehow manages to convey the boundlessness of the Buddha’s
omniscience. To show how the method of the Abhidhamma gives rise to
endless joy and happiness, he recounts the story of a certain elder, Venerable
Mahāgatigamiyatissa, who traveled by boat to India (from Lanka) to worship
the Bodhi Tree at Bodh Gaya.

66.  As 11:  Vinayadharabhikkhūnañhi vinayatantiṃ paccavekkhantānaṃ dosānurūpaṃ

sikkhāpadapaññāpanaṃ nāma—​imasmiṃ dose imasmiṃ vītikkame idaṃ nāma hotīti
sikkhāpadapaññāpanaṃ—​aññesaṃ avisayo, buddhānameva visayoti.
67.  As 11:  Ābhidhammikabhikkhūnampi khandhantaraṃ āyatanantaraṃ dhātvantaraṃ
indriyantaraṃ balabojjhaṅgakammavipākantaraṃ rūpārūpaparicchedaṃ
saṇhasukhumadhammaṃ gaganatale tārakarūpāni gaṇhanto viya rūpārūpadhamme pabbaṃ
pabbaṃ koṭṭhāsaṃ koṭṭhāsaṃ katvā vibhajanto dassesi vata no satthāti abhidhammatantiṃ
paccavekkhantānaṃ anantaṃ pītisomanassaṃ uppajjati.
 Omniscience and Immeasurability 55

While traveling by ship he sat on the upper deck and, gazing at the great
ocean, realized that he could see neither the far shore nor the near shore.
He saw only the great ocean covered with foam from the fast-​breaking
waves like a silver cloth spread out and strewn with jasmine blossoms. He
began to wonder which is more powerful: the speeding waves of the great
ocean or the basis of the method in the entire Paṭṭhāna with its twenty-​
four divisions. He knew that the great ocean is bordered by the great earth
below, by the sky above, by the ring of mountains on one side, and by
shores on the other; yet a border to the entire Paṭṭhāna cannot be known.
And so the abstruse and subtle Dhamma appears as more powerful by
one so considering. Even while seated there he was overcome with joy,
increased his insight, destroyed the defilements, and attained the highest
fruition which is arhatship.68

Here we have a truly extraordinary reception of a text. We are asked to picture a

seafaring monk whose study of the ocean leads him to consider the Paṭṭhāna—​a
book somehow not contained by its covers—​and in the course of reaching for its
unreachable borders he attains sublime joy and release. The simile is quite pre-
cise: the waves rippling outward are like the beginning of the Paṭṭhāna method;
where they stop one cannot say. But the similitude breaks down and becomes a
contrast: the ocean is in fact not endless, as it does have borders, but not so the
methods of the Paṭṭḥāna. Still, we should notice that it is the analogy itself, how
the vastness of the ocean leads him to consider the vastness and endlessness of
the Abhidhamma, that prompts this monk’s great joy and then attainment of
awakening.
The Paṭṭhāna has special status for Buddhaghosa, because apparently, when
the Buddha spent a week contemplating it on the seat of Awakening, rays of light
of many colors emitted from the Buddha’s body, rays which pierced beyond the
earth’s atmosphere through the heavenly worlds and sprang out toward the infi-
nite world systems into outer space. The rays were so vast and endless that “even

68.  As 11–​12:  [mahāgatigamiyatissadattatthero kira nāma mahābodhiṃ vandissāmīti]

paratīraṃ gacchanto nāvāya uparitale nisinno mahāsamuddaṃ olokesi. Athassa tasmiṃ samaye
neva paratīraṃ paññāyittha, na orimatīraṃ, ūmivegappabhedasamuggatajalacuṇṇaparikiṇ
ṇo pana pasāritarajatapaṭṭasumanapupphasantharasadiso mahāsamuddova paññāyittha. So
kiṃ nu kho mahāsamuddassa ūmivego balavā udāhu catuvīsatippabhede samantapaṭṭhāne
nayamukhaṃ balavanti cintesi. Athassa mahāsamudde paricchedo paññāyati—​ayañhi heṭṭhā
mahāpathaviyā paricchinno, upari ākāsena, ekato cakkavāḷapabbatena, ekato velantena
paricchinno; samantapaṭṭhānassa pana paricchedo na paññāyatīti saṇhasukhumadhammaṃ
paccavekkhantassa balavapīti uppannā. So pītiṃ vikkhambhetvā vipassanaṃ vaḍḍhetvā
yathānisinnova sabbakilese khepetvā aggaphale arahatte patiṭṭhāya.
56 Build ing Blo cks for an In t er pretat ive Pro gr a m

the Great Brahmā, able to suffuse the thirty thousand thousandfold world systems
with light became like a firefly at sunrise.”69 And the Paṭṭhāna is, Buddhaghosa
asserts, the only text where the Buddha’s omniscience can begin to find room,
just like the great leviathan Timirapiṅgala can find room only in the great ocean
84,000 yojanas in depth.70 These spatial images—​the light extending in every di-
rection in the cosmos and the vast fish needing the (nearly) unfathomable depths
of the ocean to swim—​reveal the text, and the Buddha’s omniscience it somehow
conveys—​as never finally contained. We earlier considered the temporal aspects
of the same basic puzzle here explored spatially—​how can a week’s period of om-
niscient contemplation or three months’ worth of teaching the Abhidhamma
methods somehow convey the immeasurable?
To return to the extended metaphor of the oceans, we see that contemplation
of the textual method of the Abhidhamma thus leads inexorably to the fourth
ocean, the “ocean of knowledge,” which is the Buddha’s omniscience, and this
ocean can be known (fully) “only by the knowledge of omniscience.”71 From the
standpoint of his ken alone are the others fully knowable. From here Buddhaghosa
goes into several pages painting a fantastic picture of the events during the four
weeks of the Buddha’s awakening, including the miracles performed and the rays
of light emitted by his body that gave vivid expression to his grasp of the “abstruse
and subtle Dhamma.”72 We have already considered how this knowledge, “endless
and immeasurable” when “thought out in the mind” in the fourth week of awak-
ening, came to be eventually granted to Sāriputta in the “gift of the method”
whereby we come to have the Abhidhamma.
We thus see in a general way that it is through the methods of textual practice
that engagement with scripture becomes immeasurable (the details of how they
do this remain to be worked out). The idea of method (naya) is fundamental
to the entire Pali commentarial project, and the term had wider resonance

69.  As 14–​15:  Tisahassimahāsahassilokadhātuyā ālokapharaṇasamattho mahābrahmāpi

sūriyuggamane khajjopanako viya ahosi.
70.  As 13:  Yathā hi timirapiṅgalamahāmaccho caturāsītiyojanasahassagambhīre
mahāsamuddeyeva okāsaṃ labhati, evameva sabbaññutaññāṇaṃ ekantato mahāpakaraṇeyeva
okāsaṃ labhi. Timirapiṅgala is a huge fish, a thousand leagues long, who roams the deepest
parts of the sea (Jātaka v.462, no. 537).
71.  As 12:  “The ocean of knowledge is called the knowledge of omniscience which can be
known only by the knowledge of omniscience and it cannot by known otherwise” (aññena na
sakkā jānituṃ, sabbaññutaññāṇeneva sakkā jānitunti sabbaññutaññāṇaṃ ñāṇasāgaro nāma).
72. As 15: saṇhasukhumadhammaṃ.
 Omniscience and Immeasurability 57

within the Indic tradition, particularly Jainism, as a schema for understanding.73

Its verbal root, nayati, to lead, is at the heart of the conception of both scrip-
ture and commentary. In scripture, which has been our concern here, naya are
teaching methods that the Buddha deployed, and which were, as Buddhaghosa
often asserts, many and various. The Buddha was said to have taught (or led
people to understand) by the use of various methods, especially in the case of
the Abhidhamma: “the Perfectly Awakened One never failed to make use of a
method on the right occasion when it came to the Abhidhamma.”74 He also taught
methods that people could in turn use to expand and grow their knowledge, as we
have seen on several occasions already. Naya can also mean method of interpre-
tation, and comes to mean very specific interpretative methods or guidelines in
the hermeneutical texts, the Nettippakaraṇa (“The Guide,” where the word netti,
guide, is from the same verbal root) and the Peṭakopadesa (“Instruction in the
Piṭaka”).75 (We consider in the next chapter the influence of these two guides on
the aṭṭhakathā literature). For now, what we have learned is that the piṭakas were
conceived as methods of teaching that could give students access to the immeas-
urable reach of the Dhamma.

Conclusions
We have seen that Buddhaghosa interpreted scriptural texts as the product of a
unique ken, the omniscient ken of buddhas. Omniscience, however, is best seen in
practice, as an exercise, rather than as a collection of stock knowledge such as an
encyclopedia holding all the facts of the world. Instead, what it is for the Buddha
to exercise his omniscience is to advert his attention to know a thing “unob-
structed” by any ignorance that could impede understanding. The Dhamma itself
is the infinite workings of experience, and the ability to follow these workings, all

73. Ram-​Prasad glosses naya as a “schema,” or a “ ‘lead,’ a way of approaching an infinite mul-

tiple reality,” citing the medieval Jain philosopher Malliṣeṇa that “as many as are the ways of ex-
pression, just so many are there schema” (Ram-​Prasad, Indian Philosophy and the Consequences
of Knowledge, 33 n77). While for Jains naya are “viewpoints” or epistemological commitments
(rather than methods or practices as they are for Buddhaghosa), they are important in
their multiplicity; along with syādvāda and anekāntavāda, nayavāda is a principle of “Jaina
multiplism,” the “claim that no single schema can present a total or determinate description
of reality” (31).
74. As 339: sammāsambuddhena hi abhidhammaṃ patvā nayaṃ kātuṃ yuttaṭṭhāne nayo akato
nāma natthi.
75. For systematic discussion of the technical sense of naya in the two hermeneutical texts, see
Ñāṇamoli, The Guide, xli–​xliii, and Bond, The Word of the Buddha, 49–​77.
58 Build ing Blo cks for an In t er pretat ive Pro gr a m

the way through, is precisely what is grasped by the mind of all buddhas in their
attainment of supreme awakening.
In Buddhaghosa we get the interesting (and overlooked) move of considering
what the entailments of the Buddha’s omniscience means for being a Buddhist.
He sees Buddhist practice as tasked not with coming up with a formal definition
of the Buddha’s omniscience, but rather as exploring how Buddhist thinkers live
in relationship to it, by practicing and enacting its methods. Texts and the in-
terpretative practice engaging them permit one, in a limited way, to engage the
workings of this extraordinary omniscient activity that is the Dhamma and the
Buddha at once.
Scripture in this picture must be something quite specific and directly
authenticated as the words of the Buddha, for it is only through what the
Buddha taught that we (as unawakened persons) might glimpse the immeas-
urable workings of this mind and the Dhamma that it knows. Thus the First
Council, in its codification of a “canon,” signals a complete and knowable body
of buddhavacana, contained in books and competent masters. But as important
as it might be to see this collected material as complete or closed in these ways,
it is also crucial that the books or the recitations of scripture cannot fully con-
tain or exhaust the immeasurability of the teachings. We thus find a number of
practices named as the generative exercises that describe the work the piṭakas do
in the world and the practices that they entail for students of them. Some of the
main ones I have identified here are learning (pariyatti), analysis (paṭisambhidā),
and methods (naya), though tracing out the distinctive modes of knowledge that
specific discourses and genres make possible is the task of the following chapters.
The encounter with texts that Buddhaghosa is imagining and constructing is also
shaped by richly metaphorical conceptions of oceanic depth, cosmic spectacle,
and above all, a fabulous and astonishing Buddha.
To sum up, we may draw back and point to the specific ways that Buddhaghosa’s
enterprise shares some of the features of scholasticism that José Cabezón has
identified in his useful development of this term as a comparative category.76
Cabezón’s work helps us to see certain patterns and features of commentarial dis-
course and how they are related to one another. The traits of scholasticism include
notions of the “completeness and compactness” of the scriptural canon together

76. See Cabezón, “Introduction” of Scholasticism: Cross-​Cultural and Comparative Perspectives,

4–​6, for the key terms discussed in the next two paragraphs. Cabezón also describes “reasoned
argument and noncontradiction” as important features of scholasticisms; this chapter, focusing
more on the literary and evocative qualities of Buddhaghosa’s project, has not demonstrated
that these are features of Buddhaghosa’s systematic thinking, but later chapters do suggest that,
at least in the trenches of his commentarial work, he values the careful building of arguments
and coping with (real or apparent) contradictions.
 Omniscience and Immeasurability 59

with claims and practices of its proliferativity. As we have seen, Buddhaghosa

works to establish an identifiable body of revered tradition through the idea of
the First Council (a contrivance that embodies the strong sense of tradition that
Cabezón also finds in scholasticisms). The First Council is said to have produced
a corpus of texts that is complete (nothing is missing)77 and compact (in the sense
that nothing is extraneous). Yet this body is also generative and expansive, and its
meanings and possibilities can endlessly proliferate. Additionally, Buddhaghosa’s
reflections about texts explored in this chapter identify and analyze his first-​order
practices, which constitute the second-​order processes of the sort that Cabezón
refers to as the self-​reflexivity of scholasticism.
Moreover, the ideas about Buddha’s omniscience and how they are intertwined
with features of scripture and textual practice might be helpfully illuminated by
Cabezón’s scholastic feature of “the epistemological accessibility of the world.” This
claims that the universe is basically intelligible and that, at least in principle, every
phenomenon can be known. Seen in this light, the claims about omniscience are
a way that the Pali scholastics give expression to the idea that everything is know-
able, at least by the Buddha. As Cabezón points out, this feature is related to
both the proliferative and complete aspects of scripture: the basic knowableness
of the world tells us something about what scripture is. It becomes the enacted
exploration of the world. The Buddha’s infinite knowledge is configured to be
both complete, while also endlessly generative, with both expressed through
buddhavacana.
A further feature described by Cabezón is scholasticism’s concern with sacred
language, and its confidence in “the communicative ability” of both scriptural
word and commentarial exposition; these also betray the scholastic commitment
to “conceptual thought and categories.” Our exploration of the well-​spoken
words of the Buddha shows the reverence Buddhaghosa places in the special kind
of language that buddhavacana is for its capacity to generate meaning and expe-
rience (as well as its capacity to generate commentary). The next chapter explores
several concrete features of the unique speech of the Buddha that communicate
the very skills required to understand it.

77. Buddhaghosa says explicitly that the dhamma taught by the Buddha was complete (kevala),
entire (sakala) (Sv i.177; Ps ii.20).
 2

Scripture, Commentary,
and Exegetical Distinctions
Who is there able to understand in every way the words of
the Bhagavan, skilled in many methods, originating for var-
ious kinds of dispositions, perfect in meaning and phrasing,
possessing various marvels, deep in penetrating the teaching
of the meaning of the Dhamma, suitable for all beings each in
their own language as it enters their own ears?1

This chapter explains various hermeneutical distinctions, stemming

from qualities purported to be inherent in the Buddha’s words, that Buddhaghosa
described and invoked in his commentarial practice. Where the last chapter set
out in general terms Buddhaghosa’s theory of buddhavacana, here we look at the
details of method and practice. As the commentators interpreted buddhavacana,
they took the features of the Buddha’s speech as instructions on how to interpret
it. Most of these features are well known from other studies of Buddhist ideas
about language and discourse, but the particular ways that the Pali commentarial
tradition interpreted and used them may not be as widely understood. Embedded
in them are ideas about what texts are and how they work.
First, we continue our study of what Buddhaghosa had to say about each genre
or piṭaka and how he understood commentary (aṭṭhakathā). We then turn to the
various registers of language and styles of discourse he defined and employed.
These involve a series of contrasts and distinctions about scriptural language, in-
cluding distinctions of meaning (attha) and phrasing (byañjana), speaking briefly
(saṅkittena) or with detail (vitthārena), uses of language for furthest meaning
(paramattha) and for conventional (sammuti) purposes, definitive (nītattha)

1.  Sv 1.27; Ps i.3–​4; Spk i.5; Mp i.5; Pj I, 101:  nānānayanipuṇaṃ anekajjhāsayasamuṭṭhānaṃ

atthabyañjanasampannaṃ vividhapāṭihāriyaṃ dhammatthadesanāpaṭivedhagambhīraṃ
sabbasattānaṃ sakasakabhāsānurūpato sotapathamāgacchantaṃ tassa bhagavato vacanaṃ
sabbappakārena ko samattho viññātuṃ.
 Scripture, Commentary, and Exegetical Distinctions 61

and interpretable (neyyattha) speech, and contextual (pariyāya) and categorical

(nippariyāya) teachings. These function for Buddhaghosa both as sites for theo-
retical reflection about buddhavacana and how it works, and as useful building
blocks of the commentarial system, allowing a commentator to draw out or de-
cide meaning on the basis of what kind of discourse or teaching a passage is taken
to be. They are, as I will show, best understood as pragmatic methods for peda-
gogy (on the Buddha’s part) and thus programmatic methods for interpretation
(on the commentator’s part).

Three Genres of Buddha’s Pedagogy:

Vinaya, Sutta, and Abhidhamma
We have already encountered certain distinctions drawn about what the three
piṭakas are (they are “methods” and “areas of expertise” which involve “trainings,”
“avoidances,” and “depths”), but there are further points of definition that
Buddhaghosa mentions in the three introductions to his commentaries on each
genre, where he reflects about the distinctiveness of each genre as a whole. In his
treatment of what each piṭaka does, he develops the ways they involve methods
of expansion of the sort that we saw indicated in the last chapter; see Appendix
C for a translation of this discussion. He begins with Vinaya, deploying a nirutti
(linguistic or phonological) definition of this genre that makes much of the
sounds in the word “vinaya”: “the Vinaya is called ‘vinaya’ by those wise in the
meaning of vinaya because it contains various and distinctive methods and be-
cause it disciplines body and speech.”2 He explains that the Vinaya is itself com-
posed of various kinds of material (the Pātimokkha recitation, rules involving the
seven offenses starting with defeats, matrices, and exegesis [vibhaṅga], et cetera),
and that it is distinctive due to supplementary rules that can soften rigid rules.3
We have already seen that the Vinaya is “higher moral precepts” (adhisīla), and
here we see reiterated that it is scripture that disciplines the body and speech. The
suggestion that it contains supplementary rules to soften its more rigid stances
signals a dynamic and flexible quality in the text.

2. As 19; Sp i.18; Sv i.17: Vividhavisesanayattā, vinayanato ceva kāyavācānaṃ; vinayatthavidūhi

ayaṃ, vinayo vinayoti akkhāto.
3.  As 19; Sp i.18; Sv i.17:  “Various here refers to the methods classified into the five-​fold
Pātimokkha Uddesa, the matrix of the sections on the seven kinds of offenses beginning with
defeats, and the analysis, et cetera. These have become distinctive because the methods of sup-
plementary regulations aim to make flexible what is rigid” (Vividhā hi ettha pañcavidhapāti
mokkhuddesapārājikādi satta āpattikkhandhamātikā vibhaṅgādippabhedā nayā. Visesabhūtā ca
daḷhīkammasithilakaraṇappayojanā anupaññattinayā).
62 Build ing Blo cks for an In t er pretat ive Pro gr a m

Buddhaghosa next describes the Suttanta:

It indicates meaning classified into meaning for oneself and meaning for
others, et cetera. It is well-​spoken in that the meanings are spoken in ac-
cordance with the inclinations of those being taught right here. It flows
is said [to indicate] that it bears fruit in meaning, like crops. It yields is
said [to indicate] that it discharges as a cow does milk. . . . And there is
the sharing in common of a sutta. And just as a thread is a measure for
carpenters, in this way [a thread/​sutta] is for the wise. And just as flowers
strung by a thread do not scatter and disperse, so too by it, meanings are
gathered.4

This names as an important quality of the suttas that they speak to the people
in them according to their disposition and needs and that suttas speak directly
to self and to others. The idea that a sutta contains meaning for both self and
other is, as we explore in ­chapter 3, important in that one feature that interested
Buddhaghosa about Suttanta is the way a sutta speaks simultaneously to its orig-
inal audience and to subsequent readers quite distant from it. The passage also
indicates the fecundity of Suttanta knowledge in that it grows meaning like crops
and nourishes people like a cow her calf. The carpenter’s measuring line is also
notable: it measures or probes the depth of something.
The word “Abhidhamma” is parsed according to the possibilities for its prefix,
abhi, attached to dhamma, which here refers to the phenomena described in this
genre. Here “the word ‘abhi’ is shown to refer to how [the dhammas] can be grown,
characterized, honored, defined, and [seen as] surpassing.”5 Dhammas describe
experiences like loving-​kindness (according to his example) that, when cultivated,
grow and expand one’s awareness; he cites a passage from the Dhammasaṅgaṇī,
the first book of the Abhidhamma, where loving-​kindness practices cause one’s
awareness to “pervade” the directions.6 That abhi can also have the sense of

4.  As 19; Sp i.19; Sv i.17:  Tañhi attatthaparatthādibhede atthe sūceti. Suvuttā cettha atthā
veneyyajjhāsayānulomena vuttattā. Savati cetaṃ atthe, sassamiva phalaṃ, pasavatīti vuttaṃ
hoti. Sūdati cetaṃ, dhenu viya khīraṃ, paggharatīti vuttaṃ hoti. . . . Suttasabhāgañcetaṃ. Yathā
hi tacchakānaṃ suttaṃ pamāṇaṃ hoti evametampi viññūnaṃ. Yathā ca suttena saṅgahitāni
pupphāni na vikiriyanti na viddhaṃsiyanti evametena saṅgahitā atthā. Like most of these
definitions of genres, meaning is developed through niruttis, here deploying words starting
with sounds similar to “su” in sutta: sūceti, suvuttā, savati, sūdati. See Appendix C for the whole
passage.
5. As 19; Sp i.20; Sv i.17: Ayañhi abhisaddo vuḍḍhilakkhaṇapūjitaparicchinnādhikesu dissati.
6. As 19; Sp i.20; Sv i.17.
 Scripture, Commentary, and Exegetical Distinctions 63

“surpassing” is evidenced in passages that refer to something having “extraordi-

nary beauty.”7 Elsewhere, Buddhaghosa understands “Abhidhamma” to refer to a
more expansive method of treating terms and ideas than the way they were some-
times presented in the Suttanta. Even very fundamental doctrinal topics like the
five aggregates are “detailed only with regard to one instance in the Suttanta, not
unrestrictedly” (as they are in the Abhidhamma);8 where the Suttanta uses more
limited methods, the Abhidhamma “details Suttanta classifications (bhājanīya),
Abhidhamma classifications, and the method of listing questions unrestricted
to [any single] instance.”9 Thus the Abhidhamma “is distinguished from and
surpasses the Dhamma” as it is stated in the Suttanta.10
Buddhaghosa also contrasts the three genres in the following terms:11
Vinaya is a teaching that uses commands, while Suttanta teaches in colloquial
language, and Abhidhamma in language that analyzes in the “furthest sense.”
These distinctions concern different kinds of language:  injunctive, colloquial,
and analytical distinctions important to the exegetical enterprise at partic-
ular junctures. Additionally, Buddhaghosa says, Vinaya is the Dispensation
according to offenses (in the monastic code), Suttanta follows the ways of the
world according to the “various inclinations, biases, practices, and dispositions”12
of beings, and Abhidhamma teaches according to phenomena (dhammas) to
beings who perceive “I” or “mine” in the heap of phenomenal experience.13 These
distinctions suggest something of the content of each of the three collections,
but are also cues for interpretative practice; for example, as we will see in the next
chapter, interpreting suttas according to the “inclinations, biases, practices, and
dispositions” of people described in the narrative of the sutta is taken to be an

7. As 19; Sp i.20; Sv i.17: “[Dhammas are] surpassing in the sense of such passages as ‘because of
extraordinary beauty’ ” (Abhikkantena vaṇṇenā’’tiādīsu adhike).
8. As 2: Suttantañhi patvā pañcakkhandhā ekadeseneva vibhattā, na nippadesena.
9. As 2: suttantabhājanīyaabhidhammabhājanīyapañhapucchakanayānaṃ vasena nippadesato
vibhattā.
10. As 2: Evaṃ dhammātirekadhammavisesaṭṭhena abhidhammoti veditabbo.
11. This paragraph discusses this set of distinctions and how Buddhaghosa glosses them: etāni
hi tīṇi piṭakāni yathākkamaṃ āṇā vohāra paramatthadesanā yathāparādha-​yathānuloma-​
yathādhammasāsanāni, saṃvarāsaṃvaradiṭṭhiviniveṭhanāmarūpaparicchedakathāti ca
vuccanti (As 21; Sp i.21; Sv i.19).
12. As 21; Sp i.21; Sv i.19: anekajjhāsayānusayacariyādhimuttikā sattā.
13.  As 21; Sp i.21; Sv i.19: dhammapuñjamatte ‘ahaṃ mamā’ti saññino sattā yathādhammaṃ
ettha sāsitāti yathādhammasāsananti vuccati.
64 Build ing Blo cks for an In t er pretat ive Pro gr a m

explicit exegetical principle.14 Finally, Vinaya is a teaching about the kinds of res­
traint (concerning monastic precepts), Suttanta is unraveling views (in treating
the sixty-​two views that open the first sutta in the collection), and Abhidhamma
is about differentiating the processes of name and form (nāmarūpa).15 These
distinctions indicate different kinds of content and purposes for each of the
piṭakas, and will be useful for recognizing the methods Buddhaghosa deploys to
interpret particular passages.

Conceptions of Commentary
and Vibhajjavāda Analysis
Buddhaghosa sees commentary as a method of analysis that is generative and
expansive. As we have seen and as other scholars have noted, commentarial ex-
egesis began in the canon, sometimes by the Buddha himself. Even in the suttas
we have “analytical exposition” (vibhaṅga) that shows the Buddha providing a
detailed drawing out of analytical meaning from shorter summary texts (such
as the Mahākammavibhaṅga Sutta).16 Certain disciples were known for their
capacities in exposition, such as Sāriputta (who is, as we have seen, associated
with Abhidhamma analysis), Mahākaccana (renowned for expatiating in de-
tail what the Buddha said in brief ), and Mahākoṭṭḥita (second to the Buddha
in analysis, paṭisambhidā).17 The Saṅgīti Sutta contains both a catechism list
of enumerated topics and a detailed exegesis of those topics as expounded by
Sāriputta. The Abhidhamma and the Vinaya also have vibhaṅga books, though
the kind of material that constitutes analytical drawing out varies significantly
between them. In the Vinaya, vibhaṅga refers to narrative commentarial exegesis
in which the monastic rules are embedded (in its Suttavibhaṅga). The second
book of the Abhidhamma is called Vibhaṅga, which is a quite formal terminolog-
ical exposition rather than narrative. Another category of commentary is niddesa,

14. As discussed in Netti 24–​25, for example, where the Buddha is said to teach people according
to their temperaments, and his teachings are interpretable in accordance to the audience or the
people whom the teaching is describing (see Bond, In the Buddha’s Words, 82, 85).
15. See fn 11: saṃvarāsaṃvaradiṭṭhiviniveṭhanāmarūpaparicchedakathāti ca vuccanti (As 21; Sp
i.21; Sv i.19).
16.  The “Great Exposition on Kamma” (M iii.207–​15) can be contrasted with the “Short
Exposition on Kamma” immediately preceding it in the Majjhima. Law argues that the first is
“the Sutta basis of Abhidhamma exposition of the Sikkhāpadavibhaṅga” in the second book of
the Abhidhamma (Law, Buddhaghosa, 57). Anālayo, The Dawn of Abhidharma, 79–​86, has a
useful account of the various forms of canonical commentary.
17. Law, Buddhaghosa, 57–​58. See also Bond, The Word of the Buddha, 101–​2.
 Scripture, Commentary, and Exegetical Distinctions 65

and the Mahāniddesa, included in the Khuddaka collection, is itself an exegetical

commentary (on words and phrases of the Suttanipāta); a niddesa in general can
also be the expanded exposition of an outline (uddesa).
The term aṭṭhakathā, literally, “explanation of the meaning,” appears
in the canon only in reference to material in the Abhidhamma (a chapter
of the Dhammasaṅgaṇī), and thus enters the tradition only at this level of
development; it is also a distinctively Pali genre. In general, Buddhaghosa
identifies the aṭṭhakathā as the “doctrines of the Teachers” (ācariyavāda),
and as something paired with or contrasted to the canonical text (pāḷi).18
As mentioned earlier, in his colophon to his commentaries on each nikāya,
Buddhaghosa claims that the aṭṭhakathā (also called the Mahāṭṭhakathā)
was recited at the First Council; he does not go so far as to say that this
material is buddhavacana, but rather suggests that it is an adjacent body of
exposition accompanying the Buddha’s word and preserved by elders from
very early on. Buddhaghosa also describes aṭṭhakathā as “questioning”
(paripucchā) in a context where he is explaining that one can become adept
in analysis (paṭisambhidā) by being learned in scripture and being questioned
by others about it.19 The aṭṭhakathā is also a distinct scholastic method
(aṭṭhakathānaya), distinguished from scriptural methods (pāḷinaya), and the
views of particular teachers (ācariyanaya).20
Another term for commentary is vaṇṇanā. George Bond has shown that the
commentaries use a distinction between “word commentary” (padavaṇṇanā)
and “meaning commentary” (atthavaṇṇanā). The “word commentary” glosses
words in a text according to their lexical or conventional meaning, while the
“meaning commentary” defines the religious significance of the ideas in view
of the whole Dhamma, what he calls “dhamma-​contextual” meaning, that is,
“the specialized meaning of the word in the context of the dhamma.”21 In the
example Bond cites in the Paramatthajotikā, Buddhaghosa comments on what the
scriptural line “living in a suitable place” means. Its “word commentary” defines

18. Sv ii.567, for example. Vism 107: idaṃ . . . n’eva pāḷiyaṃ na aṭṭhakathāyaṃ āgataṃ kevalaṃ
ācariyamatānusārena vuttaṃ.
19.  Vibh-​a 388–​89:  ‘paripucchā’ nāma aṭṭhakathā. Uggahitapāḷiyā atthaṃ kathentassa hi
paṭisambhidā visadā honti.
20.  Puggalapaññati commentary called the Pañcappakaraṇaṭṭhakathā, 172, and discussed in
Law, Buddhaghosa, 92–​94.
21. Bond, The Word of the Buddha, 150. An instance of Buddhaghosa saying explicitly what a
commentary on meaning involves occurs at Ps ii.203, where Buddhaghosa says that the Buddha
“teaches with meaning” not in matters of glossing terms like “rice,” “meal,” “woman,” or “man,”
but rather when teaching on subjects like the four foundations of mindfulness.
66 Build ing Blo cks for an In t er pretat ive Pro gr a m

place according to ordinary word definitions (“place” means village, town, city, or
country, for example). Its “meaning commentary” defines this line in terms of its
connection to the Buddhist Dhamma (“place” means where the fourfold com-
munity [of monks, nuns, laymen, and laywomen] are present, where giving can
be meritorious, where the sāsana is present, et cetera).22 This twofold hermeneu-
tical device is situated within a larger approach to meaning and definition, Bond
argues persuasively (sharing the view of Bhikkhu Ñāṇamoli), that resists essen-
tialist definitions in favor of a highly contextual approach to defining terms and
within a general aim of stating their significance in relationship to the Buddhist
teaching as a whole.23
Buddhaghosa refers also to the commentaries and the Visuddhimagga as
“expositions” (vinicchaya).24 “Vinicchaya” means a type of exposition based on
a certain style of analytical questioning. In the Visuddhimagga this often takes
the form of a “set of questions” (pañhākamma), which lists questions that struc-
ture exegesis. As his commentary answers each question it expands the meaning
(much as one expands meaning from a mātikā). Bond has shown that word com-
mentary is distinguished from the questioning and analytic commentarial prac-
tice of vinicchaya.
Buddhaghosa tells us in the Samantapāsādikā that his editorial hand is
conservative—​he is merely transmitting commentaries to teach the Vinaya in
keeping with the methods of the “ancient teachers,” which are the methods of
the perfectly awakened buddhas.25 But he also acknowledges that he “corrected
scribal errors.”26 Jayawickrama has suggested that this might be a generous way
of attributing any “mistaken” interpretations of his predecessors to clerical

22. Pj I 132; Bond, The Word of the Buddha, 146–​52. Another example is As 124.
23.  Bond, The Word of the Buddha, 148–​52; Ñāṇamoli, “Translators Introduction” to The
Guide, vii–​x.
24. Sp i.1 and the Visuddhimagga nigamana.
25. Sp i.1: He expounds the Vinaya, “placing reliance on what conforms to the ancient teachers.
The Vinaya is expounded with pleasure with the methods [practiced by] the mind that is the
excellent lineage of Perfectly Awakened buddhas, (according to) the assembly of teachers, who
are pure of the taints washed away from the waters of knowledge, whose analyses of knowl­
edge are purified, who are skilled in the commentary on the Good Dhamma, who have no
equal in their attainment of ascetic practice, and who are like banners of the Mahāvihāra”
(nissāya pubbācariyānubhāvaṃ. Kāmañca pubbācariyāsabhehi; Ñāṇambuniddhotamalāsavehi;
Visuddhavijjāpaṭisambhidehi; Saddhammasaṃvaṇṇanakovidehi; Sallekhiye nosulabhūpamehi;
Mahāvihārassa dhajūpamehi; Saṃvaṇṇitoyaṃ vinayo nayehi; Cittehi sambuddhavaranvayehi).
26. Sp i.2: vajjayitvāna pamādalekhaṃ
 Scripture, Commentary, and Exegetical Distinctions 67

mistakes.27 Buddhaghosa says that in addition to translating the Sinhala texts into
Pali he condensed certain passages, while not leaving out anything in keeping
with scripture.28 He identifies the commentaries he is preserving as adhering to
the sectarian affiliation “Mahāvihāra” (the “Great Monastery”) preserved in Sri
Lanka, and his interpretations in the Visuddhimagga are  in keeping with the
“Vibhajjavādins” (the “Analysts”).29
Buddhaghosa’s identification with the Vibhajjavādins is instructive for dis-
cerning the intellectual and scholarly values he affirms. In commenting on a finer
point of interpreting dependent origination, he describes the following series of
exegetical practices. This passage is one of the very few places he describes general
criteria for commentarial practice in expanding meaning.

The commentary on the meaning of this should be done by one who stays
within the circle of the Vibhajjavādins, does not misrepresent the teachers,
does not launch into his own view, does not quarrel with the views of
others, does not deviate from the Sutta, who stays in accordance with
the Vinaya, considers the great authorities (mahāpadesa), illuminates the
Dhamma, takes up the meaning and then returns again to that meaning by
explaining it with different methods (pariyāya).30

This is not a scholarly style championing one’s own positions or engaging in philo-
sophical disputation; it is instead cast as ever faithful to scripture and the teachers
in preference to one’s own view. It respects the “great authorities” (mahāpadesa),
which refer to the Buddha, a community of distinguished elders, a group of
learned elders, and a single learned elder, insofar as the latter three conform to the
Suttanta and the Vinaya. (This refers to a passage on scriptural authority in the
Dīgha.31) What I think is most illuminating in this passage is not Buddhaghosa’s
stated traditionalist approach to texts, but rather the explicit naming of the

27. Jayawickrama, The Inception of the Discipline, 96 n.11. See Endo, Studies in Pāli Commentarial
Literature, ch. 4, for a discussion of, and review of scholarship on, Buddhaghosa’s editorial
principles.
28. Sp i.2: tato ca bhāsantarameva hitvā;vitthāramaggañca samāsayitvā;vinicchayaṃ sabbamase
sayitvā;tantikkamaṃ kiñci avokkamitva.
29. Sp i.2; Vism 522.
30.  Vism 522 and Vibh-​a 130:  tassā atthasaṃvaṇṇanaṃ karontena vibhajjavādimaṇḍalaṃ
otaritvā ācariye anabbhācikkhantena sakasamayaṃ avokkamantena parasamayaṃ anāyūhantena
suttaṃ appaṭibāhantena vinayaṃ anulomentena mahāpadese olokentena dhammaṃ dīpentena
atthaṃ saṅgāhentena tamevatthaṃ punarāvattetvā aparehipi pariyāyantarehi niddisantena.
31. D i.123–​27.
68 Build ing Blo cks for an In t er pretat ive Pro gr a m

practice of identifying meaning and then finding numerous methods to explain

it. Commentarial practice was for the Analysts just this very method of devel-
oping numerous classification schemes to analyze in brief and in detail, some-
times apparently endlessly, possibilities for meaning. Analysis (vibhajja) is a style
of pedagogy and interpretation that seeks out methods and ways of explaining
and illuminating what the commentaries took to be immeasurable. It also has
canonical precedent where it is recognized as a style of discourse that is about
making distinctions.32 It concerns a way of answering questions that proceeds by
distinguishing things; in fact, even the Buddha in one place identifies himself as a
“Vibhajjavādin,” as he does not speak one-​sidedly, and avoids general declarations
when answering questions.33
Modern scholars are divided on whether in referring to themselves by this
term the monks of the Mahāvihāra lineage were simply indicating a proper name
for their sectarian identity, or whether they were suggesting something distinctive
about their method as “those who advocate analysis,” or both. Rupert Gethin sees
it as a sectarian identifier.34 C. S. Prasad, on the other hand, thinks that something
of the Vibhajjavādins’ “indifference to metaphysical speculations,” as against the
more metaphysical leanings of the Sarvāstivādins, is implicit in their choosing
this term to identify themselves.35 The passage quoted earlier is the only instance
where the term is treated with any substance in Buddhaghosa’s work, so I am not
sure just how he meant it. But what we might say is that if he means it here as sec-
tarian identifier then implicit in this passage is the idea that other Buddhists do
not proceed in this way perhaps in not following what he considers to be the tra-
dition or by asserting their own views. In any case, what is clear is the valorization
of “Analysts,” as those who use multiple methods to work out meaning.
Further, analysis in the sense of making distinctions and providing classifi-
catory schemes is also foundational to the contemplative practices described by
Buddhaghosa. George Bond has discerned the importance of the aṭṭhakathā’s

32.  The Aṅguttara describes different ways of answering questions (categorically, by making
distinctions, by asking a counterquestion, or by setting aside the question). The method of
making distinctions is the “vibhajja” approach:  vibhajjabyākaraṇīyaṃ pañhaṃ vibhajja
byākaroti (A i.197).
33. M ii.197, the Subha Sutta.
34.  Gethin, “Was Buddhaghosa a Theravādin? Buddhist Identity in the Pali Commentaries
and Chronicles,” 15.
35.  Prasad, “Theravāda and Vibhajjavāda:  A Critical Study of Two Appellations,” 112. On
Vibhajjavāda as a sectarian name, see also Bareau, Boin-​Webb, and Skilton, The Buddhist
Schools of the Small Vehicle, and Cousins, “On the Vibhajjavādins:  The Mahiṃsāsaka,
Dhammaguttaka, Kassapiya and Tambapaṇṇiya branches of the ancient Theriyas,” 131–​82.
 Scripture, Commentary, and Exegetical Distinctions 69

methods of textual exegesis in relation to its practical purposes in ethical and so-
teriological development by drawing attention to the term “method of develop-
ment” (bhāvanānaya). This term, referring to the practical steps of changing and
cultivating experience, is one way the Visuddhimagga describes what it is doing.
The text is an account of the “method of development” consisting of defining
moral precepts (sīla), concentration practices (samādhi), and understanding
(paññā), structured as a sequential path. In certain cases, the textual exposition
itself constitutes the contemplative practices, as in, for example, a certain exer-
cise that involves identifying, defining, and analyzing the “elements.” As part
of a meditation exercise that involves attention to the experiences of earth, air,
wind, and fire (to help one loosen one’s attachment to food and the body), one
engages in the recitation of questions about the properties of these elements first
in brief, and then with more analytical and classificatory detail if necessary. This
text functions then as both commentarial exegesis and meditative practice.36
Another example would be the Recollections of the Buddha and the Dhamma,
which we have already had occasion to study, and which are both exegesis in the
commentary on the Vinaya (and elsewhere) and contemplative practices in the
Visuddhimagga (see Appendices A  and B). Put another way, much meditative
practice is just the very analytic methods of textual practice aspired to by the
Analysts—​that is, exploring phenomena “with different methods.” As Bond puts
it, the explanation of the meaning (aṭṭhakathā) and the method of development
(bhāvanānaya) are “related as two sides of a coin.”37
The aṭṭhakathās that Buddhaghosa handles developed many practices that
were shared by the formal exegetical system outlined in the Nettippakaraṇa and the
Peṭakopadesa, but the explicit influence of these two manuals on Buddhaghosa’s
works is not clearly evident. Perhaps the omission of their full exegetical system is
in keeping with the antiquity of the commentaries he was editing and his overall
adherence to them (the exegetical manuals likely postdate the codification of the
canon, and go unmentioned by Buddhaghosa in his account of the First Council).
We know of only one place where he quotes the Netti by name, and except for this
one instance he does not invoke and never uses the Netti’s commentarial category
of “hāra,” as such, which is one of the hallmarks of its system.38 At the same time

36. This is sketched out in Vism 351–​52 (XI.39–​45), where he describes this particular method
of development in brief and in detail.
37. Bond, The Word of the Buddha, 180.
38.  This is Ps i.31:  ettāvatā ca yvāyaṃ—​‘‘Vuttamhi ekadhamme, ye dhammā ekalakkhaṇā
tena;Vuttā bhavanti sabbe, iti vutto lakkhaṇo hāro’’ti.—​Evaṃ nettiyaṃ lakkhaṇo nāma hāro
vutto. This is cited in Ñāṇamoli, The Guide, x.  The Visuddhimagga and Atthasālinī each
mention and apparently quote the Peṭakopadesa at Vism 141 and As 165, but the passages they
70 Build ing Blo cks for an In t er pretat ive Pro gr a m

many of the exegetical terms and techniques that the Netti and the Peṭaka for-
malize were also used in the aṭṭhakathās and the Visuddhimagga, and we can see
clearly that all of these texts drew from the same well.
To sum up this section, we have seen that Buddhaghosa deploys several
different terms for commentary, each with its own nuances. At the heart of his
project is his work as a self-​identified Analyst (Vibhajjavādin) who does not
himself make claims but rather explores meaning by explaining it with various
methods. His description of the practices of the Vibhajjavādins is the closest we
get to a list of general criteria for exegesis—​for what is to count as appropriate
expansion of the teachings. It is a conservative and traditional approach in many
ways—​commentary is to be tethered to the “great authorities”—​even as the de-
velopment of meaning through analysis was thought to be fecund, both intellec-
tually and for meditative practice.
The remainder of this chapter takes up particular distinctions claimed about
the Buddha’s words, which were then used as commentarial protocols:  the
distinctions between speaking briefly or in detail, between meaning and phrasing,
conventional and absolute teachings, definitive and interpretable teachings, and
contextual and categorical teachings. While these distinctions are sometimes
thematized and treated in general terms in Buddhaghosa’s commentaries and the
Visuddhimagga, they are not grouped together and promoted as a generalized and
generalizable set of principles or rules for commentarial practice. Buddhaghosa’s
introductory sections discussed to this point are the most general and self-​
reflexive that he gets about text and genre, and the distinctions discussed in what
follows are not outlined systematically in them. Rather they occur in the course
of commentarial discussion, in and through which we might consider what they
mean and what they suggest about commentarial practice. We learn about them
and how to use them by looking over his shoulder, as it were, in watching him
work. Richard Nance has noted a similar feature in the Vyākhyāyukti, an exe-
getical manual attributed to Vasubandhu. While this manual articulates formal
principles on occasion, according to Nance, “most of the text amounts to a se-
ries of examples.” Instead of general procedural instructions for commentary, the
Vyākhyāyukti, like Buddhaghosa’s work, suggests that commentarial practice is

cite are not been found in the Peṭaka (Ñāṇamoli, The Guide, xiii). On the other hand, Bond (In
the Buddha’s Words, 168–​80) and Ñāṇamoli (The Guide, liii–​liv) argue that the commentators
were greatly indebted to the Netti. Ñāṇamoli says that the Netti was the “scaffolding” of
the Visuddhimagga and the aṭṭhakathā, which then is taken down (and thus invisible) once
their structures were complete. De Silva, in contrast, argues (rightly, I think) that the Netti’s
method as a whole is not present until Dhammapāla’s work (Dīghanikāyaṭṭhakathāṭīkā
Līnatthavaṇṇanā, xliii).
 Scripture, Commentary, and Exegetical Distinctions 71

learned through example and by “paying close attention to the shifting demands
made by particular texts and audiences—​not by internalizing a set of interpretive
principles that could be applied across all texts and contexts.”39 Commentarial
principles are learned in the manner of an apprentice through the practice itself
and it is through practice and the discussion that emerges in that practice that we
find them.

In Brief and in Detail

We begin with a hermeneutical distinction that notes the ways that texts can wax
and wane. That the Buddha could speak both briefly (saṃkhitta) and in detail
(vitthāra) was a feature of his speech noted in the canon itself, and runs deep in
the tradition’s understanding of pedagogy, preaching, and commentarial prac-
tice. The Buddha is often asked to explain in detail what he has taught in brief,
and, sometimes, vice versa.40 We have observed how Mahākaccāna was foremost
of those able to expand in detail what was said in brief.41 The Netti develops this
into two types of teachings that the Buddha gave according to audience: those of
dull faculties required teachings in detail while those of sharp or even medium
sharp faculties could get them in brief (though this judgment about the intelli-
gence of the audience is not made in the canonical materials).42
We have already seen two devices that set forth a teaching in brief: uddesa
(outline) and mātikā (matrix). These are often contrasted with commentary that
expands them in detail: vibhaṅga (expansion of an outline) and niddesa (expos-
itory commentary or description). These kinds of summary lists and developed
expansions are evident in all three branches of the canon and run deep in the tra-
dition. For example, in the Vinaya the recitation of the monastic rules stripped
of commentary is referred to as the Pāṭimokkha uddesa or just the “sutta” (which
Buddhaghosa glosses as the “matrix”).43 Elaborated with their narrative context
and commentary, the rules are given at length in the Suttavibhaṅga, that is, “the
expansion of the sutta.” (This is perhaps a confusing use of the term sutta here
applied to Vinaya texts, but the usage here indicates the barebones rule without

39. Nance, Speaking for Buddhas, 116.

40. For example, A i.56; M i.489 (cf. M i.286, 291; M iii.46, 53).
41. A i.24: Saṃkhittena bhāsitassa vitthārena atthaṃ vibhajantānaṃ yadidaṃ mahākaccānoti.
42. Netti 100; Cf. Netti 125.
43. The Cullavagga (Vin ii.96) distinguishes between the “received sutta” and the Suttavibhanga
(tassa neva suttam āgatam hoti no suttavibhaṅga). The commentary (Sp vi.1197) glosses “re-
ceived sutta” as the “received matrix” (mātikā āgatā).
72 Build ing Blo cks for an In t er pretat ive Pro gr a m

narrative context.) The other main section of the Vinaya Piṭaka is the Khandhaka
(comprising the Mahāvagga and the Cullavagga), which is also rich in narrative
framing and detail. But a further contraction occurs in the final book of the
Vinaya corpus, the Parivāra, a question-​and-​answer type matrix or outline of the
whole Vinaya system.
Commentary is, of course, the expansion of texts, and thus it should not
surprise that the aṭṭhakathā puts creative and dynamic use to the notion of the
contraction and extension of texts. The Visuddhimagga deploys commentarial
practices of expansion, as mentioned earlier, including the technique of providing
a set of questions (pañhākamma), that structures its commentary on a subject; a
good example of this is Buddhaghosa’s treatment of the main topics of the text.
He says that the mere listing of sīla, samādhi, and paññā is “extremely brief ”
and “so in order to help everyone” he teaches each of these in detail with sets of
questions (pañhākamma).44 He then lists eight questions in the case of sīla (e.g.,
what is sīla? what are its benefits? how many kinds of it are there? Et cetera), that
become the seeds of the long, expansive answers that constitute the rest of his
chapter on the topic.
Buddhaghosa often refers to the Visuddhimagga as the extensive, detailed
account of the teachings. Often in the course of the aṭṭhakathās he says he is
treating a subject briefly in the aṭṭhakathā but the reader should refer to the
treatment of it in the Visuddhimagga so that he need not repeat himself.45 Thus
within the commentarial genre the categories of brief and extensive provide a way
in which he describes the distinctive contributions of the different kinds of work
attributed to him.
The distinction between brief and extensive is a site for considering the varied
expression of the “whole.” Very often when providing the contextual details of a
sutta—the where, when, why and by whom it was spoken—​Buddhaghosa will
report that he is giving first the brief summary of the narrative and then will
follow with a fuller account. The “whole” account can be packed into a nutshell,
as it were, and then expanded through further narrative embellishment.46 This
provides a useful technique for dealing with the hermeneutical circle, of how

44. Vism 6 [I.16]: sīlasamādhipaññāmukhena desito pi pan’esa visuddhimaggo atisaṅkhepadesito

yeva hoti; tasmā nālaṃ sabbesam upakārāyā ti vitthāram assa dassetuṃ sīlaṃ tāva ārabbha idaṃ
pañhākammaṃ hoti. He introduces and structures samādhi similarly (Vism 84 [III.1]), and so
also paññā (Vism 436 [XIV.1]).
45. For example, As 168.
46. “This is the answer in brief summary to these questions, but in detail . . .” (Ayaṃ tesaṃ
pañhānaṃ saṅkhepavissajjanā. Vitthārato pana . . . [Pj I 158]), as just one example.
 Scripture, Commentary, and Exegetical Distinctions 73

parts must be interpreted in reference to the whole, and vice versa: we get a brief
statement of the whole first, from which the parts that come to compose it may
be understood.
Buddhaghosa has intriguing things to say about how Abhidhamma fits into
this distinction about brief and detailed teachings in ways that resist a tendency
to see these as fixed qualities of texts or genres. In one place he claims, “every-
thing in the seven books of the Abhidhamma is a ‘brief ’ teaching, even the com-
plete Paṭṭhāna with its twenty four divisions.”47 Since much of the Abhidhamma
consists of matrices, this makes sense. On the other hand, Abhidhamma is itself
detailed exposition that can go on, as we have seen, immeasurably. To this point,
he suggests when explaining the myriad ways consciousness (viññāṇa) conditions
the complex of nāma-​rūpa, that “when the method of conditions is to be shown
in detail one would have to cite the whole Paṭṭḥāna,”48 the seventh—​and most
oceanic and expansive—​of all books of the Abhidhamma. This remark suggests
that the Paṭṭhāna is the enactment of detailed expansion (and we come to see
just how huge and expansive the Paṭṭhāna is in ­chapter 4). I think that the way to
understand the possible discrepancy suggested here—​is the Abhidhamma a brief
teaching or a detailed teaching?—​is always to ask, relative to what? Abhidhamma
is, recall, a method. Sometimes it is a brief and potent mātikā making possible ex-
position that carries its topics and relations further. But it is also the very practice
of exposition in detail that expands the matrices. Relative to the whole Dhamma
that it “grows,” Abhidhamma is brief; but in its (particularly the Paṭṭhāna’s) ca-
pacity to explore conditionality and relationality, it is expanded and expansive.

Meaning and Phrasing

We have already had occasion to encounter the distinction between meaning
(attha) and phrasing (byañjana). We have seen that in the Buddha’s speech both
are “perfect” and “immeasurable” (as we saw in ­chapter 1, meaning is immeas-
urable through analysis, and phrasing through wording, expression, and ways
of showing; they are “perfect” as we see in this chapter’s epigraph). This is a
distinction that goes back to the canon. Their importance and pairing in the
suttas is well expressed by an Aṅguttara passage that says “there are two things,

47. Ps iii.200 (on M i.489, a sutta in which the Buddha says he can teach both in brief and
in detail):  catuvīsatisamantapaṭṭhānampi hi sattapakaraṇe abhidhammapiṭake ca sabbaṃ
saṃkhittameva.
48. Vism 561 [XVII.201]: vitthārato pana tassa paccayanaye dassiyamāne sabbā pi Paṭṭhānakathā
vitthāretabbā hotī.
74 Build ing Blo cks for an In t er pretat ive Pro gr a m

monks, that lead to the establishment, endurance, and nondisappearance of the

Good Dhamma. Which two? Well-​put phrasing of the words and well-​heard
meaning.”49
The hermeneutic use of the distinction is widely attested. The Nettippakaraṇa
uses meaning and phrasing to structure its entire toolkit of hermeneutical devices.
According to this text, the Buddha’s “phrasing” concerns six formal features of
phrasing: “syllables, words, phrasing, modes, language, descriptions.”50 Each of the
ways of putting ideas through phrasing is said to be “immeasurable”: when teaching
the Noble Truths, for example, “there are immeasurable words, immeasurable
syllables, immeasurable phrasing, immeasurable modes, immeasurable language,
and immeasurable descriptions.”51 For its part, meaning (attha) involves these six
practices:  “showing (making visible), making known, opening up, distinguishing,
making clear, and denoting.”52 As we will see, attha gets at something like the deno-
tative capacity of meaning, while byañjana can be, in some places, helpfully under-
stood as the connotative.
Further, meaning and phrasing are intimately and precisely connected in spe-
cific ways. When teaching the Four Noble Truths, for example,

the Bhagavan shows by letters, makes evident by words, opens up by

phrasing, distinguishes by modes, makes clear with language, and denotes by
description.53

That is, the use of the formal features of the language collectively called “phrasing”
allow for different possibilities for meaning to develop. The text also states that
the different forms of both meaning and phrasing allow for both closing off and
expanding meaning:

49. A i.59: Dveme, bhikkhave, dhammā saddhammassa ṭhitiyā asammosāya anantaradhānāya

saṃvattanti. Katame dve? Sunikkhittañca padabyañjanaṃ attho ca sunīto.
50. Netti 9: Chappadāni byañjanaṃ: akkharaṃ padaṃ byañjanaṃ ākāro nirutti niddeso.
51.  Netti 8:  Tattha aparimāṇā padā, aparimāṇā akkharā, aparimāṇā byañjanā, aparimāṇā
ākārā neruttā niddesā. This appears to be drawn from S v.430, where there are “immeasur-
able details, immeasurable phrasings, immeasurable ways of showing” (tattha aparimāṇā vaṇṇā
aparimāṇā byañjanā aparimāṇā saṅkāsanā) the Four Noble Truths.
52. Netti 9: Tattha chappadāni attho saṅkāsanā pakāsanā vivaraṇā vibhajanā uttānīkammaṃ
paññatti.
53. Netti 9: Tattha bhagavā akkharehi saṅkāseti, padehi pakāseti, byañjanehi vivarati, ākārehi
vibhajati, niruttīhi uttānīkaroti niddesehi paññapeti.
 Scripture, Commentary, and Exegetical Distinctions 75

There the Bhagavan removes with letters and words, and expands with
phrasing and modes, and details with language and descriptions.54

This suggests that the letters and words close off meaning: when I say the word
“red” I  remove blue, yellow, and so forth, to denote “red.” The qualities of the
“phrasing” and the “mode,” or register used, begin to open up connotative
possibilities with further implications:  I can describe red variously as “scarlet
red” or “blood red,” which make finer distinctions and deliver a more precise im-
pact. Further opening up meaning by providing details, I can develop phonetic
features of the language (nirutti) and add more description by referring to the
“rosy bloom of her cheek” or the way he “flushed ruby red.” Nirutti, as we will see
later, often refers to the phonetic qualities of language and how they lead to ad-
ditional connotations. To get at this quality of the phrasing my rather pedestrian
examples in English pick up on the “r” sound in red to get “rosy” and “ruby,” a
kind of phonetic exercise to expand meaning that is very common in Pali inter-
pretation of the Buddha’s words. The description of the “bloom” suggests both
the rose and the cheek and thus opens up meaning in two directions; the sound
and sense of a “flush” of red add movement and perhaps emotion to the image.
We consider later how phrasing and meaning are connected in Pali
commentarial practice in Buddhaghosa’s work. For now, I  wish to emphasize
that the Nettippakaraṇa’s formal discussion of these properties of meaning and
phrasing indicate one way that practitioners of the Pali commentarial tradition
put together the formal properties of scripture with the meaning they evoke.
I also wish to draw notice to the Netti’s claim about the qualities of phrasing being
“immeasurable”: there are innumerable ways to put ideas into language, and the
phrasings we use can help us develop meaning by closing off some possibilities
while opening up others. To return to the example the text itself gives, there are
endless ways to show the meaning of the Noble Truth of suffering—​consider the
examples of sorrow and pain in the world and the ways we can evoke them in
choices of word and sound. We should also note that the idea of vyañjanā in
Sanskrit has the sense of implied or suggested meaning and comes to have great
currency in the literary traditions. The capacity of literary language to richly
suggest implicit meanings is a noted quality of the Buddha’s words, as it is, of
course, for the kāvya tradition theorized centuries later.

54.  Netti 9:  Tattha bhagavā akkharehi ca padehi ca ugghaṭeti byañjanehi ca ākārehi ca
vipañcayati, niruttīhi ca niddesehi ca vitthāreti. See Rhys Davids and Stede, The Pali Text Society’s
Pali-​English Dictionary, under vipañcita for the grammatical use of ugghaṭanā, vipañcita, and
vitthāraṇā; see also Ñāṇamoli, The Guide, 19.
76 Build ing Blo cks for an In t er pretat ive Pro gr a m

Buddhaghosa deploys the same distinctions about meaning and phrasing that
we have just seen in the Netti. In explaining the passage we considered at some
length in the Introduction (see also Appendix A)—​“He teaches a dhamma beau-
tiful in the beginning, beautiful in the middle, and beautiful in the end, with
meaning and phrasing; he makes known a religious life that is entirely perfect
and pure”—​the Visuddhimagga defines “with meaning” and “with phrasing,”
with the same formulations we just saw in the Netti. “With meaning” is “putting
together word and sense via showing, making evident, opening up, distinguishing,
making clear, and denoting.”55 “With phrasing” concerns “excellence in syllables,
words, phrasing, mode, language, and description.”56 Buddhaghosa also says
that “with meaning” is about “depth in comprehension and depth in meaning,”
and involves a scope of [two of the four kinds of ] analysis (paṭisambhidā)—​the
analyses of things (attha) and of knowledge (paṭibhāna); it inspires “joyful faith
in like-​minded people because of being experienced by the wise, and it has deep
intent.”57 The emphasis on the meaning in the Buddha’s words is about the un-
derstanding and analysis they prompt, how the well-​spoken words of the Buddha
create comprehension.
The phrasing of buddhavacana is also said to inspire “serene delight in or-
dinary people because of their faith,” and its “words are clear.”58 Phrasing refers
thus to the pleasing and clear qualities of the Buddha’s preaching. Elsewhere

55.  Vism 214 (VII.72):  sankāsaṇa-​pakāsana-​vivaraṇa-​vibhajana-​uttānīkaraṇa-​paññatti-​

atthapadasamāyogato sātthaṃ. Note that this follows very closely Netti 8.
56. Vism 214 (VII.72): akkhara-​pada-​byañjanākāra-​nirutti-​niddesa-​sampattiyā sabyañjanaṃ.
Exactly what “mode,” “aspect,” or “register” (ākāra) means in the list of kinds of phrasing as a
linguistic feature is not entirely clear; it may be about “style” or “mood.” Ñāṇamoli translates
ākāra as “style” in this list in his translation of the Vism (The Path of Purification, 211), and
as “mood” in The Guide, 107. As Hallisey has explored, Buddhaghosa suggests that ākāra can
sometimes concern the particular way spoken speech addresses a particular person; when
Ᾱnanda utters “evaṃ me sutaṃ,” Buddhaghosa says, he is reporting the particular mode in
which he, Ᾱnanda, heard the Buddha’s sermon. Ᾱnanda says, “there was one mode which
was heard by me,” suggesting that each hearer will hear and understand buddhavacana in a
mode particular to them. Hallisey suggests “ākāra” is the difference between reading an obit-
uary about a person’s death and getting a telegram that says, “Dad has died, come home”: the
meaning is the same, but there are different registers in these cases and they have different force
depending on the recipient of such news. News comes to us in different registers, and we fur-
ther inflect how it is received by who we are and our relationship to it (Hallisey, “Buddhaghosa
on the Futures of Scriptures”).
57. Vism 214 (VII.72): atthagambhīratā-​paṭivedhagambhīratāhi sātthaṃ . . . atthapaṭibhānap
aṭisambhidāvisayato sātthaṃ . . . paṇḍitavedanīyato sarikkhakajanappasādakanti sātthaṃ . . .
gambhīrādhippāyato sātthaṃ.
58.  Vism 214 (VII.72):  saddheyyato lokiyajanappasādakan ti sabyañjanaṃ  .  .  . uttānapadato
sabyañjanaṃ.
 Scripture, Commentary, and Exegetical Distinctions 77

Buddhaghosa defines “phrasing” as the qualities or features of verbal speech,

and he enumerates ten qualities regarding it. The speech of a teaching may be
“indistinct or pronounced, long or short, heavy or light, nasalized, connected,
separated, and free” of certain other languages like Tamil or that of tribal peoples
and barbarians.59 These qualities indicate features of the spoken word and also
notable claims about the ideal language of teaching, suggesting even that certain
languages be avoided.
The linguistic features of phrasing are said to include “depth in the dhamma
and in teaching,” and involve the scope of the other two (of the four) “kinds of
analysis (paṭisambhidā)—​the analysis of the Dhamma and the analysis of lan-
guage.”60 Buddhaghosa elsewhere gives several different senses of what dhamma
means, emphasizing “condition” (paccaya), but also, in a different sense, “what is
spoken” (bhāsita), referring to scripture. I think that here he means the Dhamma
in the sense of the spoken scripture because of his attention to the sound of the lan-
guage in this discussion.61 The analysis of language (nirutti) concerns, according
to Buddhaghosa, a knowledge that one can acquire through analyzing the sound
of the “natural language” (by which he means Māgadha language, that is, Pali);
this analysis “has sound as its object, not the concept.”62 This is to say that one

59.  Ps ii.203:  ‘Sithilaṃ dhanitañca dīgharassaṃ, garukaṃ lahukañca niggahītaṃ;

sambandhaṃ vavatthitaṃ vimuttaṃ, dasadhā byañjanabuddhiyā pabhedo’’ti  .  .  . tassa
damiḷakirāsavarādimilakkhūnaṃ bhāsā viya byañjanapāripūriyā abhāvato abyañjanā nāma
desanā hoti. Law, Buddhaghosa, 85, is helpful on interpreting this discussion, but his edition is
somewhat different from the Chaṭṭhasaṅgayana.
60. Vism 214 (VII.72): dhammagambhīratādesanāgambhīratāhi sabyañjanaṃ . . . dhammaniru
ttipaṭisambhidāvisayato sabyañjanaṃ.
61.  Vism 441 (XIV.23); Vibh-​a 386:  “Dhamma is in brief a term for condition. For since a
condition sets up, causes, or allows something to occur, it is called dhamma. But dhamma can
be further divided into these five kinds of dhammas: 1. whatever cause gives rise to a result,
2. The noble path, 3. what is spoken, 4. what is good, and 5. what is not good. For one reviewing
that dhamma, the knowledge which falls into the category of that dhamma, is the analysis of
dhamma” (dhammo ti sañkhepato paccayassa ‘etaṃ adhivacanaṃ. Paccayo hi yasmā taṃ taṃ
dahati pavatteti vā sampāpuṇituṃ vā deti tasmā dhammo ti vuccati. Pabhedato pana yo koci
phalanibbattako hetu ariyamaggo bhāsitaṃ kusalaṃ akusalaṃ ti ime pañca dhammā dhammo
ti veditabbā. Taṃ dhammaṃ paccavekkhantassa tasmiṃ dhamme pabhedagataṃ ñāṇaṃ
dhammapaṭisambhidā). As 38 also defines dhamma to include scripture (pariyatti) and con-
dition (hetu), among other things (on dhamma, see also Ps i.17). Nance describes how north
Indian commentators advanced various interpretations of dharmapratisaṃvid (Speaking of
Buddhas, 72–​78).
62. Vibh-​a 387: The whole passage is “for one reviewing that natural language by making the
sound of the natural language the object [of his reflection] when there is an utterance of it,
that knowledge which falls into the category of an utterance of the natural language is the
analysis of language. In this way, the analysis of language comes to have the sound as its object,
not the concept as its object” (tassā abhilāpe taṃ sabhāvaniruttiṃ saddaṃ ārammaṇaṃ katvā
78 Build ing Blo cks for an In t er pretat ive Pro gr a m

can acquire knowledge by attending to the phonetic properties of the language.

It is hard to overemphasize just how important the sound of language is when
interpreting the Buddha’s words, evident in the extensive use of nirutti analysis as
a commentarial device for expanding meaning. Buddhaghosa goes on to explain
that one can hear grammatical mistakes in how the words sound and can thus
analyze language grammatically (though he allows that some people, wrongly he
thinks, take this kind of analysis to mean being adept in many languages—​a skill
he takes to require great knowledge, but not analysis).63
As is well known, the Pali commentators insisted that the Buddha’s teaching
was originally delivered in the Māgadha language (what we call Pali), a lan-
guage they regarded as having special status. Māgadha is considered to be the
language one would speak if one’s parents did not intervene with another lan-
guage, or if one grew up feral. It is thought to be the language prevalent in all
the realms—​not only deities and humans speak it, but even ghosts, animals, and
hell-​beings,64 marking its universality rather than its exclusivity (in contrast to
Sanskrit). Kate Crosby suggests that such claims about its primacy, universality,
and innateness contrast Pali to other languages seen as derived.65 Because of its
unique qualities, Buddhaghosa insists that Māgadha is the language chosen by
the Buddha for his buddhavacana, the scripture of the Tipiṭaka.66 He gives this
reason for this:

Why? Because in this way the meaning is easy to grasp. For [when] the
scripture of the Buddha’s words is delivered in the Māgadha language
the only delay [in understanding it] for those who have acquired analysis
(paṭisambhidā) is the time it takes to reach their ears; for just when the ear
is struck the meaning arises in a hundred methods, a thousand methods.
But the scripture delivered in other languages must be learned, correcting

paccavekkhantassa tasmiṃ sabhāvaniruttābhilāpe pabhedagataṃ ñāṇaṃ niruttipaṭisambhidā.

Evamayaṃ niruttipaṭisambhidā saddārammaṇā nāma jātā, na paññattiārammaṇā).
63. Vibh-​a 387. Here Buddhaghosa is indicating his awareness of alternative Buddhist views.
In certain Mahāyāna texts, nirukti-​pratisaṃvid is taken to be a linguistic capacity exhibited
by advanced bodhisattvas to understand “any and all languages” (Buswell and Lopez, eds., The
Princeton Dictionary of Buddhism, 589; Nance, Speaking of Buddhas, 60–​61).
64. Vibh-​a 387–​88.
65.  Crosby, Traditional Theravada Meditation and Its Modern-​Era Suppression, 82–​84. See
also Granoff, “Buddhaghosa’s Penance and Siddhasena’s Crime,” 23–​24, which suggests that
Buddhaghosa aligned Māgadha with certain features of Sanskrit as a sacred language.
66.  Vibh-​a 388:  Sammāsabuddhopi tepiṭakaṃ buddhavacanaṃ tantiṃ āropento
māgadhabhāsāya eva āropesi.
 Scripture, Commentary, and Exegetical Distinctions 79

it again and again. And for the ordinary person even who studies a lot,
there is no acquiring [this] analysis.67

These intriguing claims suggest an immediacy and directness in Māgadha that

scripture given in other languages cannot match, at least for those primed for
it. The emphasis on the uniquely potent possibilities of the Māgadha version
of scripture for generating hundreds even thousands of methods for grasping
meaning becomes the very reason that buddhavacana was originally in Māgadha.
Additionally, this theory is in keeping with a style of interpreting buddhavacana
by phonetic or linguistic (nirutti) analysis of the spoken sound of language that
became one of the most prevalent interpretative techniques in Buddhaghosa’s
commentaries. Nirutti analysis is based on the idea that there is an innate meaning
encoded in the phonetic properties of words in this scriptural language.
The ways that Pali phrasing works to stimulate analysis and the production
of meaning require an example to be appreciated. The importance of the pho-
netic qualities of the Buddha’s speech led to commentarial practices that develop
meaning based on the sounds of the syllables and words in the suttas (a prac-
tice widely used and sometimes theorized in other Indic commentarial traditions
also68.) It scarcely overstates the matter to say that Buddhaghosa makes use of
nirutti analysis on nearly every page of his commentaries. This attentiveness to
the sounds of buddhavacana in Pali allows Buddhaghosa to develop implicit
meaning, though this practice can pose a challenge for the translator to convey
what is going on, because niruttis seldom work in translation. But once one
becomes attentive to nirutti analysis as a commentarial device one can see just
how generative the use of the phonetic qualities of buddhavacana can be for the
production of meaning.

67. Vibh-​a 388: Evañhi atthaṃ āharituṃ sukhaṃ hoti. Māgadhabhāsāya hi tantiṃ āruḷhassa

buddhavacanassa paṭisambhidāppattānaṃ sotapathāgamanameva papañco; sote pana
saṅghaṭṭitamatteyeva nayasatena nayasahassena attho upaṭṭhāti. Aññāya pana bhāsāya tantiṃ
āruḷhaṃ pothetvā pothetvā uggahetabbaṃ hoti. Bahumpi uggahetvā pana puthujjanassa
paṭisambhidāppatti nāma natthi. Cf. Vism 441–​42 (XIV.25). I take sodhetvā for pothetvā here,
on the advice of Rhys-​Davids and Stede, The Pali Text Society’s Pali-​English Dictionary, 475.
68. See Kahrs, Indian Semantic Analysis, for a close study of nirukta and nirvacana analysis in
Sanskrit texts. He notes that the use of nirukta analysis in Yāska’s work is based on a theory that
analysis of the syllables and sounds of words shows how the word itself conveys meaning di-
rectly and immediately, without reference to other words. A term can be said to be “anvartha,”
meaning that “there is a direct semantic link between the term and the nature of that which it
denotes” (47–​48). The sounds of the term itself contain the elements necessary for interpreting
its meaning. While Kahrs does not mention the Pali tradition, what he describes has close
affinities with Buddhaghosa’s theory about the immediacy of Māgadha and his practices with
nirutti.
80 Build ing Blo cks for an In t er pretat ive Pro gr a m

For example, Buddhaghosa provides many nirutti analyses in his treatment of

the various epithets of the Buddha in the opening lines of the Vinaya, where the
Buddha is praised by the oft-​mentioned list of nine epithets: “for the Bhagavan
is an Arahat, Perfectly Awakened, Accomplished in Knowledge and Conduct,
Well-​gone, Knower of Worlds, Highest Coachman of Men to be Tamed, Teacher
of Deities and Humans, the Buddha, the Bhagavan.”69 In his commentary on this
passage (which also occurs in the Visuddhimagga), Buddhaghosa provides nirutti
analyses on each epithet; we can focus on his treatment of the first of these, Araha
(“Worthy”) to show something of how nirutti works as a commentarial device
(for a translation of this entire passage see Appendix B). On the word Araha,
Buddhaghosa says:

The Bhagavan should be understood as an “Araha” because of these

reasons: because of [his] condition of being aloof (ārakattā), because of
enemies (arīnam), because of destroying the spokes (arānañca hatattā),
because of worthiness (arahattā) of requisites, et cetera, and because of
the lack of secret (raha) evil doing.70

These definitions pick up on the sounds “a” and “ra” and “ha” in “araha” to gen-
erate possible meanings. We can follow Buddhaghosa further as he elaborates on
these words:

“Being aloof ” (āraka) means being established far and removed from all
kilesas, [that is] the condition of having destroyed the kilesas together with
their traces with the Path, and so [he is] “Araha” because of being aloof;
and by this Path these enemies that are the kilesas are destroyed (hatā),
and so he is Araha due to the condition of having destroyed (hatattā) his
enemies (arī). Now he is Araha also because of his condition of destroying
enemies, in that all enemies are destroyed, his having wielded with the

69. Vin 3.1: ‘itipi so bhagavā arahaṃ sammāsambuddho vijjācaraṇasampanno sugato lokavidū

anuttaro purisadammasārathi satthā devamanussānaṃ buddho bhagavā. This stylized string of
epithets is quoted extensively in the canon, and is the itpiso gāthā, a widely chanted verse used
in many contexts. We discuss these qualities of the Buddha, and Buddhaghosa’s commentary
on them in ­chapter 5.
70. tattha ārakattā, arīnaṃ arānañca hatattā, paccayādīnaṃ arahattā, pāpakaraṇe rahābhāvāti
imehi tāva kāraṇehi so bhagavā arahanti veditabbo. This sizable chunk of commentary occurs
in the Vinaya aṭṭhakathās (Sp i.112 ff.) and also in the Visuddhimagga (198–​201, VII.4–​25). In
the Visuddhimagga, this extended commentary on the nine epithets occurs in the context of
the samādhi practices of Buddha recollections (see Ñāṇamoli, trans., The Path of Purification,
192–​209).
 Scripture, Commentary, and Exegetical Distinctions 81

hand of faith the hatchet of knowledge that puts an end to kamma,

standing firm on the ground of morality71 with feet of heroism, on the
grounds of the Bodhi Tree, having destroyed the wheel of saṃsāra with
its spokes (ara) that are the constructions such as merit, et cetera, with
its hub consisting of ignorance and craving for becoming, with its axle
consisting of the origins of the oozings (āsava), with its rim of old age
and death, joined to the chariot of the three forms of becoming, revolving
from beginningless time.72

The first definition, “being aloof,” is itself elaborated by use of another

nirutti:  the phonetic properties of āraka evoke, with the ka sound, what the
Buddha is aloof from—​the kilesas, or the defilements from which arahats are
free. Further, as Araha he has destroyed (hatā) his enemies (arī), a further devel-
opment of all three phonemes (a, ra, and ha) in araha. These elaborations allow
Buddhaghosa to work in a small biographical moment of the Buddha, praising
him for his victory at the Bodhi Tree, that also manages to include the next def-
inition of what he destroyed, namely the spokes (ara) of the wheel of saṃsāra.
Once introduced, the wheel of saṃsāra becomes a metaphor that gets additional
development. In fact, he offers several different readings of the wheel of saṃsāra
that take him through the next three pages of his commentary, which involve a
series of alternative modular mappings of the wheel of saṃsāra on the twelvefold
dependent origination schema (which would take us too far beyond our task here
to go through).
We are not done yet: still to be elaborated in the original nirutti definitions
are the phrases “because of worthiness (arahattā) of requisites, et cetera, and be-
cause of the lack of secret (raha) evil doing.”

71. Attentive readers will notice the distant echo of the very first words of the Visuddhimagga
in this: sīle patiṭṭhāya, here slightly broken up but still present as sīlapathaviyaṃ patiṭṭhāya,
“standing firm on the ground of morality” (Vism 1). The pleasures of attending to the phrasing
of the words begin to reveal Buddhaghosa’s playful erudition with language, as well as the way
that meaning can referenced, modular fashion, from other contexts: the foundation of mo-
rality that begins the Visuddhimagga was initiated during (and should be understood in re-
lation to) the time when the Buddha stood firm on morality at the Bodhi Tree. I am grateful
to Charles Hallisey and his Fall 2016 “Readings in Pali Commentary” class for letting me join
their class to read this passage with them and share these observations.
72.  Ārakā hi so sabbakilesehi suvidūravidūre ṭhito, maggena savāsanānaṃ kilesānaṃ
viddhaṃsitattāti ārakattā arahaṃ; te cānena kilesārayo maggena hatāti arīnaṃ hatattāpi
arahaṃ. Yañcetaṃ avijjābhavataṇhāmayanābhipuññādiabhisaṅkhārāraṃ jarāmaraṇanemi
āsavasamudayamayena akkhena vijjhitvā tibhavarathe samāyojitaṃ anādikālappavattaṃ
saṃsāracakkaṃ, tassānena bodhimaṇḍe vīriyapādehi sīlapathaviyaṃ patiṭṭhāya saddhāhatthena
kammakkhayakaraṃ ñāṇapharasuṃ gahetvā sabbe arā hatāti arānaṃ hatattāpi arahaṃ.
82 Build ing Blo cks for an In t er pretat ive Pro gr a m

And because of being foremost among those deserving alms, he is worthy

(arahati) of requisites such as robes and of distinction in worship. And
because of this when a Tathāgata has arisen, eminent deities and humans
do not worship anyone else. Indeed, Brahmā Sahampati worshipped the
Tathāgatha with a jewel garland the size of Mt. Sineru, and other deities
and humans such as Bimbasāra and the king of Kosala did likewise
according to their means. And the great king Asoka established 84,000
monasteries throughout Jambudīpa, having dispersed 96 koṭis of wealth
dedicated to the Bhagavan who had attained final nibbāna. Why speak
of any distinction in worshipping anyone else? In this way, he is Araha
because of being worthy of requisites, et cetera. Moreover, in the world
there are fools proud of their cleverness who do evil in secret (raho) with
fear of blame; he never does this, and so is worthy because of the lack of
secret evil-​doing.73

In addition to elaborating on the sound araha, for “worthy,” this commentary

further develops the “pa” in the original “paccayādīnaṃ arahattā,” “worthy of
requisites, et cetera,” to develop the notion of worship (pūja) with these impres-
sive examples of royal largesse. Finally, the Bhagavan is an “Araha” “because of
his lack of secret evil-​doing” (pāpakaraṇe rahābhāvāti imehi tāva kāraṇehi), de-
veloped here with playful consonance of “pa” and “ka” to distinguish him from
whatever (keci) fools proud of their cleverness (paṇḍitamānino) who do clandes-
tine wrongdoing.
By following this commentarial device in practice we can see how it works.
The sounds of the epithet lead to other words that then can be elaborated to
develop further meaning and possibility. From the single word “Araha” we have
begun to learn what Buddhaghosa thinks it is important to know and say about
the Buddha—​his destroying the kilesas, his dismantling the wheel of saṃsāra (and
further how saṃsāra can be understood), the kings who have worshipped him,
and so forth. The practice may seem to us to be more “free association” than his-
torical etymology, but it occurred in scholarly contexts of people deeply immersed
in the spoken qualities of the Buddha’s words and for whom developing these

73. Aggadakkhiṇeyyattā ca cīvarādipaccaye arahati pūjāvisesañca; teneva ca uppanne tathāgate

ye keci mahesakkhā devamanussā na te aññattha pūjaṃ karonti. Tathā hi brahmā sahampati
sinerumattena ratanadāmena tathāgataṃ pūjesi, yathābalañca aññepi devā manussā ca
bimbisārakosalarājādayo. Parinibbutampi ca bhagavantaṃ uddissa channavutikoṭidhanaṃ
visajjetvā asokamahārājā sakalajambudīpe caturāsītivihārasahassāni patiṭṭhāpesi. Ko pana
vādo aññesaṃ pūjāvisesānanti! Evaṃ paccayādīnaṃ arahattāpi arahaṃ. Yathā ca loke keci
paṇḍitamānino bālā asilokabhayena raho pāpaṃ karonti; evamesa na kadāci karotīti pāpakaraṇe
rahābhāvatopi arahaṃ.
 Scripture, Commentary, and Exegetical Distinctions 83

phonetic associations displayed sophisticated practices of memory, erudition, and

phono-​aesthetics in the intellectual and literary culture in which Buddhaghosa is
participating. Moreover, nirutti analysis is concordant with both Buddhaghosa’s
theory of the immediacy and universality of Māgadha—​a language that when
it strikes one’s ear suggests meaning through “a hundred methods, a thousand
methods”—​and his claim that the Buddha’s words are immeasurable in meaning
and phrasing. Recall too that all of this discussion concerns just the first of the
nine epithets of the Bhagavan in the itipiso formulation; each of the other eight
garner many pages of commentary that provide extensive teaching on who the
Buddha is by modeling praise and remembrance of his qualities.74 The entire
passage occurs in both the buddhānussati section of the Visuddhimagga (chapter
VII), where the Recollections of the Buddha are contemplative exercises, and in
Buddhaghosa’s commentary on the opening section of Vinaya, where praise and
contemplation of the Buddha is essential stage-​setting for interpreting the Vinaya
itself. In both, these practices have the force of making an immediate and existen-
tial connection of the practitioner with the Buddha: as Buddhaghosa puts it, by
practicing Recollections of the Buddha one “has the awareness of living together
with the Teacher.”75
As we pull back from these considerations to look at their implications, we
can see that meaning and phrasing are tightly woven together in Buddhaghosa’s
conception in a way that differs from certain Mahāyāna texts that raise the pos-
sibility of a conflict between them. Buddhaghosa never allows that attha and
byañjana might conflict and present the interpreter with the predicament of de-
ciding which should prevail. We do not find a notion of the “spirit” of a teaching
trumping the “letter,” as Étienne Lamotte has described as a hermeneutic prin-
ciple in the Catuḥpratisaraṇasūtra, a Mahāyāna text, for example.76 While on

74.  The nirutti analyses of these epithets of the Buddha are worth considering next to the
example Kahrs gives of the Sanskrit exegesis by Abhinavagupta on the tantric deity Bhairava
(Kahrs, Indian Semantic Analysis, ch. 3). The name “Bhairava” communicates directly the
nature of this deity, since it is a name thought to be inherently significant; analysis consists of
drawing out the verbal roots heard in the phonemes of this name that establish the nature of
Bhairava as the absolute deity. Kahrs is right to note how this practice occurs within a larger
belief system and is an authorized discourse of power that enables “those who mastered it to
enforce or modify beliefs by encoding meaning into already existing terms” (62). An entire
metaphysic of the absolute deity is promoted as inherent in the name itself. I think that these
same processes are going on in Buddhaghosa’s treatment of the nine epithets of the Buddha.
75. Vism 213 (VII.69): buddhānussatim anuyutto . . . satthārā saṃvāsasaññam paṭilabhati. We
discuss this entire passage at greater length in ­chapter 5.
76.  Lamotte, “Assessment of Textual Interpretation in Buddhism,” 13–​14, and “La critique
d’interprétation das le bouddisme,” 342, 44.
84 Build ing Blo cks for an In t er pretat ive Pro gr a m

certain occasions Buddhaghosa suggests that the intent (adhippāya) of the text
should prevail over the phrasing,77 he nowhere suggests that as a general prin-
ciple meaning (attha) is more important than phrasing (byañjana). As Bond has
shown, “intent” in certain contexts is a matter of stating the “gist” of a word or
idea as part of the padavaṇṇanā; it is not the same as the “meaning” or the kind
of definitional work of the atthavaṇṇanā in specifying the meaning in relation to
the Dhamma.78 Lamotte gets closer to the Pali view when he discusses “the two-
fold perfection” repeated frequently (as we have seen), where the Good Dhamma
is “perfect in attha and byañjana.”79 The perfect coincidence of perfect meaning
and perfect phrasing in the unique speech of the Buddha does not make a clash
between them possible, whereby a commentator could appeal to the former in-
stead of the latter.
Lamotte cites an exchange in the canonical Pali Vinaya between the monk
Assaji and Sāriputta, where Assaji is new to the Dhamma and is not confident
about explaining it “in detail.” Sāriputta then urges: “tell me a little or a lot, but
tell me the meaning; [just give] me the meaning with its purport, why bother
with long phrasings?”80 As I  read it, this request is not a general license to a
reader to give greater weight to the Buddha’s meaning over his phrasings, but
for Assaji to describe in brief something he would have difficulty relating in de-
tailed wording and illustration. Buddhaghosa does not pick up this passage and
discuss it at all; the instruction to Assaji does not become a general principle for
preaching or interpretation. It should also be said that Buddhaghosa is not above
accusing those of contesting views of being overly literal in their interpretations.
At one point he criticizes an alternate reading of an issue by saying “do not slander
the Bhagavan by seeking just the shelter of the phrasing; for the Buddha’s word
is deep and should be understood from the intent, having attended closely to the
teachers.”81 Here again, phrasing and meaning are not being opposed, but rather
the opponent is accused of doing a superficial and literal reading of a passage’s
phrasing that should be read deeply for its intent.

77. Vism 322: tasmā byañjanacchāyāmattaṃ gahetvā mā Bhagavantaṃ abbhācikkhi. Gambhīraṃ

hi buddhavacanaṃ. Taṃ ācariye payirupāsitvā adhippāyato gahetabbaṃ.
78. Bond, The Word of the Buddha, 152 (citing in support Pj I.123, 238).
79. Lamotte, “Assessment of Textual Interpretation in Buddhism,” 13.
80.  Vin i.41:  Appaṃ vā bahuṃ vā bhāsassu, atthaṃyeva me brūhi; Attheneva me attho, kiṃ
kāhasi byañjanaṃ bahu”nti. As Horner notes, there is some modern controversy on this
passage (The Book of the Discipline, IV, 53 n.3).
81. Vism 322: tasmā byañjanacchāyāmattaṃ gahetvā mā Bhagavantaṃ abbhācikkhi. Gambhīraṃ
hi Buddhavacanaṃ. Tan ācariye payirupāsitvā adhippāyato gahetabbaṃ.
 Scripture, Commentary, and Exegetical Distinctions 85

To sum up, we have seen that the Buddha’s meanings and phrasings are both “im-
measurable” and that the reader should consider both of these qualities of scripture
in a task of discovering infinity rather than closing down with totality. Moreover,
as suggested by the epigraph for this chapter, each one is “perfect” in itself as well as
related to the other: the formal and phonetic elements of a word or phrase deliver a
proliferation of meanings that advance the reader in his exploration of the Buddha’s
limitless knowledge and attainments. We have attended especially to nirutti analysis
among the different kinds of analyses, as nirutti is the most widely and explicitly
practiced. Drawing meaning from the phonemes in a Māgadha word of scripture
suggests that words themselves contain their meaning. In the example considered, an
entire buddhology can be heard in the sounds of the epithets given to the Buddha.

Conventional Teachings and

Teachings in the Furthest Sense
The well-​known contrast between “conventional” teachings (sammutidesanā)
and teachings in the “further sense” or “furthest sense” (paramatthadesanā)
is not made in the canon. In the earliest sources, the two terms sammuti and
paramattha are not paired or opposed in a distinction about truth or language. It
was subsequent Buddhist traditions, initiated likely by Nāgārjuna, that came to
deploy a distinction in a Mahāyāna context between what may be translated as
“conventional and ultimate truth” as a useful hermeneutical device in ways rel-
evant to often quite significant epistemological and metaphysical questions.82
Buddhaghosa’s treatment of this contrast is quite different, as I demonstrate ex-
tensively in this section.
While the contrast is not present in the Pali canonical sources, the terms
sometimes appear independently; sammuti occurs infrequently, referring vari-
ously to conventional terms, views, or agreed on practices or usage.83 Paramattha
can mean “highest good” or “highest attainment,” as for example when nibbāna

82. The idea of two truths appears to have begun with Nāgārjuna (Mūlamadhyamakakārikā
XXIV. 8–​9), though the ways that it developed in the Sanskrit traditions differed from the role
it played for Buddhaghosa and Dhammapāla. As discussed by Kalupahana, it was Vasubandhu
who developed this distinction into a metaphysical claim about different realities, saṃvṛti-​sat
and paramārtha-​sat (Akb p.  334, as cited by Kalupahana, Nāgārjuna:  The Philosophy of the
Middle Way, 332).
83.  For example, S i.135 uses sammuti to describe our conventional name of “being” for a
collection of “aggregates” (like the name “chariot” for a certain collection of parts); Vin v.222
uses sammuti for the sense of “agreement” between monks in the context of formal legal
proceeding. Miln 3 uses sammuti contrasted with suti in a manner echoing the Sanskrit dis-
tinction of revealed or “heard” (śruti) as opposed to received or “remembered” (smṛti) lore.
86 Build ing Blo cks for an In t er pretat ive Pro gr a m

is defined as “the [noble] truth of the highest attainment.”84 When the Pali
commentarial tradition begins to pair sammuti and paramattha, it does so to
make a distinction about different kinds of teachings (desanā) or discourses
(kathā),85 rather than truth claims about reality. For reasons that will become
clear, it can be useful in this context to avoid the usual translations of paramattha
as “ultimate truth” or “absolute meaning,” which would seem to place this kind
of discourse at a higher level of truth, because this is not how Buddhaghosa takes
it. I prefer “furthest sense” or “furthest meaning,” which indicate that this kind of
language or teaching goes further than conventional discourse in making analytic
distinctions.
Buddhaghosa explains that the Buddha’s choice of which of two kinds of
teachings to give depends on the topic under discussion and the capacities of his
audience. When the Buddha uses such terms as “person, being, woman, man,
kshatriya, Brahmin, deva, and Mara,” he is using a “conventional teaching.” But
when he speaks in such language as “impermanence, suffering, no-​self, aggregates,
elements, bases, and the foundations of mindfulness,” he is giving teachings in the
“furthest sense.”86 Words like “person,” “woman,” and “man” are readily under-
stood by most people, but “aggregates” and “elements” take a more specialized an-
alytic knowledge to grasp as they are further analytically reduced. The Buddha’s
choice of language also depends on whom he is addressing: “if someone, having
heard a conventional teaching, is able to attain distinction, having abandoned
delusion and comprehended the meaning, then [the Buddha] teaches a con-
ventional teaching.” So too for a teaching in the furthest sense.87 Both kinds of

Thī-​a 216 uses sammuti in the sense of “agreement.” Vibh-​a 164 uses sammuti to refer to the
common or customary term for an idea.
84. Mp iii.161; Spk i.238, 329; Ps v.59; Sv iii.1022: “he shows that nibbāna is the truth of the
highest attainment” (paramatthasaccaṃ nibbānaṃ dasseti). Kalupahana suggests that the
Buddha as we get him in the nikāyas was not interested in positing ultimate views or ulti-
mate reality, but rather spoke of ultimate fruits, consequences, and results (Kalupahana,
Nāgārjuna: The Philosophy of the Middle Way, 331–​32).
85. Ps i.137; Mp i.94: “The teachings of the Lord Buddha are twofold: conventional teachings
and teachings in the furthest sense” (buddhassa bhagavato duvidhā desanā sammutidesanā,
paramatthadesanā cāti). Cf. Sv ii.382 and KvA 34: buddhānaṃ pana dve kathā sammutikathā
ca paramatthakathā ca.
86.  Mp i.94–​95; Ps i.137 (Cf. Spk ii.77; ItiA 82; KvA 34):  Tattha ‘‘puggalo satto itthī puriso
khattiyo brāhmaṇo devo māro’’ti evarūpā sammutidesanā, ‘‘aniccaṃ dukkhaṃ anattā khandhā
dhātū āyatanāni satipaṭṭhānā’’ti evarūpā paramatthadesanā.
87.  Mp i.94–​95; Ps i.137 (ItiA 82):  Tattha bhagavā ye sammutivasena desanaṃ sutvā atthaṃ
paṭivijjhitvā mohaṃ pahāya visesaṃ adhigantuṃ samatthā, tesaṃ sammutidesanaṃ deseti. Ye
pana paramatthavasena desanaṃ sutvā atthaṃ paṭivijjhitvā mohaṃ pahāya visesamadhigantuṃ
samatthā, tesaṃ paramatthadesanaṃ deseti.
 Scripture, Commentary, and Exegetical Distinctions 87

language are therapeutic and can help people abandon delusion, but they need to
be deployed appropriately.
This pragmatic approach to terminology is refined further when Buddhaghosa
gives eight specific topics on which the Buddha reserved the use of conventional
terms: for illuminating (1) shame and apprehension, (2) a person’s own karma,
(3)  individual agency, (4)  the immediate (effects of karma), (5)  the sublime
abidings, (6)  former dwellings (in previous lives), (7)  the purification of gifts,
and (8) for the sake of not abandoning the conventions of the world.88 As he goes
on to explain, this is eminently practical. Technical analytic language is confusing
when talking about shame, loving-​kindness, the workings of karma and rebirth,
and so on. For these topics one is better off using ordinary talk of “persons”:

When it is said that “the aggregates, elements, and bases have shame and
have apprehension,” the crowd does not understand, becomes confused,
and rebels, [saying], “how can it be that aggregates, elements, and bases
have shame and have apprehension?” But when one says that “a woman has
shame and apprehension, a man has shame and apprehension, a Brahmin,
a god, or Māra,” then [people] understand, are not confused, and do not
rebel. Because of this, the Buddha taught a teaching on persons for the
purpose casting light on shame and apprehension, et cetera.89

So too, with the sublime abidings, constituting as they do the varieties of love
toward “persons” (rather than collections of “aggregates”); and karma, and the
other topics mentioned as appropriate for conventional buddhavacana. Thus the
Buddha’s choice of conventional language is not configured as a concession to
the ignorant ways of the world, but as a pragmatic and skillful decision about
what sorts of language to apply in different contexts. Language that goes furthest
(parama) analytically resulting in aggregates, elements, and bases, has its place
too. As Charles Hallisey has shown, these two uses of language are not ranked.
They are just different uses of language, such as we might use in talking about a

88.  Mp i.95; Ps i.138 (ItiA 82):  Apica aṭṭhahi kāraṇehi bhagavā puggalakathaṃ katheti—​
hirottappadīpanatthaṃ, kammassakatādīpanatthaṃ, paccattapurisakāradīpanatthaṃ,
ānantariyadīpanatthaṃ, brahmavihāradīpanatthaṃ, pubbenivāsadīpanatthaṃ,
dakkhiṇāvisuddhidīpanatthaṃ, lokasammutiyā appahānatthañcāti.
89.  Mp i.96:  “Khandhadhātuāyatanāni hiriyanti ottappantī” ti hi vutte mahājano na jānāti,
sammohamāpajjati, paṭisattu hoti “kimidaṃ khandhadhātuāyatanāni hiriyanti ottappanti
nāmā” ti? “Itthī hiriyati ottappati, puriso khattiyo brāhmaṇo devo māro” ti vutte pana jānāti, na
sammohamāpajjati, na paṭisattu hoti. Tasmā bhagavā hirottappadīpanatthaṃ puggalakathaṃ
katheti.
88 Build ing Blo cks for an In t er pretat ive Pro gr a m

star as part of a constellation in the night sky or, in contrast, describing it in the
technical terminology of astrophysics.90 Neither is more true than the other, nor
does the truth of one demonstrate the falseness of the other, and which is “better”
depends on the context. As Hallisey points out, the distinction is formulated not
to pit them against each other as opposites but to create a larger framework that
gives value to both (this is true for the other pairings described in this chapter as
well).91
Further, Buddhaghosa likens the two kinds of teaching to the use of different
languages. A  good teacher can switch between the Andhra language and the
Tamil language depending on which language his audience speaks:

It is just as a teacher skillful in regional languages commenting on the

meaning of the three Vedas ascertains that they [his students] know the
meaning when spoken to in the Tamil language, then addresses them in
the Tamil language. In the case of another language such as the Andhra
language, [he speaks] in this or that language. In this way the Brahmin
students, having access to a skilled and experienced teacher, learn their
subjects very quickly. The Lord Buddha is like the teacher, the three estab-
lished piṭakas are like the three Vedas when they are to be discussed, being
skilled in conventional and furthest-​sense [language] is like being skilled
in regional languages.92

This is an eminently practical use of language geared to the understanding of

one’s audience. The distinction is not marking different accounts of “ultimate re-
ality” as it may be in other Buddhist traditions and modern scholarship.
The idea that one’s choice of language should be geared to the capacities of
one’s audience is also evident in the commentary on the Kathāvatthu, which
suggests that the Buddha will sometimes need to preach conventional teachings
first until his hearers are prepared to understand the more technical terminology
of paramattha. Indeed, this text says that teachings in the (analytically) furthest

90. Hallisey, “In Defense of Rather Fragile and Local Achievement,” 128–​29. Jayatilleke, Early
Buddhist Theory of Knowledge, 364, and Karunadasa, “The Buddhist Theory of Double Truth,”
33–​34, make the same point about the lack of ranking in the Pali approach to these terms.
91. Hallisey, “In Defense of Rather Fragile and Local Achievement,” 133.
92. Mp i.95; Ps i.137: yathā hi desabhāsākusalo tiṇṇaṃ vedānaṃ atthasaṃvaṇṇanako ācariyo
ye damiḷabhāsāya vutte atthaṃ jānanti, tesaṃ damiḷabhāsāya ācikkhati. Ye andhabhāsādīsu
aññatarāya bhāsāya, tesaṃ tāya tāya bhāsāya. Evaṃ te māṇavakā chekaṃ byattaṃ
ācariyamāgamma khippameva sippaṃ uggaṇhanti. Tattha ācariyo viya buddho bhagavā, tayo vedā
viya kathetabbabhāve ṭhitāni tīṇi piṭakāni, desabhāsākosallamiva sammutiparamatthakosallaṃ.
 Scripture, Commentary, and Exegetical Distinctions 89

sense may be too “severe” a register of language for people at first.93 There is some-
thing about talking with and about people on matters of loving-​kindness and
giving gifts that would seem to make the spare, reductive, and technical language
of “aggregates” baffling and off-​putting, to say the least.
We have seen that something of this distinction about language or termi-
nology corresponds to Buddhaghosa’s descriptions of genre: paramattha is asso-
ciated with Abhidhamma knowledge, and colloquial talk (vohāra) with Suttanta:

Here the Vinaya piṭaka is called teaching in the form of commands be-
cause of being a teaching with a preponderance of commands [spoken
by] the Bhagavan who was worthy of making commands; the Sutta
piṭaka is called a colloquial teaching due to a preponderance of the collo-
quial taught by the Bhagavan who was skillful in colloquial talk; and the
Abhidhamma piṭaka is called teaching in the furthest sense because being
a teaching with a preponderance of furthest sense taught by the Bhagavan
who was skillful in furthest sense. 94

This claim does not restrict any one of the registers of speech exclusively to its
associated genre (and in practice there is much fluidity across genre on this),
but rather suggests that each genre has a preponderance of one of the three. The
notion of colloquial or transactional talk (vohāra) that predominates in the suttas
suggests both the idiomatic speech of regular people and the language of trade,
business, and legal transactions. While the suttas are also full of commands and
more technical teachings in the paramattha register, one of the features of Suttanta
discourse that particularly interests Buddhaghosa is its dialogical back-​and-​forth
nature of Buddha’s engagement with particular interlocutors (as I explore in the
next chapter). Vohāra, perhaps more than sammuti, gets to this “back-​and-​forth”
in both trade and conversation.
For its part, paramattha analysis is a use of language that dismantles ideas and
constructs to show how they can be broken down into smaller parts in order to
discern the conditional relations between them. The Abhidhamma deploys the

93.  KvA 34:  pakatiyā pana paṭhamameva paramatthakathaṃ kathentassa desanā lūkhākārā
hoti. Lūkha can mean “rough, coarse, unpleasant, poor, bad,” indicating here some sort of
unpleasant mode (ākāra) of speech. I follow Law’s translation of “severe” here (The Debates
Commentary, 42).
94.  As 21; Sp i.21; Sv i.19:  ettha hi vinayapiṭakaṃ āṇārahena bhagavatā āṇābāhullato
desitattā āṇādesanā; suttantapiṭakaṃ vohārakusalena bhagavatā vohārabāhullato desitattā
vohāradesanā; abhidhammapiṭakaṃ paramatthakusalena bhagavatā paramatthabāhullato
desitattā paramatthadesanāti vuccati. As 223 claims Abhidhamma is paramattha teachings.
90 Build ing Blo cks for an In t er pretat ive Pro gr a m

language of analytic categories such as dhammas, the smaller descriptions we

begin to arrive at when breaking down experience, as a way to dismantle larger
constructs (like “person”); analytical treatment of dhammas also allows for var-
ious classificatory schemas and groupings that show how causality works in ex-
perience. Such dismantling also can rid of us our fixed notions of composite,
reified, or abstract entities. Buddhaghosa describes paramattha practice of
breaking down composite entities into their parts as a “correct way of seeing”
(yathābhūtadassana), where such practice is embedded in a larger contempla-
tive and analytical exercise of purifying view (diṭṭhivisuddhi).95 For example, he
describes the work an advanced meditator does in trying to see the workings of
name and form (nāmarūpa) instead of a “person” or “being.” He suggests that
we can exercise a style of seeing that discerns that there is no fist in what are dis-
crete but connected fingers arranged a certain way; no lute beyond a body and
strings; no army beyond elephants, horses, and so forth; no tree beyond trunk,
branches, leaves; and so forth.96 Such fine-​grained seeing helps us disengage
from the conceptual reifications that can dominate our thought and action in
problematic ways.
Thus, for Buddhaghosa the paramattha/​sammuti distinction is about different
kinds of language or teachings; it is not an epistemological or ontological distinc-
tion. He makes only one mention of the idea that the terms can refer to two
different types of “truth,” and he quickly qualifies what this means. Embedded
in a larger discussion about teachings (desanā) that we have been discussing,
Buddhaghosa quotes several verses that describe two kinds of truth (sacca). In this
context, Buddhaghosa makes the following claims about the two truths declared
by the Buddha:

The Perfectly Awakened One, the best among speakers, taught that
there are two truths: conventional and furthest sense. A third [truth] is
not found.
An agreed-​upon statement is true because it is the performance
(kāraṇā) of the conventional terms of the world; an expression in the
furthest sense is true because it is the performance (kāraṇā) in which
dhammas are brought about.

95.  Vism XVIII is an exercise of “purifying view” (diṭṭhivisuddhi) (Vism 587 [XVIII.2]),
where Buddhaghosa describes paramattha as a way of seeing only naming and forming [rather
than persons] (paramatthato pana nāmarūpamattam eva atthī ti. Evam passato hi dassanaṃ
yathābhūtadassanaṃ nāma hoti [Vism 593 (XVIII.28)]).
96. Vism 593 (XVIII.28).
 Scripture, Commentary, and Exegetical Distinctions 91

There is no false speech of the Teacher, who is the leader of the

world and skillful in transactional discourse, when he makes use of the
conventional.97

Here we find the idea of these two truths being acted out, performed, or tasked
(all possible with kāraṇā) with these different kinds of language; sammuti truths
are the performance of ideas commonly understood, but paramattha is a use of
language tasked with having brought about (bhūta) the highly reductive terms
that are the phenomena (dhamma) of analysis. These verses do not put these two
truths in a hierarchy or suggest that, from the standpoint of a putative “ultimate
reality,” the Buddha’s conventional teachings are false. The notion that conven-
tional truth could in any way be construed as “false speech” is in fact explicitly
denied, both in the verses and Buddhaghosa’s subsequent discussion.98 For how
could anything the Buddha uttered be regarded as in any way deficient in truth?
One key difference between this formulation and Sanskrit doctrines on
the two truths, is that the Pali tradition uses the term sammuti, rather than the
Sanskrit term saṃvṛti, and they are not cognates (as is sometimes suggested in
the scholarship).99 Sanskrit saṃvṛti can have the sense of covering, obstructing,

97. Ps i.138; Mp i.95 (and ItiA i.82); note that Sv ii.383 and Spk ii.77 have just the first two of
these three verses. Cf. Pind’s translation; see his discussion of these verses in “Pāli Miscellany,”
515–​22. Duve saccāni akkhāsi, sambuddho vadataṃ varo; sammutiṃ paramatthañca, tatiyaṃ
nūpalabbhati. //​Saṅketavacanaṃ saccaṃ, lokasammutikāraṇā; paramatthavacanaṃ saccaṃ,
dhammānaṃ bhūtakāraṇā. //​Tasmā vohārakusalassa, lokanāthassa satthuno; sammutiṃ
voharantassa, musāvādo na jāyatī ti. It is unclear where these verses originally came from,
though they are cited in several places in the commentarial material, but Law suggests that
they correspond to Nāgārjuna’s statement on two truths (Buddhaghosa, 85).
98. KvA 34 also denies that there can be anything false in the conventional discourse taught
by the Buddha (te sammutikathaṃ kathentāpi saccameva sabhāvameva amusāva kathenti).
The picture I  am describing here for Buddhaghosa appears to be different from that of the
modern Theravada thinker Ledi Sayadaw, who argues that from the standpoint of ultimate
truth, conventional truth is “erroneous” (“Some Points in Buddhist Doctrine,” 129), some-
thing Buddhaghosa never says. Karunadasa (“The Buddhist Theory of Double Truth,” 35–​
36) also notes this discrepancy. But my reading of the Pali tradition also departs from that
of Karunadasa, who says that for the Pali tradition and the Sarvāstivādins alike it was taken
“for granted that what cannot be further analysed, what is irreducible, is real (paramattha,
dravyasat) and that which is further analysable is nominal (sammuti, prajñaptisat)” (27). I do
not find evidence that for Buddhaghosa (at least) paramattha means anything like dravyasat,
substantially real (a term he never uses). As the Pali tradition developed, the distinction about
paramattha and sammuti may have come to have epistemological and metaphysical signifi-
cance (such as in the Abhidhammatthasaṅgaha in the ontological reading of Bhikkhu Bodhi, A
Comprehensive Manual of Abhidhamma, 25–​27). Scholars have not always clearly distinguished
the commentarial approach of Buddhaghosa and Dhammapāla from this later tradition.
99. As for example, Buswell and Lopez, The Princeton Dictionary of Buddhism, 762. The entry
on saṃvṛti: identifies it with Pali sammuti and says, “in Sanskrit the term carries a connotation
92 Build ing Blo cks for an In t er pretat ive Pro gr a m

or concealing, which indicates in the hands of certain Mahāyāna philosophers

a kind of worldly understanding associated with ordinary people that is, for
some thinkers, considered ultimately unreal in contrast to ultimate truth. But
the Pali term sammuti has no connotations of concealment or dissimulation,
and refers simply to conventional terminology used in ordinary discourse. While
Sanskrit paramārthasatya is variously conceived in different systems, the Pali
term paramatthasacca refers to a technical description of the dhammas, the phe-
nomena that, in an analytically further way, can be said to describe experience. It
is the knowledge one arrives at through the kinds of contemplative or definitional
analysis usually associated with Abhidhamma practices.
The Abhidhamma text, the Kathāvatthu, and its commentary have a lengthy
discussion of how “person” is understood that is relevant here. The discus-
sion is embedded in the context of a philosophical debate with another school
of Buddhists called the Puggalavādins. This context is highly relevant for
interpreting its discussion since the Puggalavādins apparently sought to argue
for a much more ontologically substantial interpretation of the “person” than
the authors of the Kathāvatthu and its commentary were willing to abide. This
appears to cast the discussion in rather ontological terms, but one can also read
the Vibhajjavādin treatment of the debate in the Kathāvatthu without assuming
that it concerns what exists metaphysically. The debate is framed in terms of how
one arrives at the notion of “person”: “does one arrive at [the notion of ] ‘person’
by way of a true sense, a furthest sense?”100 The commentary explains these terms
as follows:

Therein, “person” means self, being, soul. “One arrives at” means one
arrives, having approached with the understanding, the meaning is “one
knows.” Here “by way of a true sense, a furthest sense”: “true sense” means
real, not to be understood as something unreal like a delusion, mirage, et
cetera; “furthest sense” means highest, not to be understood on the basis
of hearsay, et cetera.101

of ‘covering, concealing,’ implying that the independent reality apparently possessed by ordi-
nary phenomena may seem vivid and convincing, but is in fact ultimately illusory and unreal.”
For clarification on the derivation of sammuti from √sam-​man (not √sam-​var or √sam-​vṛ) see
Collins, A Pali Grammar for Students, 12.
100. Kv 1: Puggalo upalabbhati saccikaṭṭhaparamatthenāti.
101. KvA 9: Tattha puggaloti attā, satto jīvo. Upalabbhatīti paññāya upagantvā labbhati, ñāyatīti
attho. Saccikaṭṭhaparamatthenāti ettha saccikaṭṭhoti māyāmarīciādayo viya abhūtākārena
aggahetabbo bhūtaṭṭho. Paramatthoti anussavādivasena aggahetabbo uttamattho.
 Scripture, Commentary, and Exegetical Distinctions 93

These definitions concern how one knows the idea of “person,” and the Vibhajjavādin
position is to assert that “person” is not known by the reductive practices associated
with analysis. The text contrasts this to the way that the categories of Abhidhamma
analysis are known (which the text lists here as fifty-​seven dhammas, as they are
classified among the aggregates, sensory bases, elements, faculties, et cetera).102
Against the Puggalavādins, the commentary holds that one arrives at the dhammas
quite differently than one does for the idea of “person,” and can know them truth-
fully and further in the way one cannot know “person.” As an Abhidhammic text the
Kathāvatthu describes an analytical method that is meant to demonstrate that there
is a difference between the categories this method produces—​the dhammas—​and
conventional terms like “person,” which dissolve like mirages when so analyzed. In
the context of this discussion, against people who would argue that “person” is like
“dhamma,” the category of “person” is not usefully and skillfully analytic.
And yet, despite this discussion, we must note that the rejection of the useful-
ness of “person” as a category of analysis is contextual because in other contexts,
even within the Abhidhamma, “person” is a useful category. We have an en-
tire Abhidhamma text called Describing Persons (Puggalapaññatti), which uses
“person” as its category of analysis. Persons are hardly off-​limits in Abhidhamma
discourse. This text begins in ways we would expect by analyzing the category of
“person” into the various reductive schemas that constitute it, such as aggregates,
elements, bases, truths, the faculties, and so on. But then the bulk of the text
goes on to describe and redescribe persons as relatively unproblematic “wholes,”
according to moral status, character, and action, social class, temperament, and
spiritual capacity. One way to understand this is to return again to the pragmatic
and contextual use of language that Buddhaghosa notes, where perhaps no term
or idea is “inherently” sammuti or paramattha; the highly context-​sensitive na-
ture of the distinction would also militate against the ontological reading. That is
to say, when arguing against Puggalavādins, “puggala” needs to be dismantled as
analytically problematic, at least in comparison to the use of “dhammas,” which
can make it dissolve. But puggala can elsewhere be a useful category, helpfully
described in diverse ways by the classificatory schemes of the Puggalapaññatti.
To sum up, as Karunadasa puts it, the distinction between paramattha and
sammuti “refers not to two species of truth as such, but to two modes of explaining
the truth”; it pertains to “method, and not to content.”103 A  crucial corollary
to this—​and one we explore at length in c­ hapter 4, so I only signal it here—​is

102. KvA 9–​10: Ubhayenāpi yo parato ‘‘puggalo upalabbhati saccikaṭṭhaparamatthena, rūpañca

upalabbhatī’’tiādinā khandhāyatanadhātuindriyavasena sattapaññāsavidho dhammappabhedo
dassito.
103. Karunadasa, “The Buddhist Theory of Double Truth,” 36–​37.
94 Build ing Blo cks for an In t er pretat ive Pro gr a m

that Buddhaghosa does not make the leap into ontology that would view the
resulting paramattha listings of dhammas as a final, irreducible list of ontological
reals, as has been suggested by some scholars, including (oddly, given his remarks
earlier) Karunadasa.104 The paramattha method of analysis, even of dhammas, is
an endless activity or practice rather than an attempt to arrive at a final and def-
inite listing of what is ultimately, in an ontological sense, real. Paramattha does
not mean ontologically real in Buddhaghosa’s work, and he does not see himself
doing the metaphysical argumentation that would produce such claims. But this
discussion awaits ­chapter  4. For now, it is enough to see that paramattha and
vohāra/​sammuti refer to uses of language, ways of seeing, or analytical exercises
efficacious for certain contexts within the larger teleology he advances.

Definitive and Interpretable Statements

Another distinction, and one sometimes related to the distinction between
furthest-​sense and conventional language, is the distinction between definitive
statements and those requiring further interpretation. The Aṅguttara gives what
appears to be clear instruction about how the Buddha’s words are to be interpreted
by advising against the way that people can misrepresent the Tathāgata. The
Buddha sometimes taught explicitly, in a manner in which the meaning is “de-
finitive” (literally, “already guided,” nītattham), and other times he taught implic-
itly, in a manner in which the meaning needs to be interpreted (literally, “to be
guided” neyyatham).

Monks, there are two ways of slandering the Tathāgatha. What are the
two? One explains a sutta whose meaning is interpretable as a sutta whose
meaning is definitive, and one explains a sutta whose meaning is definitive
as a sutta whose meaning is interpretable.105

104.  Karunadasa, “The Buddhist Theory of Double Truth,” 36–​37. Throughout his paper
Karunadasa refers to the dhammas arrived at through the Abhidhamma method as “real
existents,” an ontological leap neither required nor carried out by Buddhaghosa at least.
Examples of other scholars who read the Pali Abhidhamma tradition as giving an ontological
description of ultimate reals include Ronkin, who argues that at the aṭṭhakathā level, dhammas
were reckoned as “the ultimate independently existing constituents of experience” (Early
Buddhist Metaphysics, 122). Bhikkhu Bodhi calls dhammas in the Abhidhamma texts “onto-
logical actualities,” stepping away, I believe, from Nyanaponika’s phenomenological approach
on which he comments (“Editor’s Introduction” to Nyanaponika Thera’s Abhidhamma Studies,
xviii).
105.  A  i.60:  dveme, bhikkhave, tathāgataṃ abbhācikkhanti. Katame dve? Yo ca neyyatthaṃ
suttantaṃ nītattho suttantoti dīpeti, yo ca nītatthaṃ suttantaṃ neyyattho suttantoti dīpeti. Netti
21 explicitly demonstrates asking which category a text belongs to.
 Scripture, Commentary, and Exegetical Distinctions 95

By taking what is stated plainly and reading inferred meaning into it, one distorts;
by not doing the work of inference of something that needs to be drawn out, one
also distorts. So the first task is to figure out which kind of statement is which.
But how are we to know what kind of statement or utterance it is that one is
explaining? As is generally the case, Buddhaghosa does not give general herme-
neutical criteria for distinguishing statements that need to be interpreted from
those stated definitively; but he does suggest in the following example that it is
out of immaturity or foolishness (bālatā) that one would confuse them.
His commentary on this passage provides his most systematic treatment
of this distinction, and one that is also relevant to our understanding of the
sammuti/​paramattha distinction.

Suttas that say such things as “Monks, there is one person,” “monks,
there are two persons,” “monks, there are three persons,” “monks, there
are four persons” need to be interpreted. Although “monks, there is one
person” was said by the Perfectly Awakened One, ultimately there is no
name for “person” in the furthest sense and so the meaning of it should
be interpreted. But [a fool], because of his own immaturity, explains this
sutta as definitive. For the Tathāgatha would not say “monks, there is
one person” when there is no person from the [standpoint of ] furthest
sense. But since this was said by him [i.e., the Tathāgatha], he [the foolish
person] explains a sutta that is to be interpreted as a sutta that is defini-
tive, understanding it [to say] that there is a person from the [standpoint
of ] furthest sense. The definitive meaning is the stated meaning, such as
“impermanent, suffering, nonself.” For here the meaning simply is imper-
manent, suffering, nonself. But [a fool], because of his own immaturity
explains a definitive sutta as an interpretable sutta, saying “this sutta is in-
terpretable, I need to bring out its meaning.”106

106. Mp ii.118: Tattha ‘‘ekapuggalo, bhikkhave, dveme, bhikkhave, puggalā, tayome, bhikkhave,

puggalā, cattārome, bhikkhave, puggalā’’ti evarūpo suttanto neyyattho nāma. Ettha hi kiñcāpi
sammāsambuddhena ‘‘ekapuggalo, bhikkhave’’tiādi vuttaṃ, paramatthato pana puggalo nāma
natthīti evamassa attho netabbova hoti. Ayaṃ pana attano bālatāya nītattho ayaṃ suttantoti
dīpeti. Paramatthato hi puggale asati na tathāgato ‘‘ekapuggalo, bhikkhave’’tiādīni vadeyya.
Yasmā pana tena vuttaṃ, tasmā paramatthato atthi puggaloti gaṇhanto taṃ neyyatthaṃ
suttantaṃ nītattho suttantoti dīpeti. Nītatthanti aniccaṃ dukkhaṃ anattāti evaṃ kathitatthaṃ.
Ettha hi aniccameva dukkhameva anattāyevāti attho. Ayaṃ pana attano bālatāya ‘‘neyyattho
ayaṃ suttanto, atthamassa āharissāmī’’ti ‘‘niccaṃ nāma atthi, sukhaṃ nāma atthi, attā nāma
atthī’’ti gaṇhanto nītatthaṃ suttantaṃ neyyattho suttantoti dīpeti nāma. Bhikkhu Bodhi notes
how this was an important issue debated among the early Indian schools and carried forward
in Mahāyāna contexts (trans. The Numerical Discourses of the Buddha, 1624, n.243). Lamotte
describes some of these debates (“Assessment of Textual Interpretation in Buddhism,” 16–​19).
96 Build ing Blo cks for an In t er pretat ive Pro gr a m

In this example, the interpretable/​definitive distinction is mapped onto the

sammuti/​paramattha distinction. What this mapping does here is to use the
well-​known distinction about the word “person”—​that further description does
not use this term and replaces it with analytically more precise categories (as
explained in the previous section)—​to make this distinction clearer.
While we might be tempted to take this mapping of the two distinctions as
a broader principle and thus align definitive statements with furthest-​sense lan-
guage, and interpretable statements with conventional language, Buddhaghosa
does not make any such larger claim, and we should avoid taking examples as
general principles. He nowhere claims that all paramattha terms are interchange-
able with nītattha, nor are all sammuti terms neyyattha.107 Nor does he say that all
Abhidhamma discourse is definitive and that all Sutta discourse is interpretable,
as other Buddhist interpreters did.108 Lamotte is correct that there can be reti-
cence among Buddhist authors on how to distinguish definitive and interpretable
texts, and that “we can only examine their method of procedure in each particular
case.”109
Indeed, as I have indicated elsewhere, Buddhaghosa is little given to stating
general principles of hermeneutics; instead, we have to watch his practice. His
approach for this distinction (and others) tends to be ad hoc, deploying the
know-​how of practice and experience rather than invoking a formal and gener-
alizable principle. For example, the Atthasālinī combats an opponent with this
device by saying that the opponent is confused about both legalities and genre.
This particular issue concerns a question over whether a monk who fails to make
a confession of an infraction at the Uposatha is committing a violation of speech
(as the Vinaya would have it), or whether the violation is occurring in the “mind
door,” which is a technical description of how actions can occur in the mind (the
general context here is the subject of karma in the Abhidhamma commentary).
The opponent has presumably taken the monk’s silence in not confessing the in-
fraction as a mental action occurring at the “mind door”—​that is, too literally

107. Pind agrees: “in spite of this interesting connection, Buddhaghosa does not explicitly cor-
relate neyyattha with sammuti nor paramattha with nītattha” and they do not “belong in the
same context” (“Pāli Miscellany,” 522).
108. Vasubandhu made this distinction along these lines (Gold, Paving the Great Way, 117).
109. Lamotte, “Assessment of Textual Interpretation in Buddhism,” 17. In contrast, Bhikkhu
Bodhi seems more willing to generalize on the basis of this passage: “the Pāli commentators
decide this issue [of which discourses are which] on the basis of the Abhidhamma distinction
between ultimate realities and conventional realities” (Bodhi, trans., The Numerical Discourses
of the Buddha, 1624 n.243). (His putting it thus is also more ontological than I would frame
the paramattha/​sammuti distinction, as discussed previously.) Karundasa also sees a general
correspondence between the two distinctions (“The Buddhist Theory of Double Truth,” 36).
 Scripture, Commentary, and Exegetical Distinctions 97

or definitively, in Buddhaghosa’s view. Buddhaghosa scolds him that he has for-

gotten that the Vinaya counts only physical and verbal actions—​not mental
actions—​as culpable.110 The opposing view has taken something at face value that
needs to be interpreted. Here the matter has nothing to do with the furthest-​
sense/​conventional distinction, but is rather a question of legal discourse. In an-
other example from the legal context of the Vinaya, we have an instance where
deciding that a matter is interpretable becomes the occasion to try to discern the
intent (adhippāya) of a passage.111
The definitive/​interpretable device is thus handy for Buddhaghosa when
disputing opponents, where one should ask them pointedly about whether they
are taking a passage definitively or in a manner to be interpreted, as we find him
doing in several instances.112 Indeed, the toolkit of every exegete needs a device
for accusing those holding a divergent view of being overly literal and not doing
enough work to draw out a less obvious reading, or alternatively, of not taking
claims at face value that should be so taken.

Contextual, Qualified, and Figurative versus

Categorical and Abstract Ways of Teaching
Our final distinction about discourse is perhaps the most consequential for my
analysis in the rest of the book: the contrast between pariyāya and nippariyāya
teachings. Buddhaghosa defines pariyāya as meaning three things:  a way of
teaching (as presented), the activity or performance of a teaching, or a “turn” as
when something is said in turn.113 Used on its own, pariyāya often just refers to
various “ways of teaching” that the Buddha deployed pragmatically for whatever

110. As 92–​93.
111. Sp iv.847.
112. Ps ii.363; iii.14.
113.  Sv i.36:  pariyāya-​saddo tāva vāradesanākāraṇesu vattati. Cf. Ps i.17. Pariyāya when used
on its own can in fact mean a range of different things, including a method of teaching or
exposition (sometimes Buddhaghosa will offer “another way” (aparopi pariyāyo) to interpret
something using this term, Spk i.99), or a figurative use of speech (as for example, As 63; Pj
I 15). It is notable that the Jains used the term paryāya in a technical way to describe modal
knowledge: the viewpoint of modes (paryāyarthika-​naya) in contrast to the viewpoint of sub-
stance (dravyārthika-​naya). The first “considers the modifications and conditions of an object”
and “indicates the infinite standpoints possible when Reality is analyzed from the modes it
possesses” (Grimes, A Concise Dictionary of Indian Philosophy, 230). This modal logic works
well for many instances of Buddhaghosa’s use of the term. The second, a standpoint based on
an object’s substance, is something Buddhaghosa would never say: dravya is not part of his
thinking.
98 Build ing Blo cks for an In t er pretat ive Pro gr a m

purpose was at hand, and where the emphasis is on the skillful activity or per-
formance of a pedagogical approach.114 Our interest here is when pariyāya is
contrasted to nippariyāya as two different ways an assertion can be made or taken;
pairing and contrasting the two was a commentarial development based on a
single canonical occurrence of the pairing of these terms in the Aṅguttara. When
contrasted in this way, when the Buddha spoke in a manner that is pariyāya, he
was speaking contextually, modally, figuratively, or in a qualified sense; when he
spoke in a manner that is nippariyāya, he spoke categorically, abstractly, literally,
or unqualifiedly. Because the contrast between these terms developed variously
and was used to different purposes depending on context, it is helpful to resist
thinking about this distinction as concerning a single dimension of language or
about either term as managed easily by a single translation.115 As often, contextual
examples can show us what this meant.
The Aṅguttara passage in which this distinction makes its single appearance
in the canon is an account of a contemplative practice. The Buddha describes
progressive meditative freedom from “confinement” as finding “space” from the
usual trappings of sensory experience. The meditative progression finds space
first “in a qualified sense” (pariyāya) and then “categorically” (nippariyāya).116
Buddhaghosa says that pariyāya means “by one instance,” (and so “in a qualified
sense” seems a good way to translate it here), while nippariyāya (“categorically,”
“unrestrictedly”) means “not by way of a single instance but in every way and
entirely the taints are destroyed, all confinement is eliminated, and open space
is found.”117 This suggests that in this case the two are related as the particular
is to the general, as Charles Hallisey observes.118 It can also refer to the way that
nippariyāya knowledge is abstracted from the thicker contextual particularities of
pariyāya usage that can constrain sense.

114.  We see this usage of pariyāya as the activity or performance (kāraṇa) of a teaching, as
Buddhaghosa glosses it (Spk iii.97), in the Bahuvedanīya Sutta, which I discuss in c­ hapter 4.
115. Sasaki settles perhaps too quickly on a single point of contrast in the Pali sources, where
pariyāya refers to the “reason” (Grund) something comes to be and nippariyāya refers to the
“fact” (Tatsache) of its existence or the “result” (Folge) of what prompted it (“Pariyāya und
Nippariyāya,” 50). As the examples I describe here show, the terms get at varying vectors of
meaning.
116. A iv.450–​51.
117. Mp iv.205–​206: pariyāyenāti ekena kāraṇena . . . nippariyāyenāti na ekena kāraṇena, atha
kho āsavakkhayo nāma sabbasambādhānaṃ pahīnattā sabbena sabbaṃ okāsādhigamo nāmāti.
118. Hallisey, “In Defense of Rather Fragile and Local Achievement: Reflections on the Work
of Gurulugomi,” 132.
 Scripture, Commentary, and Exegetical Distinctions 99

This distinction between pariyāya and nippariyāya can also be used to distin-
guish two different soteriological ambitions. For example, “maintaining a life of
purity” can be interpreted both relatively and categorically: in a qualified sense it
refers to following the five precepts in order to achieve a good rebirth, while cate-
gorically, it refers to the perfectly “stainless” purity of the path of nibbāna, which
cannot be qualified in any way.119 Statements about nibbāna tend to be deemed
categorical and unqualified.
Buddhaghosa uses the distinction elsewhere to refer to the style of teachings
given in the context of a conversation between the wise monk Assaji (whom we
met before in his terse teachings to Sāriputta) and an arrogant and clever Jain
called Saccaka, known to be a formidable debater. Assaji, Buddhaghosa says,
used abstract or categorical discourse (nippariyāyakathā) rather than qualified
discourse (pariyāyakathā) when introducing the Buddha’s teaching to him. He
got straight to the point in teaching that “form is impermanent,” using very di-
rect language. This interlocutor, who was predisposed to refuting the teachings
and finding fault, needed language that was categorical rather than qualified or
contextualized; he was “not established,” that is, he had no shared ground on
which a qualified or figurative teaching would work.120
In other cases, the distinction is more like that between the “contextual”
and the “categorical.” The Visuddhimagga notes that words are often contextual
(pariyāya): some, like “long” and “short,” apply only relative to other things.121
We can say that the material experience of one kind of deities is inferior to that
of another, and thus the idea of “inferior” is a concept that here only makes sense
relative to something else.122 But, unlike long and short, we can also speak of “in-
ferior” more categorically (nippariyāya) as when we say that a certain rebirth is
inferior because it is the result of bad (akusala) karma. The categorical use of
language is often definitional:  because bad karma by definition produces infe-
rior, that is, bad, rebirths, there is nothing contextual about using “inferior” in
this way. Note that there is nothing essentially either contextual or categorical in
the word “inferior”—​the word can go either way depending on the context in

119. Mp ii.9: suddhaṃ attānaṃ pariharatīti ettha duvidhā suddhi—​pariyāyato ca nippariyāyato

ca. Saraṇagamanena hi pariyāyena suddhaṃ attānaṃ pariharati nāma.  .  .  . Arahattaphale
patiṭṭhito. . . . nippariyāyeneva suddhaṃ nimmalaṃ attānaṃ pariharati paṭijaggatīti veditabbo.
120.  Ps ii.271:  dosaṃ āropeyya, tasmā paravādissa pariyāyakathaṃ kātuṃ na vaṭṭati. Yathā
esa appatiṭṭho hoti, evamassa nippariyāyakathaṃ kathessāmīti cintetvā, ‘‘rūpaṃ, bhikkhave,
anicca’’nti imaṃ aniccānattavaseneva kathaṃ katheti. Note that Buddhaghosa does not pull in
the language of nītattha (definitive) here to draw this contrast, as we might imagine he might.
121. As 415.
122. Vism 473.
100 Build ing Blo cks for an In t er pretat ive Pro gr a m

which it is used. There thus remains some important sense in which some context
matters, even with categorical usage, to discern which is which.123
In classifying the first Noble Truth of suffering, an Abhidhamma commentarial
discussion describes pariyāya suffering and nippariyāya suffering as listed in a ma-
trix along with other classifications of suffering.124 Categorical suffering is the
bodily and mental pain that is suffering, literally “the suffering that is suffering.”
But other types of suffering are more qualified or contextual such as the suffering
of birth, aging, death, and so on, as formulated in the classic description of the
four Noble Truths.125 The sufferings attending birth and old age are not always
literally and unqualifiedly painful in the way that pain is by its very nature, pain.
We might extrapolate that this means that one must look at the birth of a baby
in a very specific way to see that it is suffering, since in other respects it might be
an occasion for joy. To see birth as suffering one may consider how even a healthy
delivery is painful for the mother or the infant, or one might ponder the sad truth
that whoever is born will someday die. Old age, too, can have it merits and may
not always be suffering in an unqualified sense, though in some senses it can in-
volve suffering. This qualified or contextual sense is quite unlike the definitional
sense of suffering qua suffering.
A further way of understanding the contrast is in terms of “figurative” and
“literal” claims. This can be seen in the definition of “rebirth”:  rebirth can be
stated literally as the first manifestation of the aggregates, since that is how re-
birth is defined. But we can also refer to it more figuratively as exiting the mother’s
womb.126 Another example can be found in an Abhidhamma commentary on a
very technical issue: moral precepts (sīla) are literally the morality of restraint and
the morality of nontransgression; only figuratively can we speak of sīla as the mo-
rality of intention (cetanā) and mental phenomena (cetasika).127 This means that,
strictly speaking, moral precepts are restraining and refraining from wrongdoing,

123.  Hallisey makes this point as well (Hallisey, “In Defense of Rather Fragile and Local
Achievement,” 132).
124. Vism 499 and Vibh-​a 93: pariyāyadukkhaṃ nippariyāyadukkhanti.
125.  Vism 499 and Vibh-​ a 94:  ṭhapetvā dukkhadukkhaṃ sesaṃ dukkhasaccavibhaṅge
āgataṃ jātiādi sabbampi tassa tassa dukkhassa vatthubhāvato ‘pariyāyadukkhaṃ’ nāma.
Dukkhadukkhaṃ ‘nippariyāyadukkhaṃ’ nāma. The classic formulation of the first noble truth
is the passage on which this commentary is discussing as given in the Vibhaṅga (Vibh 99).
126. Vism 499.
127. Vibh-​a : 331: Ettha ca saṃvarasīlaṃ avītikkamasīlanti idameva nippariyāyato sīlaṃ; cetanā
sīlaṃ cetasikaṃ sīlanti pariyāyato sīlanti veditabbaṃ. This issue of sīla as the absence (through
restraint and lack of performing) of wrong actions (rather than a matter of positive intention)
is discussed in my book, Heim, The Forerunner of All Things, ch. 1.
 Scripture, Commentary, and Exegetical Distinctions 101

even though we might speak more informally as their being a matter of positive
intention and mental states.
For me, one of the most interesting alignments Buddhaghosa makes about
this distinction occurs when he associates pariyāya discourse with Suttanta and
nippariyāya with Abhidhamma. In a small reference to them made in passing
in the Atthasālinī, he indicates that teachings may have less or more context in
which they are delivered: pariyāya teachings that are given in the Suttanta can
include background knowledge of the people involved (such as their moral
habits), while nippariyāya teachings are more abstract and given without refer-
ence to context or persons. In the midst of a very spare “Abhidhammic” account
of the eightfold path (the details here are not pertinent for the point), he offers
a different possibility for describing it, which he describes as a “Suttanta” style of
thinking about it:

For this pariyāya teaching is a Suttanta teaching, where he [the

Buddha] says: “his conduct of bodily action and verbal action were for-
merly highly purified.” But this [the discussion at hand] is a nippariyāya
teaching.128

This small comment suggests that contextual facts about a given person’s previous
moral action (the insertion is from a Majjhima passage describing an actual set
of practices carried out by an ideal meditator) are a matter of pariyāya teaching
(associated here with Suttanta), and that this is contrasted with the rather
starker (we might suggest) categorical renderings of the Dhamma, in which such
interjections about particular persons are irrelevant (of the sort we get in the
Abhidhamma).
Buddhaghosa also associates these terms with the two genres in a rather tech-
nical discussion of different experiences that can be had in the first jhāna, an ad-
vanced contemplative state. The Dhammasaṅgaṇī distinguishes three experiences
possible in this jhāna:  “emptiness” (suññatā), “signless” (animitta), and “not
desired” (appaṇihita). The details of these are not as important for our purposes
as noticing what Buddhaghosa says about how they get their names:

By the Suttanta method something gets its name from its associated qual-
ities and by its object. This is a contextual teaching (pariyāyadesanā). But

128.  As 154:  Ayañhi suttantikadesanā nāma pariyāyadesanā. Tenāha—​‘‘pubbeva kho panassa

kāyakammaṃ vacīkammaṃ ājīvo suparisuddho hotī’’ti. Ayaṃ pana nippariyāyadesanā. The in-
terjection is in M iii.289.
102 Build ing Blo cks for an In t er pretat ive Pro gr a m

Abhidhamma discourse is a categorical teaching (nippariyāyadesanā).

And so in this case something gets its name not by its associated qualities
or object, but just by its arrival.129

This suggests that Abhidhamma’s abstract or categorical teachings involve names

for experiences that arrive unmediated by related qualities or by their objects.
Instead, such experiences get their names from the fact of their occurrence: “for
it is just its arrival that is foremost; and [that arrival] is twofold: the arrival of
vipassanā and the arrival of the Path.”130 This suggests a view of language in the
Abhidhamma/​nippariyāya register whereby its terminology involves an imme-
diate and direct naming of phenomena that are known in contemplative expe-
rience. This is contrasted to a naming process that names things via features and
objects other than the experience itself. Whatever we might think about the
claim that names for things can occur directly and unmediated by other ideas
(an assertion likely based on theories the tradition held about the primacy, im-
mediacy, and universality of Pali we discussed earlier, as well as claims about the
kinds of knowledge contemplative experience delivers), it is here instructive for
interpreting how Abhidhamma discourse was understood and contrasted to
Sutta discourse.
To begin to sum up what we have learned about this distinction: it is impor-
tant to note that nowhere do the commentators rank hierarchically these two
styles of teaching: nippariyāya is not better or truer than pariyāya.131 Nor do they
map exactly onto other distinctions about language made by the commentators
such as conventional (sammuti) and furthest-​sense (paramattha) teachings or de-
finitive (nītattha) and interpretable (neyyattha) teachings, although there might
sometimes be alignments of these distinctions. They are also not identified ex-
clusively with one genre; while pariyāya teachings are prevalent in the suttas,
nippariyāya teachings are also given in them (as in the case of Assaji, whose

129.  As 222 (on Dhs 99):  Tattha suttantikapariyāyena saguṇatopi ārammaṇatopi nāmaṃ
labhati. Pariyāyadesanā hesā. Abhidhammakathā pana nippariyāyadesanā. Tasmā idha
saguṇato vā ārammaṇato vā nāmaṃ na labhati, āgamanatova labhati.
130.  As 222:  Āgamanameva hi dhuraṃ. Taṃ duvidhaṃ hoti—​vipassanāgamanaṃ
maggāgamananti. Note that Pe Maung Tin and Rhys Davids say that some manuscripts have
dhuvaṃ (“certain”) in place of dhuraṃ (foremost or principal) (The Expositor, 300 n. 2.).
131.  Hallisey has also argued for the significance of the lack of hierarchical ranking in this
and other distinctions about language made in the Pali sources (including also the sammuti/​
paramattha and the neyyattha/​nītattha distinctions). He discusses the various ways that each
distinction was conceived, and notes that as much as the terms in each pair are contrasted, they
are also alternatives within “a larger framework” or a “single field in which one element consist-
ently gives value to the other” (“In Defense of Rather Fragile and Local Achievement,” 132–​34).
 Scripture, Commentary, and Exegetical Distinctions 103

nippariyāya teachings occur in the Majjhima). Nor do any of these distinctions

indicate a fixed or essential quality of language or words; that is, a word in one
context might be deemed pariyāya but in another it can have the categorical force
of nippariyāya (as we saw with “inferior”). And, as we can see from this latter
point, context never fully goes away: it depends on what is being considered in-
ferior. Context is also often decisive for whether a teaching is one or the other, as
when the audience to whom that teaching is given is relevant (Assaji’s decision
to deliver nippariyāya teachings to Saccaka was based on what he gleaned about
Saccaka).
With these qualifications well taken, I see this distinction between contextual
and categorical as one that Buddhaghosa deploys widely and productively in his
reading of the piṭaka genres. The pariyāya teachings associated with the Suttanta
method are teachings in which the content is mediated through or inflected by
some salient feature of the context: they are teachings that are configured by and
pertain to the circumstances, or what Buddhaghosa calls the “single instance,” to
which they speak. In the commentators’ treatments of the suttas discussed in the
next chapter we see a principled focus on narrative context as a singular set of
circumstances to which the Buddha’s teaching spoke and through which it is to
be interpreted. This focus is an enactment of their insistence that some teachings
can be indexed to context and circumstance, and that the wisdom we can glean
from the teachings relies, at least in part, on exploring this relation. In contrast,
the analyses of the Abhidhamma are deemed to be abstracted, categorical, un-
qualified and unqualifiable, and not mediated by particular circumstances. This
abstract mode of knowledge requires its own style of interpretation, as we shall
see in ­chapter 4.

Conclusions
Throughout this chapter I have sought to draw a distinction between setting for-
ward general principles or rules of interpretation, on the one hand, and practicing
methods through commentarial examples, on the other. Both can offer guidelines,
but they do so in different ways. My reading of Buddhaghosa is that he treats the
many distinctions described in this chapter not as general principles that can be
articulated in advance and then applied to particular cases, but more as methods
or rules of thumb that emerge through the practice of exposition. In fact, he often
(though not always) resists the tendency to step out of the thickets of practice to
assert general principles, a style of teaching similar to Vasubandhu’s hermeneutic
manual, as noted previously. The pedagogy is cumulative rather than declarative,
particularist rather than universal (though we will want to look for patterns), and
methodical rather than summative.
104 Build ing Blo cks for an In t er pretat ive Pro gr a m

Another way to discern his style of work is to mention the ancient Greek idea
of mētis, the practical knowledge that comes from doing the activity, usually by
working as an apprentice with a master. Mētis is learning by doing, and while it
involves rules of thumb (of the sort we have considered in this chapter), “knowing
how and when to apply the rules of thumb in a concrete situation is the essence of
mētis. The subtleties of application are important precisely because mētis is most
valuable in settings that are mutable, indeterminate (some facts are unknown)
and particular.”132 Mētis is local knowledge that can adapt to constantly changing
conditions; it is not evident how or if it might be abstracted and applied to other
contexts, or how far it can travel. Mētis can be contrasted with the Greek notion
of techne, the formalized, universal, hard-​and-​fast rules and propositions imper-
vious to context that constitute technical knowledge; techne is “characteristic,
above all, of self-​contained systems of reasoning in which the findings may be
logically derived from the initial assumptions.”133 Techne would be the rules of
navigation deployed on the open sea by anyone who had studied them; mētis is
the know-​how of the experienced sea captain who has piloted his ship into a par-
ticular port countless times and knows its local idiosyncrasies and how to adjust
to changing weather and conditions there. Buddhaghosa gives us the guidance of
mētis, which we should not mistake for techne.
And so his methods do not aim at the promulgation of a decisive set of rules
formulated to guide interpretative practice in all cases, such as those we find
described by Lamotte in the Mahāyāna Catuḥpratisaraṇasūtra. This text has sev-
eral straightforward principles, such as prioritizing meaning over phrasing, and
the literal meaning over the interpreted meaning.134 Such principles can be very
useful in a commentator’s hands because they provide in a consistent and system-
atic way rules for coping with apparent inconsistencies in scriptural texts that must
have posed immediate and explicit dilemmas for many Mahāyāna interpreters.
Buddhaghosa’s approach, as a self-​described Vibhajjavādin, is different; he does
not allow for the possibility of even apparent contradictions in the scriptures
themselves (for they are allegedly perfect in meaning and phrasing), his tradi-
tion is represented as coming to him whole and unbroken, and he does not see
the role of the commentator as a matter of declaring and arguing his own views.
His role, as he himself tells us, is rather to make distinctions and point out a mul-
tiplicity of methods for understanding as he goes along. His pedagogy is no less

132. Scott, Seeing Like a State, 316. Scott is getting this rehabilitation of the idea of mētis from
Detienne and Vernant, Cunning Intelligence in Greek Culture and Society.  I am grateful to
Charles Hallisey for pointing me to this distinction. 
133. Scott, Seeing Like a State, 320.
134. Lamotte, “Assessment of Textual Interpretation in Buddhism.”
 Scripture, Commentary, and Exegetical Distinctions 105

systematic for that,135 for he aims to be ruthlessly consistent and unremitting in

his attention to detailed exposition through the very methods we have started
to outline. But his is a style, a mētis, of intellectual practice that has attracted less
scholarly notice than more overtly and explicitly formal efforts of the Sanskrit
Buddhist traditions, and it has not always been recognized as the distinctive con-
tribution to intellectual history that it is.
Moreover, as we have suggested, the methods described here are for the most
part understood to be derived from the unique features of buddhavacana as these
were intimated in canon and developed in commentary. Buddhaghosa saw the
Buddha as practicing multiple methods of teaching (he “was skilled in many
methods” as stated in the epigraph of this chapter) and sought to emulate this. As
George Bond has argued, for the Pali tradition, “the only way to understand the
unique dhamma is to interpret it in terms of itself: in accord with its own logic
and in terms of its own fundamental ideas.” “In this way,” Bond continues, “the
Buddha, in effect, becomes the interpreter of his own teachings.”136 Buddhavacana
itself tells us how to interpret it. The nimble flexibility identified in the Buddha’s
speech whereby he can speak with meaning and with phrasing, categorically or
contextually (and so on), indicate the ways that the Buddha was highly attuned
to the different registers of language required to communicate different subjects
to diverse audiences. The Buddha’s words are seen as pragmatic tools for making
understanding possible. And thus the commentator’s practice, as Buddhaghosa
sees it, is to make use of the tools embedded in buddhavacana itself. Recall that
the entire project is grounded in the dogma (or discovery) that the Buddha was
omniscient.
Finally, I  conclude by lifting up the last distinction considered in this
chapter—​that between qualified or contextual styles of discourse associated with
the Suttanta and the categorical utterances associated with the Abhidhamma—​
because it structures the next three chapters. We turn first to the narrative,
give-​and-​take, and dialogical modes of buddhavacana in the suttas, looking par-
ticularly at their framing narratives (nidānas) as the contexts whereby scripture
may be interpreted. Reading the suttas with Buddhaghosa is to attend to the

135. Erich Frauwallner initiated a tradition in modern scholarship that suggests that compared
to the Sarvāstivāda, the Pali tradition fell far short of the “doctrinal system, theoretical
considerations, and clear systematic thought” of the Sarvāstivādins. This view continues to
be echoed in current work, as for example, when Bronkhorst suggests that “the Sarvāstivāda
tradition fundamentally distinguished itself from the Pāli school in its attempt to order the
doctrine systematically.” See Bronkhorst’s discussion of “systematizing” (Buddhist Teaching in
India, 109–​10; he quotes the above quotation by Frauwallner).
136. Bond, The Word of the Buddha, 43.
106 Build ing Blo cks for an In t er pretat ive Pro gr a m

colloquial and contextual modes of buddhavacana, noting in particular how the

Buddha’s sermons were prompted and received by particular people whose stories,
inclinations, and dispositions are relevant to their interpretation. In the Suttanta
and Vinaya, the Buddha’s omniscience is performed by how he knows the sit-
uation of his interlocutors and speaks directly to it, in ways that Buddhaghosa
develops to have an impact, both literary and existential, on the ideal reader. And
then we consider in ­chapter  4 the analytical methods of categorical teachings
of the Abhidhamma, those true in every instance, which allow us to experience
buddhavacana in a quite different register as an analytic practice, all the way
down, that enacts the Buddha’s unobstructed knowledge in a different way. We
then return to pariyāya discourse in ­chapter  5, on the Vinaya. These chapters
deepen our understanding of how several of these methods of buddhavacana can
work in practice to explore the workings of the Buddha’s omniscient ken.
 PART II

Interpreting the Three Piṭakas

 3

The Contexts and Conditions

of Buddhavacana in the Suttanta

In ­c hapter  1 we explored some of the different forms the Buddha’s omnis-

cience is said to take, ranging from the very concrete and particular knowledge
of all beings’ dispositions to the analytical methods of Abhidhamma knowledge.
This chapter considers the aspect of the Buddha’s omniscience that concerns the
unimpeded knowledge of concrete particulars—​when he attends to a particular
matter his direct knowledge into it continues without obstruction. Omniscience,
as we have seen, is conceived not as a matter of knowing an encyclopedic corpus
of propositional facts, but rather as a practice of knowing an infinite plurality of
possibilities. It is a process rather than a body of material, and it is unfolding and
expanding in the very course of the Buddha’s ongoing awakening throughout his
career. Though the corpus of buddhavacana that gives expression to this omnis-
cience is a finite and complete body of texts, this “complete” corpus remains in
constant tension with the immeasurable expansiveness it enacts and evokes. The
tension is resolved in Buddhaghosa’s thought by showing how each genre of the
corpus demonstrates the activity or the workings of the Buddha’s omniscience.
The workings of the Buddha’s omniscience are exemplified and instantiated in
various genres and registers of his teachings. Omniscience is at work in the ana-
lytical methods of the Abhidhamma, which we explore in ­chapter 4. Here we are
concerned with the omniscience revealed in the course of his day-​to-​day teaching
practices to his disciples, rivals, and potential converts. Since the Buddha is said
to “know completely beings’ various inclinations and deeds,”1 his conversations
with people indicate his omniscience in action on the ground. That is to say, he
knows people in their unique and singular particularity and his teachings speak

1. D i.2: sattānaṃ nānādhimuttikatā suppaṭividitā.

110 In t erpr et ing t h e T hr ee PiṬak as

directly to them in a way that, in principle at least, knows no limits. The enact-
ment of this immeasurable knowledge is expressed in the Suttanta and Vinaya
genres, which are full of stories of the particular instances of people encountering
the Buddha’s knowledge of them. Both genres present the Buddha’s teachings as
embedded in very particular narrative contexts: both canon and commentary lo-
cate every sutta and vinaya rule in a story. Suttas begin not with the Buddha’s
voice, but with his closest disciple reporting what he had heard and the occasion
on which he heard it; Vinaya rules begin with Upāli, the reciter of the Vinaya,
telling the story of the infraction that prompted the laying down of the rule. Our
concern in this chapter is with suttas, but many of the considerations we explore
here are pertinent for ­chapter 5 on the Vinaya.
The narrative occasion of the sutta was called the nidāna: Ᾱnanda says, “Thus
have I heard,” and then recounts the time, place, audience, and other particulars
in which the Buddha’s sermon is given. The commentaries often develop further
the canonical account, sometimes adding significantly to the narrative setting
in which the Buddha gave his sermons. In this way, sutta knowledge is always
embedded in a story. Buddhaghosa argues for the importance of this setting and
how exactly it works as an interpretative device. The nidāna is the principal emic
category at work in this chapter, because it suggests how important narrative
context is for the interpretation of scripture through which the ideal reader
experiences the Buddha’s omniscience; this experience is configured as a contem-
plative and transformative practice of being existentially oriented to the qualities
of the Buddha.
As suggested in the previous chapter, the emic distinction between pariyāya
and nippariyāya, which here we might refer to as “contextual” and “categorical”
teachings, is also operative for understanding how Buddhaghosa interprets nar-
ratively situated knowledge in contrast to the more abstracted knowledge of the
Abhidhamma. As we saw, pariyāya knowledge is knowledge by way of “one in-
stance,” rather than the more categorical Abhidhamma methods that explore
what can be known in every instance. To follow up on the possibilities of how
knowledge speaks to the “one instance” that prompts it and to which it speaks is
to explore the embeddedness of doctrine in the original instance in which it was
uttered.
While not invoking the idea of nidāna or the distinction between pariyāya
and nippariyāya, Richard Nance has noted how the Buddha taught in a manner
both “presently and locally” and “tenselessly and universally.” He suggests,

the language of Buddhist teaching is thus Janus-​faced. One face looks

towards the local and responds to shifting historical, institutional, cultural
and personal conditions. The other face looks towards the translocal: to
 Buddhavacana in the Suttanta 111

that which is stable and persists across time. To date, scholarly work on
Buddhist philosophy has tended to focus on the latter face, and to view
the former as a matter of dispensable (usually rhetorical) ornament.2

Nance shifts the focus to the contextual, dialogical, and pedagogically situated
teachings, and urges attentiveness to the Buddha’s “skills of orientation and ap-
plication” in addressing his audiences, and in turn their “responsiveness.”3 While
concerned mostly with Mahāyāna sources and the notion of skillful means,
Nance argues for the philosophically rich possibilities of the dialogical features of
the Buddha’s teachings in ways aligned with my approach in this chapter.
It is often easy for modern scholars, based on rationalist assumptions, to iden-
tify the significance of those kinds of textual knowledge that are delivered in terms
of propositional content, where (apparently) decontextualized assertions appear
to have universal import and relevance. A  dominant style of modern Western
philosophy, for example, is given in the form of the systematic treatise, a genre
that speaks in a highly universalist register. Systematic treatises give and appear to
require no interpretative frame through which they must be understood.
In contrast, another kind of interpretative frame widely developed and used
by modern historically minded scholars involves interpreting a text in relation-
ship to its historical contexts (variously conceived as intellectual, social, polit-
ical, economic, etc.). The idea that a historical framework is useful or required
to interpret a text (even one that makes no explicit reference to such a context
in making its propositional claims) may seem to call into question the text’s uni-
versal significance and to render parochial its knowledge. When we describe a
context outside the text whereby it was produced and received, we risk narrowing
the scope of its relevance to that context:  if a text was produced by and for a
particular historical or intellectual community, then how can it have universal
relevance?
This of course is not an unsolvable problem, for surely we must allow that
thought can reach outside of its historical location and speak broadly to human
beings even quite distant from its “original” context. Still, I think the decontex-
tualized treatise of propositional claims and arguments has shaped a prevailing
conception of philosophy as restricted to this form. This valorization of what
we might call decontextualized and universalist discourse can sometimes make

2.  Nance, “The Voice of Another,” 367. He has in mind Griffiths, “Denaturalizing
Discourse: Ābhidhārmikas, Propositionalists, and the Comparative Philosophy of Religion” as
an example of a scholar focusing on the timeless face (Nance, “The Voice of Another,” 374 n5).
3. Nance, “The Voice of Another,” 368.
112 In t erpr et ing t h e T hr ee PiṬak as

it hard for us to see the philosophical potency of contextually situated and dia-
logical discourse. It might be one reason why modern scholars often pull out the
sermon’s doctrinal “content” stripped of its dialogical setting when reading Pali
suttas today.
Buddhaghosa was not interested in reading texts to discover their histor-
ical context in the modern sense, but he did think that knowing the narrative
contexts of people and events in the Buddha’s world makes us better readers
of suttas and vinaya rules. And he has a strong sense of the history of the First
Council as well. He draws the notice of his reader to at least two moments of the
past. The first moment is when the nidāna of every sutta was spoken by Ānanda
at the First Council; asked about the particulars of the sutta—​where, when, and
to whom it was given, and so forth—​Ānanda reported these details. Ānanda’s
words take the reader back to the “original” moment in which the dialogue took
place. Buddhaghosa sees these two events as important for the interpreter to
grasp: “since it is to be asked ‘by whom was this utterance said, when, and why
was it said?’ [it should be] replied that ‘it was spoken by Venerable Ānanda at
the time of the First Council.’ So the First Great Council is the beginning that
should be understood first for the sake of proficiency with the nidāna of each
sutta.”4 This is a layered sense of time involving an account of the original context
by drawing attention first to the “historicity” of the reception and transmission
of that account.
In addition, Buddhaghosa is, as we have seen, highly attuned to different
registers of the Buddha’s teaching; for instance, the pariyāya (contextual) and
nippariyāya (categorical) distinction might seem to replicate the opposition
roughly sketched here between contextually situated and context-​free teachings.
For him,5 both are equally the Buddha’s teachings and they do not stand in a
hierarchical relation to each other; neither has a higher claim in teaching “the
truth.” Nor are they entirely mutually exclusive, nor is the distinction absolute.
(Contextualized teachings contain, embedded within them, nippariyāya claims.
And in some important sense, as we will see especially in the next chapter, even

4.  Pj I.89:  Yato vattabbametaṃ ‘‘idaṃ vacanaṃ kena vuttaṃ, kadā, kasmā ca vutta’’nti.
Vuccate—​āyasmatā ānandena vuttaṃ, tañca paṭhamamahāsaṅgītikāle. Paṭhamamahāsaṅgīti
cesā sabbasuttanidānakosallatthamādito pabhuti evaṃ veditabbā.
5.  Ramanujan’s influential essay on the context-​sensitivity of much classical Indian thought
downplays to some extent the universalist and decontextualized registers of much Indian
thought, but he also notes the “multiple diglossia” that also allowed classical Indians to be fully
capable of context-​free modes of thought as well (“Is There an Indian Way of Thinking? An
Informal Essay,” 57). Where Buddhaghosa goes further than Ramanujan is in his explicit ideas
about how these two registers allow for specific interpretative practices.
 Buddhavacana in the Suttanta 113

nippariyāya teachings are given some sort of context in which they are to be
interpreted, a context he is helping to create through his commentaries.) Still,
the idea of teachings that offer a context through which the doctrinal content of
a teaching is usefully interpreted was explicitly contrasted with an idea of catego-
rical teachings that require no additional contextual framing to interpret.
I think that Buddhaghosa’s thinking about narrative can reframe some of
our assumptions about context. Modern scholars sometimes read Pali suttas as
offering self-​contained propositional content that can be understood independ-
ently of its narrative context. When they have been interested in context it has
often been to reconstruct the social and intellectual context of the Buddha’s mi-
lieu. Neither approach is “wrong,” and certainly many important interpretations
of the Buddha’s teachings have been advanced by these methodological choices.
But they are not the only way to read a Buddhist sutta, and—​I argue throughout
this chapter—​they can miss hugely important opportunities for understanding
and experiencing doctrine that the Pali commentarial tradition deemed signifi-
cant. In what follows I emphasize two critical features of Buddhaghosa’s reading
practices of the suttas.
First, for Buddhaghosa, reading the narrative contexts in which teachings are
embedded is an essential way that the ideal monastic reader is to approach and
experience the Buddha. In each of the three nidānas to the piṭakas, Buddhaghosa
emphasizes not just omniscience but all of the “qualities of the Buddha”
(buddhaguṇa) and develops them exegetically and contemplatively. Narrative
contexts do important buddhological work to assert and elaborate the qualities of
the Buddha and his overarching narrative to which the ideal reader is to become
oriented. This is made explicit in the case of the Suttanta, when Buddhaghosa
treats the nidāna as a “threshold” or doorway to enter into this piṭaka in a way
that will reveal the Buddha’s qualities, as we will see.
Second, Buddhaghosa’s interpretative methods in reading suttas involve a lit-
erary reading sensitive to the dialogical nature of the Buddha’s teachings. He is
interested in the “on the spot” nature of the Buddha’s teaching style in which he
engages the immediate concerns of his audience in a way that constitutes and
enacts the very unfolding of his awakening mind. The sophisticated dialogical
character of the suttas demonstrates the Buddha’s transformative impact on his
immediate audience, and (ideally) implicates the reader as well.

Beyond the Fractured Text
In an instructive article about various ways of approaching suttas for historical
understanding, Jonathan Walters identifies an interpretative style of reading Pali
suttas that he calls the “textual whole mode,” and that he associates with Steven
114 In t erpr et ing t h e T hr ee PiṬak as

Collins. This method reads suttas in relation to the narrative frames in which they
occur in a manner attentive to their literary qualities. He argues that attending
only to the embedded doctrinal teaching of a sutta “depends upon fracturing the
integrity of the sutta as received.”6 Suttas themselves refer to the stories in which
the Buddha’s teachings were given, stories that can give us a layer of historicity “of
composition, of aesthetics, of reading” that can only be grasped by taking the text
as a “whole.” He argues that through attending to the larger textual whole “both
history and philosophy are enriched by considering the frames within which the
fragments are, we assume purposefully, situated.”7
Of course, we are obliged to ask exactly what is to constitute the textual
“whole” in this vast and multilayered canonical tradition. Walters is careful to
acknowledge the different types of canonical wholes that can be invoked in this
method. The Ariyapariyesanasutta (the “Sermon on the Noble Quest”) on which
he focuses for his example can be read in terms of various textual “wholes”: the
embedded story of the teaching, the Majjhimanikāya in which the sutta occurs,
the Suttanta genre as a whole, and the tipiṭaka itself.8 The canonical layer is also
locating suttas in a biographical whole composed of episodes within the Buddha’s
teaching career and his interactions with his contemporaries. The literary imagi-
nation of the jātaka stories, for example, provides a vast and imaginative literary
whole in which we can consider the dramatis personae of many of the Buddha’s
world (as can the Vinaya literature and the Therī-​ and Theragāthās, to offer fur-
ther specific examples).
Walters goes further to advance an additional method of reading suttas that
looks also at their subsequent history of reception, a reception that can be studied
through the commentaries as well as the supplementary texts that build on suttas.
In the case of his example, the Ariyapariyesana appears to be the kernel of the
Buddha’s biography on which an enormous supplementary biographical litera-
ture was developed, including such texts that might seem far removed from it,
such as Aśvaghoṣa’s Buddhacarita. (Walters says “one could write a veritable
history of Buddhology, if not the whole religion, as a process of supplementing
the original biography of [the Ariyapariyesana Sutta].”9) And of course the

6. Walters, “Suttas as History,” 266.

7. Walters, “Suttas as History,” 266.
8. Walters, “Suttas as History,” 268.
9. Walters, “Suttas as History,” 276. However, Finot and Frauwallner suggest that the canon-
ical Vinaya and the Mahāparinibbāna Sutta contain the core of later biographical elaboration
(see Frauwallner, The Earliest Vinaya and the Beginnings of Buddhist Literature, ch. 3, for this
argument).
 Buddhavacana in the Suttanta 115

commentaries provide a very specific history of readership and reception that ad-
vance their readings as the authorized interpretations of the tradition. Walters
notes how Buddhaghosa takes his commentarial role to involve much supple-
mentation as he furnishes the “background” of each sutta in his commentarial
elaborations of the nidāna. In this context, Walters’s response to Buddhaghosa’s
readings of texts is notable:

Buddhaghosa always does so much more than confirm my expectation of

the details that ought to be included: he makes me hear details that I did
not expect ought to be included and puts forth what strike me as rather
bizarre readings of his own. These no doubt spoke to the sociohistorical
and literary worlds in which Buddhaghosa, like any author, operated. But
they speak to me, starkly.10

What startle Walters are the “radically different” agendas (different from that of
modern scholars, that is) that Buddhaghosa’s commentaries advance. In partic-
ular, he describes an agenda aimed at constructing, often through “excruciating
detail,” the Buddha’s daily habits, which produces “nothing less than a docetic
Buddha, only pretending to be an ordinary human being.” Buddhaghosa’s agendas
are “so radically different from my own that it takes great effort even to fathom
what he is saying and that in turn cautions me not to be too certain about seeing
my own readings ‘in the text.’ ”11 Walters suggests that Buddhaghosa’s readings
construct a “Buddhalogical vision far removed from ‘the historical Buddha’ as he
has been conceived by many scholars and Buddhist modernists in the nineteenth
and twentieth centuries.”12
I think that Walters is perceptive about much of this, and I dwell on his dis-
cussion at length here in part to signal that Buddhaghosa’s Buddha, as we have
already begun to see, is not the same historical persona imagined by modernist
Buddhists. What I  appreciate about Walters’s remarks is how even if Walters
finds his agendas rather fanciful, he allows Buddhaghosa to give him pause in
his certainties about his own interpretations. Walters allows Buddhaghosa to
“speak to me, starkly”—​not just as an authorized reader of the Buddha’s texts far
closer to the Buddha’s world than we are (and thus providing an important his-
torical layer of reception)—​but as a thinker whose interpretations can interrupt
his own. Buddhaghosa’s readings “privilege the frame, the textual whole, over the

10. Walters, “Suttas as History,” 279.

11. Walters, “Suttas as History,” 279.
12. Walters, “Suttas as History,” 283.
116 In t erpr et ing t h e T hr ee PiṬak as

embedded fragments”13 and in so doing construct a literary world in which the

Buddha and his teachings look quite different than how modern scholars have
often understood them. For me, it is precisely this encounter with the unexpected,
the happening on “radical” and “rather bizarre” choices, that I am drawn to in
reading ancient texts, because it is in these that we find opportunities to learn
something that we do not already know. Buddhaghosa introduces us to a vast
and imaginative literary world that can potentially take us deeper into teachings
we think we understand, and perhaps even confront us with understandings we
have not even conceived. I argue below that once we understand the rationality
of Buddhaghosa’s interpretative choices, his readings, while quite different from
those of modern scholars, are not bizarre and unexpected, but part of the larger
systematic and critical understanding of scripture that I have been sketching.

The Nidāna as a Threshold to the

Majestic Buddha and Dhamma
Various forms of introductory stage-​ setting—​what the commentaries call
nidāna—​were important to both canonical and commentarial practices. The
word “nidāna” can be understood in several ways, and we need to keep all of its
possible meanings at play: a nidāna is a context, origin, source, occasion, intro-
duction, or causal condition. When used to name part of a sutta or other text it is
the introduction or narrative context in which the text occurred in the Buddha’s
teaching career. There are different types of nidānas in texts, and we explore sev-
eral to begin to appreciate their range.
As far as the canon is concerned, sutta teachings are framed by the contexts
in which they come about, and vinaya rules are embedded in the narratives of
the violations that necessitated them.14 At the commentarial level, Buddhaghosa
records that at the First Council, Upāli, the reciter of the Vinaya, and Ᾱnanda,
reciter of the Suttas, were, prior to being granted the role of reciters, interrogated
by Mahākassapa as to the narrative (vatthu), context (nidāna), and the audience

13. Walters, “Suttas as History,” 282.

14. The suttas in the Aṅguttara have rather scant nidānas, which is a feature of the numerical
schematic in which these teachings are organized; the suttas are somewhat decontextualized
from their narrative settings to be worked into the numbered schema of this text. For a close
study of the prose formulas used to describe people approaching the Buddha in the Dīgha
Nikāya suttas, see Allon, Style and Function. The Samantapāsādikā lists narrative (vatthu),
origin (nidāna), person (puggala), rule (pannatti), corollaries to the rule (anupannatti), and
offences (apatti) and exceptions (anapatti) as the major structuring devices for each rule in
the Suttavibhaṅga (Sp i.14; Bond, The Word of the Buddha, 159). We discuss Vinaya methods
in ­chapter 5.
 Buddhavacana in the Suttanta 117

(puggala) of specific rules and teachings uttered by the Buddha.15 In the case of
the Suttanta, Ᾱnanda is queried as to whether he knows where, when, and to
whom were taught the first two suttas (the Brahmajāla and the Sāmaññaphala)
of the Dīgha Nikāya, the first book of the Suttanta.16 The narrative frame of a
sutta, in which Ᾱnanda utters “Thus have I heard” and names the particulars of
the occasion of the sermon, constitutes the answer to Mahākassapa’s queries and
comes to be recorded as the nidāna of each sutta. The suttas as we have them have
this “historical” moment of transmission recorded within them. Sometimes the
final words at the end of a sutta refer back to this larger narrative frame and de-
scribe the sermon’s impact on that original audience. Such particulars of the or-
igin, circumstances, and reception of a sermon—​where, when, on whose account
it was given, and its impact—​situate the teaching in a particular moment in time
and connect it to a particular audience.
The nidāna is not limited to canonical texts, however, and the commentators
and subcommentators often expand substantially the original canonical nidāna
and furnish their own supplementary narratives on the setting of a sutta. I refer
to these as the “commentarial nidāna” as opposed to the “canonical nidāna” that
is stated in the sutta. Deploying a great body of received narrative and expository
lore, the commentators often fashioned quite extensive commentarial nidānas
that add significantly to the original setting.
Nidānas can also be the introductions to entire texts or entire genres. For ex-
ample, the aṭṭhakathā on the jātaka verses provides a Jātakanidāna for the entire
jātaka corpus (which gives us the earliest Pali attempt to fashion a more or less
complete biography of the Buddha). And each piṭaka as a whole is also said to
require a nidāna that the commentator provides:  Buddhaghosa locates the
Suttanta piṭaka within a commentarial nidāna set forth at the beginning of the
Sumaṅgalavilāsinī;17 the Vinaya is introduced through a narrative context set up
in the opening of the Samantapāsādikā;18 and the Atthasālinī provides a large
nidāna on the Abhidhamma.19 It is of course these very introductory framings for

15. Sp i.30; Sv.i.12.

16. Sp i.14; Sv i.14, Sv i.25; Pj 97.
17. Sv i.2–​50, which is Buddhaghosa’s commentary on D i.1–​2; much of this he reiterates at the
beginning of each of the nikāyas. See Bhikkhu Bodhi’s very helpful translation and discussion
of the Brahmajāla Sutta and its commentary and subcommentaries, though note that his trans-
lation of the exegetical texts is “composed after the fashion of a montage, drawing selectively
from” the commentary and subcommentary (The All-​Embracing Net of Views, vi).
18. Jayatilleke, trans., The Inception of Discipline and the Vinaya Nidāna.
19. As 1–​35. Pe Maung Tin and C. Rhys Davids, trans., The Expositor, 1–​45.
118 In t erpr et ing t h e T hr ee PiṬak as

the genres that furnish so much of what I have been exploring for Buddhaghosa’s
theory of texts, as it is in these efforts to describe and contextualize these genres
that we get his most systematic bird’s-​eye view of textual wholes.
In the context of the suttas, commentarial nidānas are presented as
elaborations of introductory matter present in the canonical sources themselves.
They can help support the authenticity and authority of the canonical material,
though Buddhaghosa does not emphasize this function of the nidāna in the way
that Dhammapāla does, as I discuss later. The nidāna is also explicitly described
in the exegetical handbooks as an interpretative tool. As a formal exegetical de-
vice described in the Nettippakaraṇa, the nidāna is a commentarial practice that
comprises, in part, the sixth of the sixteen “modes of conveyance” (hāra), that
is, part of a fourfold array of interpretation tools—​linguistic analysis (nerutta),
context (nidāna), authorial intention (adhippaya), and consecutive sequence
(pubbāparasandhi). The nidāna is an account of the occasion or narrative (vatthu)
that prompts the Buddha’s utterance of the teaching.20 Buddhaghosa uses the
nidāna as an interpretative tool, but does not make the same effort Dhammapāla
does to link it to the hāra scheme outlined in the Netti.
Though the nidāna as a literary and interpretative feature has not captured
much modern attention, where they have noticed it, modern scholars have seen
the nidāna largely as a device, as Lily de Silva puts it, to establish for an an-
cient literature “the historicity, authenticity, and the authority” of its canonical
texts.21 George Bond suggests that the nidāna provides the “setting in life” that
“furnishes the context which reveals the true intention of an otherwise vague or
abstract teaching.” From them we “can judge the authenticity and historicity of
nidānas by the extent to which they serve as indispensable guides to the meaning
of intent of a sutta.”22 Bond’s idea that the nidānas might be useful interpreta-
tive devices to help us with the meaning of the suttas can be seen as an important
advance from an earlier disdain for Buddhaghosa’s nidānas by B.  C. Law, who
states, “it goes without saying that the answers given [by nidānas] are legendary,
uncritical, orthodox and childish.” Law does allow that if one can “separate the
grain from the husk” they can offer “historical, geographical, biological, textual
and doctrinal details” that are of interest (we are not told, however, by what criteria

20. Nett 34–​35.
21.  De Silva, “Introduction” to her edition of Dīghanikāya-​aṭṭhakathā-​tīkā
Līnatthavaṇṇanā, lxvii.
22. Bond, The Word of the Buddha, 138. Nance translates nidāna (in Indian Mahāyāna texts)
as the “sound basis” used to “index where, when, why, and/​or how a teaching is transmitted”
(Speaking for Buddhas, 54; 225 fn 27).
 Buddhavacana in the Suttanta 119

we might distinguish what is grain from what is husk).23 As far as I know, however,
no modern scholar has turned to the commentators themselves to learn how they
interpreted the role of the nidānas.
Buddhaghosa’s most comprehensive discussion of the purposes of the canon-
ical nidāna occurs in his nidānas to each of the four nikāyas. He says the nidāna is
“adorned with the time, place, teacher, story, assembly, and region”24 in which the
sutta was given. The Buddha’s teachings are always given to someone, for a particular
reason, in a particular location, at a particular time. Buddhaghosa advances a highly
literary and aesthetic appreciation of the nidāna, as suggested in the following elab-
orate similes that explain what the nidāna does.

Thus far, the nidāna, adorned with time, place, teacher, narrative, assembly,
and region, is spoken by Venerable Ᾱnanda, and the commentary on its
meaning is complete. [It is spoken] for the sake of the ease [or pleasure] of
entering this sutta, which is perfect in meaning and phrasing and indicates
the power of the Buddha’s qualities. The nidāna is:
like a bathing place (tittha), a spot of ground white from strewn sand as
though its surface was spread with pearls, with a bejeweled stairway charming
and resplendent surfaced with flawless stones which is for the ease of entering
a lotus pond of pure, clear, and sweet water sparkling with lotuses and blue
water lilies
like a staircase radiant with the effulgence and flashing of the light of
a mass of gems intertwined with golden creepers, and with soft, delicate
landings made of ivory, for the ease of ascending a splendid palace with
height reaching up as though wishing to touch the paths of the stars, and
encircled by ornamented ledges and well-​proportioned walls
like a great door with a wide and well-​placed door post, shining majes-
tically with light from gold, silver, gemstones, pearls, corals, etc., for the
ease of entering a great mansion gleaming with the riches of a noble lord, a
house teeming with the sweet voices of laughter and talking mingled with
the sounds of golden anklets and bracelets jangling.25

23. Law, Buddhaghosa, 94.
24. kāladesadesakavatthuparisāpadesapaṭimaṇḍitaṃ nidānaṃ (Sv.i.50; Mp.i.119; Spk.ii.3).
a.iii.536:  Ettāvatā ca yaṃ āyasmatā ānandena kamalakuvalayujjal
25.  Sv.i.50; Ps.i.15; Pv-​
avimalasādhurasasalilāya pokkharaṇiyā sukhāvataraṇatthaṃ nimmalasilātalaracanavilā
sasobhitaratanasopānaṃ, vippakiṇṇamuttātalasadisavālukākiṇṇapaṇḍarabhūmibhāgaṃ
titthaṃ viya suvibhattabhittivicitravedikāparikkhittassa nakkhattapathaṃ phusitukāmatāya
viya, vijambhitasamussayassa pāsādavarassa sukhārohaṇatthaṃ dantamayasaṇhamudup
halakakañcanalatāvinaddhamaṇigaṇappabhāsamudayujjalasobhaṃ sopānaṃ viya, suva
120 In t erpr et ing t h e T hr ee PiṬak as

For anyone who thinks Buddhaghosa is nothing more than a dry-​bones pedant of
arid Abhidhamma lists, this literary flourish must surely occasion surprise. And
indeed, his subcommentator Dhammapāla is taken aback: in a rare expression of
impatience with Buddhaghosa, he chides him that this is not a matter for “poetic
composition.”26 What can Buddhaghosa mean by describing the nidāna in such
lavishly poetic terms?
The similes portray the nidāna as a portal, entryway, or steps for the pleasure
or ease (sukha) of entering a sutta. The sutta, in turn, “indicates the power of the
Buddha’s qualities.” The nidānas are beautiful and sparkly thresholds that give us
access to the rich places of the sutta—​the gorgeous lotus pond, the lofty palace
brushing the heavens, the magnificent lord’s palace. Buddhaghosa reads a sutta
as an aesthetically stunning place to be entered, and the nidāna that welcomes
us into it is deeply pleasing and exquisitely lovely. The first simile evokes the reli-
giously loaded language of the sacred crossing, the tittha.27 In the second simile,
the nidāna is likened to a staircase up to a magnificent and lofty palace that
reaches heavenward to touch the stars. This image suggests a theme of the infi-
nite: the nidāna is a threshold to something that reaches upward to touch the
immeasurable starry sky. We enter, in the final image in the sequence, a mansion
of a great lord (the mansion and palace are both suggestive of the mansions and
bliss worlds presided over by buddhas that we find in other forms of Buddhism).
This is an appreciation of the sutta that resonates with the extravagant terms
describing sūtras often associated with Mahāyāna textual traditions. By way of
the nidāna the sutta can bring one to the majesty of the Buddha.
The sutta, “perfect in meaning and phrasing,” always “indicates the power of
the Buddha’s qualities.” It is not merely a discursive teaching that can be held
apart from the Buddha, but rather it reveals or illustrates something about the
Buddha himself. And in fact, in his interpretative practice in the commentarial
nidānas that I have read, Buddhaghosa always tries to explain how a sutta and its

ṇṇavalayanūpurādisaṅghaṭṭanasaddasammissitakathitahasitamadhurassaragehajanavic
aritassa uḷārissarivibhavasobhitassa mahāgharassa sukhappavesanatthaṃ suvaṇṇarajata
maṇimuttapavāḷādijutivissaravijjotitasuppatiṭṭhitavisāladvārabāhaṃ mahādvāraṃ viya
ca atthabyañjanasampannassa buddhaguṇānubhāvasaṃsūcakassa imassa suttassa
sukhāvagahaṇatthaṃ kāladesadesakavatthuparisāpadesapaṭimaṇḍitaṃ nidānaṃ bhāsitaṃ,
tassatthavaṇṇanā samattāti.
26. na kabyaracanādi (ṭīkā i.72). See what follows.
27. The idea of the nidāna as a tittha (Sanskrit tīrtha, a crossing place, landing, or the steps
down to a river for bathing) also appears in the commentary on the Khuddakapāṭha (Pj I.157).
The term suggests the many aesthetic and religious qualities of a passageway into pleasure and
purification that we find in Indian religion more broadly. A tīrtha is a sacred space, and in some
conceptions it is the site of initiation into radical transcendence.
 Buddhavacana in the Suttanta 121

nidāna reveal the Buddha’s qualities—​most often, how it reveals his omniscience.
I give examples in what follows.
With this foray into the poetic and ornate, Buddhaghosa is evoking the beauty
of Ᾱnanda’s nidāna and the way it opens up the rest of the sutta and primes the
reader to encounter the magnificence of the Buddha. He is describing suttas in a
“literary” way (as the comment from Dhammapāla notes): suttas can be read for
pleasure and beauty and for the worlds they evoke or even constitute. The nidāna
plays an important aesthetic role in preparing the imagination for entering the
sutta perceptive to the beauty and pleasurable delight he associates with faith in
the Buddha. We have seen him elsewhere being quite explicit about this. Recall
that everything the Buddha said was “beautiful in the beginning, beautiful in the
middle, and beautiful at the end.” Here he follows through on the first part of his
claim that a sutta is beautiful in the beginning in its nidāna, at the end in its con-
clusion, and in the middle in the rest in between.28
I think Buddhaghosa’s literary and aesthetic choices about how to depict the
purposes of the nidāna can be cast into stronger relief if we contrast them with
Dhammapāla’s choices. We have already noted his disapproval of Buddhaghosa’s
poetic extravagance. He prefers to see the nidāna instead as establishing the
authority and authenticity of the teaching. He asks quite pointedly why the
nidāna is even necessary:

Why did they create a nidāna at the First Council for the Dhamma and
Vinaya, when surely only the Buddha’s words should be recited?
It was for the purpose of establishing the stable, unconfused, and cred-
ible nature of the teaching recited. For a teaching connected to a time,
place, teacher, story, and the recipient of the Dhamma is a long-​lasting,
unconfused doctrine, and credible, like a legal contract provided with
notations of place, date, maker, and witnesses.29

His primary concern in these remarks is with the way the nidāna demonstrates
the authenticity of a sutta. Details of its original context anchor it to the First
Council and to the authority ultimately of the Buddha, and they work like the

28.  suttaṃ nidānena ādikalyāṇaṃ nigamanena pariyosānakalyāṇaṃ sesena majjhe kalyāṇaṃ

(Vism 213–​14). See my “Introduction,” p. 11.
29.  Kasmā panettha dhammavinayasaṅgahe kariyamāne nidānavacanaṃ, nanu bhagavato
vacanameva saṅgahetabbanti? Vuccatedesanāya ṭhitiasammosasaddheyyabhāvasampāda
natthaṃ. Kāladesadesakavatthudhammapaṭiggāhakapaṭibaddhā hi desanā ciraṭṭhitikā
hoti, asammosadhammā saddheyyā ca. Desakālakattusotunimittehi upanibandho viya
vohāravinicchayo (Dīgha ṭīkā i.70).
122 In t erpr et ing t h e T hr ee PiṬak as

particular notations of witnesses, and so forth, for business and legal contracts.
This approach is markedly different in tone from that of Buddhaghosa, who does
not connect the nidāna in this way with the authority and credibility of textual
transmission.
But Dhammapāla goes on to say the nidāna also reveals the qualities of the
Buddha, with an emphasis, as with Buddhaghosa, on the Buddha’s omniscience.

Moreover, saying the nidāna is for [showing] the accomplishment of

the Teacher. [In particular it shows] that the accomplishment of Perfect
Buddhahood of the Blessed Tathāgata is free of prior composition, infer-
ence, scriptural authority, and reasoned conjecture. Because of his being
perfectly awakened there is just one authoritative means of knowledge: the
movement of his unhindered knowledge, everywhere free of prior compo-
sition, et cetera, into all knowable phenomena.30

The nidāna shows the way the Buddha spoke—​extemporaneously and without
relying on argumentation or authoritative texts. Rather he spoke from his omnis-
cience, defined always as his capacity to penetrate all things without obstacle. The
nidāna is the demonstration of this.
Dhammapāla goes on to say that the nidāna demonstrates the Buddha’s ac-
complishment “because it shows the teaching of the Dhamma with his on-​the-​
spot intelligence about the inclinations of the audience present.”31 It speaks
specifically to the particular audience and their particular needs. And the nidāna
reveals the accomplishment of the Teacher in other ways too:

Also, saying the nidāna is for [showing] the accomplishment of the

Teacher. Nothing the Blessed One does is useless or for his own sake, for
his every deed possesses knowledge and compassion. Since this is so, the
nature of the Teacher is that his every action of body, speech, and mind is
that of a Perfectly Awakened One, that is, one whose every deed occurs for
the sake of others. And so the nature of the Dispensation is not a matter
for poetic composition.32

30. Apica satthusiddhiyā nidānavacanaṃ. Tathāgatassa hi bhagavato pubbaracanānumānāgamat

akkābhāvato sammāsambuddhattasiddhi. Sammāsambuddhabhāvena hissa pubbaracanādīnaṃ
abhāvo sabbattha appaṭihatañāṇacāratāya, ekappamāṇattā ca ñeyyadhammesu (Dīgha ṭīkā i.71).
31.  nidānavacanena sampattaparisāya ajjhāsayānurūpaṃ ṭhānuppattikappaṭibhānena
dhammadesanādīpanato (Dīgha ṭīkā i.71).
32.  Tathā satthusiddhiyā nidānavacanaṃ. Ñāṇakaruṇāpariggahitasabbakiriyassa
hi bhagavato natthi niratthikā pavatti, attahitatthā vā, tasmā paresaṃyeva atthāya
 Buddhavacana in the Suttanta 123

This statement shows how the Buddha’s teachings were oriented to his audience,
and delivered entirely for its benefit. Since the nidāna describes that audience, it
displays precisely how the sermon was beneficial to them, something we would
not know without the details of the story. Since the nidāna is illuminating this
fundamental nature of the Buddha and his teaching, it is a matter too serious,
Dhammapāla thinks, to discuss in the poetic terms that Buddhaghosa does. Since
Dhammapāla thinks that the nidāna has two purposes—​showing authoritative-
ness (pamāṇa) of the Buddha and the Dispensation and showing the Buddha’s
accomplishment (siddhi)33—​these need to be clearly, and prosaically, stated.
Yet in his own way, Dhammapāla also resists totality and allows for the infinite.
In his final comments on this matter he wants to keep the range of possibilities
open for what a nidāna can do: he modestly suggests that he has “shown merely
the beginning of the purpose of the nidāna; for who is able to elucidate fully the
purposes of the nidāna spoken by the Treasurer of the Dhamma (i.e., Ᾱnanda)
whose awakening followed that of the Buddha himself ?”34
While there is an important difference in their styles in talking about the
purpose of the nidāna, Buddhaghosa and Dhammapāla share the idea that it
demonstrates the Buddha’s omniscience in action. As he demonstrates in his
practice in his commentarial nidānas, Buddhaghosa suggests that the nidāna
shows the Buddha’s infinite knowledge of beings in their particularity. Walters
is right that the nidāna creates an extraordinary Buddha. But there is a ration-
ality to this extraordinary buddhology about which the commentators are
quite reflective: the extraordinariness of the Buddha lies in the precise workings
of his omniscient ken and the task of the interpreter is to learn how to discern
those workings. In the literary interpretations of the nidānas the omniscience is
expressed in how the Buddha speaks to the particulars of his original audience.

The Nidāna in Practice

Now that we know something of what Buddhaghosa thought the nidāna does,
we may find that his commentarial supplements of the canonical nidāna are not

pavattasabbakiriyassa sammāsambuddhassa sakalampi kāyavacīmanokammaṃ satthubhūtaṃ,

na kabyaracanādisāsanabhūtaṃ (Dīgha ṭīkā i.71–​72).
33. Tena vuttaṃ ‘‘satthusiddhiyā nidānavacana’’nti. Apica satthuno pamāṇabhūtatāvibhāvanena
sāsanassa pamāṇabhāvasiddhiyā nidānavacanaṃ (Dīgha ṭīkā i.71–​72).
34.  Idamettha nidānavacanapayojanassa mukhamattadassanaṃ. Ko hi samattho
buddhānubuddhena dhammabhaṇḍāgārikena bhāsitassa nidānassa payojanāni niravasesato
vibhāvetunti (Dīgha ṭīkā i.72).
124 In t erpr et ing t h e T hr ee PiṬak as

nearly as unexpected or bizarre as first we might suppose. He usually explains

suttas according to how they illustrate the Buddha’s qualities, above all, his om-
niscience. One way a sutta can demonstrate his omniscience is in how it speaks to
the occasion in which it is situated. Much of the work of a Suttanta commentary
is to explore the complex dynamic between the doctrinal content of the sermon
and the way it speaks pointedly to its context to exhibit in action the Buddha’s
unhindered knowledge of particulars. Before exploring concrete examples of
this, some general observations about suttas and Buddhaghosa’s work with them
may be instructive. Three important features of Buddhaghosa’s nidānas are
worth pointing out first: his interest in the immediate everyday particulars of the
narrative setting, his recontextualizing these particulars in a much larger cosmic
setting, and the way that the nidāna story enacts or performs the didactic lesson
of the sutta.
The narrative nature of suttas always locates them in a dialogue: a sermon is
never stripped from the conversational location in which it takes place, and the
sheer range of interlocutors, conversations, debates, questions, confrontations,
and admonishments documented in the nikāyas is impressive. Philosophy
in this genre is fundamentally dialogical and embedded in layers and layers
of biographical stories; an enormous body of lore about these people and
conversations seems to have traveled in and with the canonical texts to inform
the commentarial project. Moreover, these contexts often have a markedly
“everyday” feel about them, a feature with which Buddhaghosa is very much
concerned. No detail of the original audience or the Buddha’s doings in his
work-​a-​day exchanges and conversations with people is unimportant. In the
Brahmajāla Sutta nidāna we consider later, details of the Buddha’s daily rou-
tine of bathing, sleeping, eating, and teaching are of considerable interest to
Buddhaghosa because it is in them that he makes present the Buddha’s knowl­
edge of immediate concrete particulars. We might suggest a literary principle
of specificity that makes possible the enactment of the nature of the Buddha’s
omniscient knowledge of particulars. Without such specifics in the narrative
setting, the Buddha would be speaking only in generalities. The power of
pariyāya discourse lies precisely in the nature of the “one instance” in which it
occurs and to which it speaks.
At the same time, Buddhaghosa is ever ready to pull back, often quite seam-
lessly, to a much grander narrative setting to evoke a cosmic timescape of past and
future lives that relocates and recontextualizes the specific instance in a breath-
takingly vast spatial and temporal perspective. The specificity is still there, but
the sense of the ordinary and everyday world of the Buddha’s interaction with
his disciples or adversaries gives way to a larger framework that includes lifetimes
that occurred eons ago, cited as a condition for the present encounter.
 Buddhavacana in the Suttanta 125

Finally, the narrative situation, whether immediate or remote, often illustrates

the philosophical or doctrinal content of the sermon. The narrative is an in-
stance of the ideas of the more general teachings of the sermon. This illustrates
the teachings with a concrete example, by making particular and specific what is
presented in more general terms in the doctrinal teachings of the sermons. The
narrative context inhabits, presents, or performs the teachings in specific, con-
crete, and singular ways.
A good example of all three of these features—​their specificity, their cosmic
context, and the way they illustrate the doctrinal content of the sermon—​occurs
in Buddhaghosa’s nidāna to the Mahānidāna Sutta, which I mention here very
briefly and partially.35 The sutta, which is about the nature of conditionality in
the teachings of dependent origination, is prompted by an “everyday” sort of ex-
change between Ᾱnanda and the Buddha. Ᾱnanda lives in close proximity to the
Buddha and takes advantage of that closeness to ask the Buddha lots of questions.
Buddhaghosa says that Ᾱnanda “ordinarily approached the Exalted One even a
hundred or a thousand times a day, but never without a reason; and so too on this
day he came forward with a question.”36 This little detail suggests the well-​known
intimacy between Ᾱnanda and the Buddha, where Ᾱnanda is forever running to
him with questions, even if, of course, he has only good reasons for doing so. On
this day, a question had occurred to him as he had finished his alms round, swept
his quarters, shook out his bed mat, completed his other chores, and meditated.
The question Ᾱnanda asks is why, when the Buddha has asserted that the nature
of dependent origination “is deep and it appears deep” that it seems “as clear as
clear can be to me?”37
The Buddha both praises Ᾱnanda for his insight and scolds him for his im-
pertinence, for the doctrine of dependent origination is always deep and al-
ways appears so. (It is, of course, a trope in the literature around Ᾱnanda that
he must regularly be praised and admonished for one and the same action.)
If conditionality seems clear to Ᾱnanda it is only because of who he is:  one
who has heard much, lived as a teacher, is a stream-​enterer, and endowed with
supporting conditions from the past.38 When Buddhaghosa explains what it

35.  Bhikkhu Bodhi has a wonderful translation of this sutta, its commentary, and its
subcommentary, that the reader is urged to read to see the many layered exegetical moves in
detail (The Great Discourse on Causation: The Mahānidāna Sutta and Its Commentaries).
36.  Sv ii.485:  So kiñcāpi pakatiyāva ekadivase satavārampi sahassavārampi bhagavantaṃ
upasaṅkamanto na ahetuakāraṇena upasaṅkamati, taṃ divasaṃ pana imaṃ pañhaṃ gahetvā.
37. D ii.55: Yāva gambhīro cāyaṃ, bhante, paṭiccasamuppādo gambhīrāvabhāso ca, atha ca pana
me uttānakuttānako viya khāyatī’’ti.
38. Sv ii.488: pubbūpanissayasampattiyā, titthavāsena, sotāpannatāya, bahussutabhāvenāti.
126 In t erpr et ing t h e T hr ee PiṬak as

means that Ᾱnanda had “supporting conditions from the past” that make it
possible for him now to discern the workings of conditionality, he steps out of
the everyday time frame to locate these conditions in a lifetime that occurred
100,000 aeons ago when Ᾱnanda was the younger brother of the Buddha
Padumuttara (the buddha at that time); his attendance on that buddha made
possible this moment of knowing even a deep teaching with reasonable clarity.39
Buddhaghosa effortlessly weaves the charmingly quotidian with the dazzlingly
cosmic in the same narrative moment. Moreover, one comes to see in the
narrative just how much the past, even the remote past, is a condition for a pre-
sent occasion—​the narrative context replicates and enacts the philosophical
teaching that follows concerning the precise workings of conditionality in de-
pendent origination.
This is a demonstration in brief; to demonstrate in detail how these features
work requires dwelling on the particularities of multiple nidānas. These are often
richly drawn where each detail is important, and Buddhaghosa’s exegeses can be
lengthy. Arguments about the particular can only proceed by way of examples of
those particulars. In an important sense the best demonstration for how this works
is reading many suttas and their commentaries, as Buddhaghosa’s nidānas them-
selves illustrate what he thinks exegesis should look like. I content myself with here
focusing on two extended examples, which I cannot treat exhaustively, but which
we can begin to explore with some depth by drawing the reader’s attention to some
of his devices. Buddhaghosa’s nidāna on the Brahmajāla Sutta offers a complex
treatment of the narrative particulars prompting a sermon, and his discussion of the
Mūlapariyāya Sutta provides an intriguing reception of a sermon.

The Brahmajāla Sutta: Grasped and Held

in a Web of Views
For Buddhaghosa, the Brahmajāla Sutta concerns first and foremost the
Buddha’s omniscience. For Buddhaghosa, both the sermon and the narrative
enclosing it are about the Buddha’s omniscience, and they demonstrate that om-
niscience. The Brahmajāla is the first sutta of the first book of the Dīgha Nikāya,
which gives it pride of place in the Suttanta. Buddhaghosa assigns it special sig-
nificance when he describes the Suttanta Piṭaka in its terms as he distinguishes
the parameters of each piṭaka: the Suttanta is a “teaching on explaining views,
for the explaining of views is taught in it as what opposes the holding of the

39. Sv ii.488–​92.
 Buddhavacana in the Suttanta 127

sixty-​two views.”40 We shall come to see how the Buddha’s treatment of views
reveals his omniscience.
The Brahmajāla begins with an encounter with a rival ascetic Suppiya, who
disparages the Buddha even while his own student praises him. The Buddha’s dis-
ciples discuss this praise and blame, prompting the Buddha to give a teaching
that begins with the importance of not heeding either. Praise of a teacher is often
rather shallow, based as it is on “minor” matters of morality instead of the knowl­
edge of the sixty-​two viewpoints the Buddha goes on to describe. The Buddha’s
understanding that transcends all sixty-​two viewpoints is his not insignificant
insight about the “basis” of views (diṭṭhiṭṭhāna): views are ideological positions
grounded in and conditioned by feeling (vedanā). Underneath the grasping
and promoting of metaphysical positions is affective experience. Because he
understands the way that feelings condition views and he knows the future states
to which they lead, the Buddha catches all views in his “net” (jāla), and so this
sutta is called “the sublime net” (brahmajāla).
The sutta says that what the Tathāgata understands about the sixty-​two views
is that they are grasped and held tight out of feeling.

Monks, the Tathāgata understands this: these bases of views, grasped in

this way, held in this way, lead in this way to certain future conditions. And
the Tathāgata understands this: he understands what transcends this, but
yet he does not hold on to this understanding, and because of not holding
on to it he alone on his own has found peace. Monks, the Tathāgata is
free of all attachment having truly known the arising and disappearing of
feelings, and the satisfactions, dangers, and escape from them.41

The insight achieved by the Buddha is that those who promote metaphysical
views do so driven by feeling: we feel attached to our views. The sutta does not

40. Sv i.19; As 21; Sp 22: dvāsaṭṭhidiṭṭhipaṭipakkhabhūtā diṭṭhiviniveṭhanā ettha kathitāti diṭṭ

hiviniveṭhanakathā.
41.  D i.16:  Tayidaṃ, bhikkhave, tathāgato pajānāti—​ ‘ime diṭṭhiṭṭhānā evaṃgahitā
evaṃparāmaṭṭhā evaṃgatikā bhavanti evaṃabhisamparāyā’ti, tañca tathāgato pajānāti, tato
ca uttaritaraṃ pajānāti; tañca pajānanaṃna parāmasati, aparāmasato cassa paccattaññeva
nibbuti viditā. Vedanānaṃ samudayañca atthaṅgamañca assādañca ādīnavañca nissaraṇañca
yathābhūtaṃ viditvā anupādāvimutto, bhikkhave, tathāgato (this formula is repeated at D i.22,
24, 29, 39). See Evans, “Epistemology of the Brahmajāla Sutta,” 76–​77, on this; I am largely
in agreement with his interpretation of this passage and his reading of the sutta as a whole,
for some of which he is guided by Buddhaghosa. See also Gethin, “Wrong View (micchā-​
diṭṭhi) and Right View (sammā-​diṭṭhi) in the Theravāda Abhidhamma,” on the psychology
of holding views.
128 In t erpr et ing t h e T hr ee PiṬak as

refute the propositional content of the sixty-​two views described in the sutta,
but instead points out the inability of those who would hold fast to views to
recognize, as the Buddha does, their conditioned nature—​the ways they are
felt or experienced (vedayita). (Anticipating the next chapter, we can suggest
that the Buddha has achieved the Abhidhammic perspective that can see the
structures in the felt experience of holding metaphysical views: feeling [vedanā]
is a key Abhidhamma category.) Other scholars too have noted that the sutta
is interested in the affective or “psychological underpinnings” of holding and
promoting views, as Anālayo puts it,42 and exposing the subjectivity involved
in advancing truth claims that purport to be objective and independent of the
subject holding them, as Evans sees it.43 Similarly, in his reading, Buddhaghosa
emphasizes the sutta’s message that grasping and holding views is conditioned
by other processes. He says that there are eight kinds of “bases of views which
are the causes of views” and enumerates them as “the aggregates, ignorance, con-
tact, cognizing, initial thought, lack of careful attention, wicked friends, and the
voice of another.”44
And so the Buddha knows where each view comes from and where it leads.
The holding of each view, according to the sutta, entails very specific consequences
(“certain future conditions”); it leads to constraints and distortions in one’s un-
derstanding and circumscribes the limits of where one can be reborn in the af-
terlife (after describing each view, the Buddha says where adherents of that view
will go after death). What the Brahmajāla captures in its net is the conditions and
results of holding fast to the sixty-​two views that collectively constitute the com-
plete stock of metaphysical dogmas. The net of views captures and traps people
in saṃsāra, and it is itself a consequence of being trapped by feeling. Notice,
too, that the passage asserts the Buddha’s own freedom from even this higher
knowledge:  he understands the conditionality of views and transcends these
conditions, but he does not himself hold fast to that higher understanding that
sees vedanā at work in them.
Buddhaghosa interprets the sermon to reveal the workings of Buddha’s
omniscience. This is evident in his reading of this key passage that is repeated
several times in the sutta:

42. Anālayo, Satipaṭṭhāna: The Direct Path to Realization, 161.

43. As emphasized by Evans, “Epistemology of the Brahmajāla Sutta,” 76–​77, 84.
44.  Sv i.107:  Api ca diṭṭhīnaṃ kāraṇampi diṭṭhiṭṭhānameva. Yathāha ‘‘katamāni
aṭṭha diṭṭhiṭṭhānāni? Khandhāpi diṭṭhiṭṭhānaṃ, avijjāpi, phassopi, saññāpi, vitakkopi,
ayonisomanasikāropi, pāpamittopi, paratoghosopi diṭṭhiṭṭhāna’’nti.
 Buddhavacana in the Suttanta 129

There are, monks, other dhammas, deep, difficult to see, difficult to un-
derstand, peaceful, lofty, beyond the sphere of reasoning, subtle, experi-
enced by the wise, which the Tathāgata, having experienced himself with
his own higher knowledge, makes known. It is due to these that those
speaking rightly would speak true praise of the Tathāgata.45

This passage is drawing on the contrast between the minor points of morality for
which teachers are often praised in contrast to the deeper, rarer dhammas taught
by the Tathāgata. The question is what dhammas means here. At first glance it
would seem to refer to the Abhidhamma categories of phenomena, like vedanā,
that describe experience. But while acknowledging that dhamma has several
different meanings including this one, Buddhaghosa says that here it means qual-
ities (guṇa), and specifically, the Buddha’s qualities:

[Dhamma] can mean teaching in such cases as “monks, I  will teach a

Dhamma beautiful in the beginning, et cetera.” It can mean study in such
cases as “here, monk, one masters a sutta, a recitation, et cetera.” It can refer
to something without essence in such cases as “on that occasion there are
dhammas, there are aggregates, et cetera” [quoting the Abhidhamma text
the Dhammasaṅgaṇī]. But here it occurs in the sense of quality. Therefore,
monks, the meaning should be seen here in the sense of the qualities of the
Tathāgata, specifically [his] knowledge.46

Even more specifically, it means here one quality: his omniscience.

But which are these qualities praised thus by the Bhagavan? The knowl­
edge of omniscience.47

45.  D i.12, 17, 22, 29, 39:  atthi, bhikkhave, aññeva dhammā gambhīrā duddasā duranubodhā
santā paṇītā atakkāvacarā nipuṇā paṇḍitavedanīyā, ye tathāgato sayaṃ abhiññā sacchikatvā
pavedeti, yehi tathāgatassa yathābhuccaṃ vaṇṇaṃ sammā vadamānā vadeyyuṃ.
46.  Sv i.99:  ‘‘Dhammaṃ, vo bhikkhave, desessāmi ādikalyāṇa’’ntiādīsu (ma. ni. 3.420)
desanāyaṃ. ‘‘Idha bhikkhu dhammaṃ pariyāpuṇāti suttaṃ, geyya’’ntiādīsu (a. ni. 5.73)
pariyattiyaṃ. ‘‘Tasmiṃ kho pana samaye dhammā honti, khandhā hontī’’tiādīsu (dha. sa.
121)  nissatte. Idha pana guṇe vattati. Tasmā atthi, bhikkhave, aññeva tathāgatassa guṇāti
evamettha attho daṭṭhabbo.
47.  Sv i.99:  Katame ca pana te dhammā bhagavatā evaṃ thomitāti? Sabbaññutaññāṇaṃ.
Buddhaghosa acknowledges that the term dhammā is in the plural, and omniscience is stated
in the singular. But since omniscience takes different forms, and has multiple objects, it can be
referred to in the plural (Sv i.99–​100: Katame ca pana te dhammā bhagavatā evaṃ thomitāti?
Sabbaññutaññāṇaṃ. Yadi evaṃ, kasmā bahuvacananiddeso katoti? Puthucittasamāyogato ceva,
puthuārammaṇato ca).
130 In t erpr et ing t h e T hr ee PiṬak as

This is to say that in the canonical praise, the Bhagavan is praising his own quality
of omniscience, which is “deep, difficult to see, difficult to understand, peaceful,
lofty, beyond the sphere of reasoning, subtle, experienced by the wise” to quote
the original verse. In this way, Buddhaghosa shows how the insight the Buddha
had about the basis of views is the enactment of his omniscient mind.
At several other key junctures Buddhaghosa reads the sermon as demonstrating
the Buddha’s omniscience. First, he argues that there are four occasions in par-
ticular that demonstrate the unique ken of the Tathāgata: (1) the declaration of
the Vinaya; (2) the [classification of phenomena] according to plane (which he
takes to be Abhidhamma knowledge); (3) the workings of conditionality (that
is, dependent origination); and (4) the different views.48 We briefly considered
the workings of conditionality earlier in the Mahānidāna Sutta, and of course
omniscience’s workings in the Abhidhamma knowledge and the Vinaya rules are
explored in the next two chapters. The Brahmajāla Sutta exemplifies the fourth
occasion, the Buddha’s unlimited capacity to understand views and what grounds
them. Its purpose is to illuminate this aspect of the Buddha’s knowing:

Having arrived at the differences of views, the thundering of buddhas

becomes great, their knowing is entered, the magnitude of the knowledge
of buddhas can be understood, and their teaching stamped with the three
characteristics and connected with emptiness becomes deep.49

Buddhaghosa argues here that the sermon allows one to begin to understand
the magnitude of the Buddha’s knowledge and to “enter” into it. This pro-
found teaching is a didactic exercise in demonstrating emptiness and the three
characteristics—​impermanence, suffering, and lack of self—​perceivable in expe-
rience. Reiterating this, he argues elsewhere that the structure of the Brahmajāla
is “to start with views and conclude with making known emptiness.”50 This survey

48. Sv i.100; ii.485: “On four occasions the thundering of buddhas becomes great, their knowl­
edge is approached, their magnitude is understood, and their teaching stamped with the three
characteristics [of saṃsāra] and connected to emptiness, becomes deep. What are these? His
declaring the Vinaya, the planes of existence, the workings of conditionality, and the differences
of doctrines” (Buddhānañhi cattāri ṭhānāni patvā gajjitaṃ mahantaṃ hoti, ñāṇaṃ anupavisati,
buddhañāṇassa mahantabhāvo paññāyati, desanā gambhīrā hoti, tilakkhaṇāhatā, suññatāpaṭi­
saṃyuttā. Katamāni cattāri? Vinayapaññattiṃ, bhūmantaraṃ, paccayākāraṃ, samayantaranti).
49.  Sv i.102:  Iti samayantaraṃ patvā buddhānaṃ gajjitaṃ mahantaṃ hoti, ñāṇaṃ
anupavisati, buddhañāṇassa mahantatā paññāyati, desanā gambhīrā hoti, tilakkhaṇāhatā,
suññatāpaṭisaṃyuttāti.
50. Sv i.123: Imasmimpi brahmajāle heṭṭhā diṭṭhivasena desanā uṭṭhitā, upari suññatāpakāsanaṃ
āgataṃ.
 Buddhavacana in the Suttanta 131

of views demonstrates emptiness by showing how views are conditioned by the

processes of dependent origination, specifically its link feeling. As conditioned by
psychological processes, such views are not independent, unconditioned truth
claims their adherents see them as, but are, instead, empty.
A further way that this particular sutta demonstrates the Buddha’s omniscience
is in how it speaks to the incident that prompted it, and this is my chief interest
here. One of the established protocols Buddhaghosa uses to explore methodi-
cally the nidāna is to discern which of the four reasons for the teaching of a sutta
(suttanikhepa) prevails in any particular case. There are four reasons or grounds
that prompt a sutta: (1) personal inclination, that is, the Buddha’s own reasons for
giving a sermon; (2) the inclinations of others; (3) the asking of a question; and
(4) the occurrence of a specific incident.51 He takes the Brahmajāla Sutta to have
been inspired by a “specific incident,” namely, the arising of praise and blame in
an encounter with the rival ascetic Suppiya.52 The specific incident is the story of
Suppiya.
The canonical nidāna states that the Buddha and his entourage were traveling
from Rājagaha to Nālandā when they found themselves walking along the same
path with Suppiya and his student Brahmadatta. Along the way Suppiya voices
vigorous disparagement of the Buddha to Brahmadatta, who resists and instead
defends the Buddha. This prompts the Buddha’s disciples in later talking among
themselves to assert:

It is wonderful and marvelous, friends, how the Exalted One, he who

knows and sees, the Worthy One, the perfectly enlightened Buddha, has
so thoroughly penetrated the diversity in the dispositions of beings. For
this wanderer Suppiya spoke in many ways in dispraise of the Buddha, the
Dhamma, and the Saṅgha, while his own pupil, the youth Brahmadatta,
spoke in many ways in their praise.53

Although the sutta itself suggests the way the Buddha “has so thoroughly
penetrated the diversity in the dispositions of beings,” his knowledge of the
inclinations of others is not cited by Buddhaghosa as the official precipitating

51. Cattāro hi suttanikkhepā—​attajjhāsayo, parajjhāsayo, pucchāvasiko, aṭṭhuppattikoti (Sv i.50).

Buddhaghosa then gives examples of particular suttas for each of the four.
52. Sv i.51.
53. This is Bhikkhu Bodhi’s translation (The All-​Embracing Net of Views, 51–​52) of D i.2.
132 In t erpr et ing t h e T hr ee PiṬak as

cause (suttanikhepa) of the sutta; rather it is this “specific incident” of the dispar-
agement of Suppiya and the praise of Brahmadatta that got the sermon rolling.54
This is because the Buddha begins his sermon with urging his disciples to disre-
gard both disparagement and praise. The sutta then teaches a teaching on mor­
ality (as the sort of “trifling and mundane” reasons a teacher might be found
praiseworthy55) and then the long enumeration of the sixty-​two speculative views
available in the world, all of which the Buddha has transcended (and which make
his ken truly superlative, as we saw earlier).
Thus far, Buddhaghosa’s reading appears to be drawing out implications of
the sutta itself. The canonical nidāna sets out the opportunity for the Buddha
to expound the insignificance of praise and blame and ultimately to dazzle the
world with a demonstration of his knowledge of what undergirds the promotion
of sixty-​two dogmas. But as we go deeper into the commentarial nidāna—​the
flesh that Buddhaghosa puts on to the opening frame story—​we come to see how
the Buddha’s sermon is interpreted as speaking pointedly and specifically to its
context.
Buddhaghosa takes great care to fill in the particulars of the story with
Suppiya, which we can only touch on here. According to Buddhaghosa, Suppiya
was a student of Sañjaya, an important teacher who was fast losing most of his dis-
ciples to the Buddha, most notably Sāriputta and Moggallāna. If he had known
he would fall in behind the Buddha on that road, he would have found another
way to go, Buddhaghosa tells us, for he is subjected to the astounding spectacle of
the Buddha lit up in multihued rays that radiate outward eighty feet all around
illuminating the ground with sparkling light as though strewing it with gems and
gemstone dust. The Blessed One is surrounded by a huge host of most estimable
disciples, gracious and composed in manner, altogether forming a beautiful and
glorious sight the splendor of which prompts Buddhaghosa to dilate at consider-
able length with many similes from the tiny—​he was “like a filament (of a lotus
flower) surrounded by petals”—​up to the vast—​he was like the Great Brahmā
surrounded by the multitude of Brahmā deities.56 On full display is the cosmic

54. Sv i.51.
55. Morality (sīla) as defined here is a matter merely of avoiding violating the five precepts and
other problematic actions and is thus the least a teacher can do to be found praiseworthy; it
is nowhere as impressive as the knowledge of the affective underpinnings of views, which is
unique to the Tathāgata. As Buddhaghosa puts it: “sīla is said to be trifling and mundane and
so much inferior only in comparison with much higher qualities” (Evameva upari upari guṇe
upādāya heṭṭhā heṭṭhā sīlaṃ appamattakaṃ oramattakanti veditabbaṃ [Sv i.59]).
56. This paragraph paraphrases and condenses the descriptions given in Sv i.39–​40.
 Buddhavacana in the Suttanta 133

and spectacular Buddha, even as the two rival parties traverse the dusty road to
Nālandā.
Suppiya finds his own retinue uninspired and pathetic in contrast; at a
moment in which he might have praised the Blessed One, he instead is consumed
with envy. Later in the rest house that night he notices again the graceful bearing
of the Buddha’s disciples, and “appraising his own followers,” finds “some of them
tossing hands about, others [tossing] feet, some prattling nonsense, others loose-​
tongued or drooling, some gnashing teeth, others snoring.”57 Buddhaghosa says
that Suppiya voiced very specific dispraises, casting aspersions on the Buddha’s
claims of omniscience and attainment, and accusing him of violating social
norms, lacking taste, being worthless, an annihilationist, a nihilist, and so on. To
add insult to injury, Suppiya’s own disciple Brahmadatta offers up a spirited de-
fense of the Buddha, demonstrating, as Buddhaghosa sees it, a lively independ­
ence from his teacher.58
Suppiya’s dispraise is explained somewhat obliquely by the teachings of the
sutta that describe how specific dogmas lead to further distortion and unfor-
tunate results. His teacher, Sañjaya, is said elsewhere to be an “eel-​wriggler,” an
adherent of one of the sixty-​two positions the Buddha describes.59 Eel-​wrigglers
come by their endless equivocation and skepticism variously, but one way they
embrace it is simply by their stupidity,60 which is said to be how Sañjaya came by
his.61 Suppiya’s failure to recognize the Buddha’s qualities and instead to falsely
and stupidly disparage him is thus accounted for, though neither the sutta nor
Buddhaghosa is so indelicate as to point this out directly (though Dhammapāla
does not hesitate to make the connection62).
But what seems to capture Buddhaghosa’s particular notice is the way the
monks discussing Suppiya the next morning marvel that the Buddha is one

57.  Sv i.42:  Paribbājako taṃ vibhūtiṃ disvā attano parisaṃ olokesi. Tattha keci hatthaṃ
khipanti, keci pādaṃ, keci vippalapanti, keci nillālitajivhā paggharitakheḷā, dante khādantā
kākacchamānā gharugharupassāsino sayanti.
58. These details are given in Bhikkhu Bodhi’s summary translation of the commentary, The
All-​Embracing Net of Views, 90–​94.
59. Bodhi, The All-​Embracing Net of Views, 170.
60. D i.26–​29.
61. The sutta immediately following the Brahmajāla, the Sāmaññaphala (D i.59–​60) describes
Sañjaya’s equivocations and evasions as a result of his being the “most stupid and confused” of
all ascetics and Brahmins in King Ajātasattu’s estimation.
62. Dīgha ṭīkā i.215.
134 In t erpr et ing t h e T hr ee PiṬak as

“who knows and sees.” This means that “he knows the biases and inclinations
of each and every being” and “he sees every knowable phenomenon as if it were
an āmalaka fruit held in the palm of his hand.”63 This inspires a list of features
of the Buddha’s omniscience, including his knowledge of all beings’ past lives,
the three knowledges, and other stock descriptions of omniscience. Buddhaghosa
also notes that the Buddha knows the very particular causal conditions that lead
beings to encounter one another, a knowledge quite astonishing given the diver-
sity of beings and how they may come into connection with each other.

In this way the Buddha knows, and knows thoroughly, with his om-
niscient knowledge the various aspirations, dispositions, undertakings,
desires, and wishes of beings as though they were lengths [known] with
a measure, or weights [known] with a scale. For it is hard to find even
two beings with the same disposition in this world. If one wants to
go, the other wants to stay, and if one wants to drink, the other wants
to eat.64

Buddhaghosa is emphatic that human beings are everywhere unique: elsewhere

he insists that even twins can be distinguished from each other.65 The range of
human difference is vast and immeasurable, but the Buddha, upon encountering a
person, knows this range and the inclinations—​their quirks, their idiosyncrasies,
their hopes. It is at this that the disciples in the sutta are marveling when they
praise the Buddha for “knowing and seeing” and penetrating “the diversity in the
dispositions of beings.” What is even more astonishing, Buddhaghosa suggests,
is that he can see the connections and shared elements that bring such various
beings together. He can understand the conditions behind the encounter with
Suppiya.
But the precipitating cause of this display of the Buddha’s omniscience
for Buddhaghosa is the “specific incident,” not the knowledge of inclinations.
The incident is the praise and blame and the ensuing discussion of the monks
about these the next morning. The sutta says that although “he knew what their
discussion was about,” the Buddha asked his monks what they were talking

63.  Sv i.43:  tena bhagavatā tesaṃ tesaṃ sattānaṃ āsayānusayaṃ jānatā, hatthatale ṭhapitaṃ
āmalakaṃ viya sabbañeyyadhammaṃ passatā.
64. Sv i.44: evaṃ sattānaṃ nānādhimuttikatā, nānajjhāsayatā, nānādiṭṭhikatā, nānākhantitā,
nānārucitā, nāḷiyā minantena viya tulāya tulayantena viya ca nānādhimuttikatāñāṇena
sabbaññutaññāṇena viditā, sā yāva suppaṭividitā. Dvepi nāma sattā ekajjhāsayā dullabhā
lokasmiṃ. Ekasmiṃ gantukāme eko ṭhātukāmo hoti, ekasmiṃ pivitukāme eko bhuñjitukāmo.
65. Sv ii.509.
 Buddhavacana in the Suttanta 135

about.66 (This is a common theme:  the Buddha asks what people are talking
about, but of course he already knows.) And in fact, the Buddha approaches the
monks because he realized that, though they were discussing his omniscience,
the “workings of omniscience are not clear to them,” and so he can now demon-
strate them.67 Omniscience is something that has to be demonstrated through
its workings and applications, and this is an opportune moment for doing so.
At this point, Buddhaghosa launches into a very specific account of the exact
happenings the night before, when it occurred to the Buddha what the disci-
ples would be discussing the following morning. Buddhaghosa goes on for sev-
eral pages describing the Buddha’s daily routine, where each day is broken into
five parts, and his activities during the morning, afternoon, and nighttime (itself
divided into three watches).68 We learn when he ate, went for alms, rested, re-
ceived offerings of reverence, held audiences, taught the Dhamma, had his feet
washed, bathed, and so on. Intriguingly the Buddha never sleeps even at night,
but at certain junctures he lies on his right side, fully aware, and rests to refresh his
body. But why is any of this relevant? Because Buddhaghosa is curious about the
exact moment in his daily routine that the Buddha came to know, with his omnis-
cience, what the disciples were talking about. In fact, this occurred in the very last
watch of the night, while he was walking up and down a path that was part of his
usual routine of pacing back and forth to work out the aches of the day. Having
discerned what they are talking about, he resolves to give the “lion’s roar” of the
Brahmajāla Sutta, as though lifting up Mount Sumeru itself.69
This demonstration of omniscience is concerned to situate it in the eve-
ryday life of the Buddha, the “specific incident” that prompts it. The context is
maximized through several pages of exacting detail to prepare the reader to receive
the mir­aculous nature of the occasion of this omniscient knowing. It is all very well,
Buddhaghosa seems to suggest, to claim omniscience for the Buddha, but what
did this kind of knowing actually look like in the course of the Buddha’s workaday
life? What was he actually doing when this knowledge hit him? Buddhaghosa
makes it immediate and real through his supplementary nidāna with its specificity.
We can leave the story here, but before doing so it is worth recapping the
various ways that the nidāna furthers the canonical sutta’s demonstration or

66. D i.2: bhagavā tesaṃ bhikkhūnaṃ imaṃ saṅkhiyadhammaṃ viditvā. The commentary says
that he knew this by his omniscience (Sv i.44: ettha viditvāti sabbaññutaññāṇena jānitvā).
67.  “These monks are discussing my qualities beginning with the knowledge of omnis-
cience. But the workings of the knowledge of omniscience are not clear to them as they
are to me” (ime bhikkhū mayhaṃ sabbaññutaññāṇaṃ ārabbha guṇaṃ kathenti, etesañca
sabbaññutaññāṇakiccaṃ na pākaṭaṃ, mayhameva pākaṭaṃ[ Sv i.48]).
68. Sv i.44–​48. See Bodhi, The All-​Embracing Net of Views, 97–​101.
69. Sv i.48.
136 In t erpr et ing t h e T hr ee PiṬak as

enactment of omniscience. In Buddhaghosa’s nidāna, not only did the Buddha

enact his omniscience in describing the sixty-​two views but also he displayed it
by knowing what his disciples were talking about, knowing the dispositions of
Suppiya and the immediate and distant conditions that produced this encounter,
and knowing what sort of teaching would benefit this particular situation.
Suppiya’s intransigence as a member of a rival sect holding views is revealed to be
the result of affective experience, a central teaching of the sutta. These narrative
particulars show the immediacy of how the Buddha’s knowing works, and in
effect, enact or perform the doctrinal teaching of the sutta.
As suggested earlier, the spectacular, cosmic Buddha is seamlessly folded into
the everydayness of the Buddha in his daily routine responding off-​the-​cuff to the
exigencies of the events and people around him. Walters suggests that there is a
docetic quality here, but I wonder if in this case, as elsewhere, the Christian origin
of such terms is misleading. In certain Christian contexts, the idea of docetism
is used to deny the reality of Christ’s physical body, as his cosmic principle only
took on the appearance of being human. There are no claims in Buddhaghosa’s
thought about what is real or unreal, and the “cosmic” and “everyday” Buddhas
(which are my distinctions) are not pitched in opposition to one another. Rather,
the Buddha who spoke to his disciples about Suppiya that morning is the same
cosmic Buddha whose body emitted six-​colored rays of radiance extending eighty
cubits all around.

The Mūlapariyāya Sutta: Knowing the Roots

I turn now to the Mūlapariyāya Sutta, the first sutta in the Majjhima Nikāya,
which provides an opportunity to look at the reception of a sermon. The
sutta does not seem to have much by way of a nidāna, so it might seem like an
odd place to make my argument. The sermon opens simply with the Buddha
addressing “the monks while dwelling at Ukkaṭṭḥa at the foot of a royal Sālā tree
in the Subhaga Grove.” He announces that he will teach them “a teaching on
the roots,” and proceeds to do so.70 What follows is a quite subtle discourse on
how ordinary beings, learners, arhats, and Tathāgatas perceive and conceive of
the elements and the various realms of beings. We get no further hints as to the
narrative frame of this teaching until the closing line of the sutta, when we are
told that upon delivery of the sermon “the monks did not delight in what was
said by the Bhagavan.”71 This startling claim contrasts pointedly with the usual

70. M i.1.
71. M i.6. I am grateful to my student, Hope Wen, for drawing my notice to this passage.
 Buddhavacana in the Suttanta 137

formulaic rejoicing at his words ending many suttas. We might note the impact
this device has on the reader: there appears nothing in the sutta at first reading
that would seem to provoke dismay. But everything changes in our reading when
we learn that the sermon met with displeasure; we find we must go back to the
sutta and look again. What about the Buddha’s words to these monks triggers
this response?
The sutta itself does not tell us, but Buddhaghosa does. He of course notices the
dismay in which the teaching was received and provides ample context and explica-
tion to make sense of it. He says that of the four reasons prompting a sutta, this one
(like the Brahmajāla) was occasioned by “a specific incident.” This was the “arising
of conceit due to learning”;72 in his telling, the monks addressed by the sutta were a
group of five hundred former Brahmins who had mastered the three Vedas; as monks,
they mastered the entire word of the Buddha, and had thus grown conceited.73 They
assured themselves that because they understood the language and grammar of the
Buddha’s teachings, there was nothing confusing in them that they did not grasp.
Due to this conceit they failed to listen and attend to the Buddha properly, and he,
“knowing the course of their thoughts,” decided to teach the Mūlapariyāya Sutta to
break down their conceit.74
The teaching he gave them shows the very subtle distortions that occur even
at the level of perception in how an “ordinary person” perceives experience. The
Buddha describes how the ordinary person, the learner, the arhat, and the Tathāgata
each perceive and conceive of the twenty “bases” or fields of experience. For example,
the first four of these twenty-​four bases are the primary elements, earth, water, fire,
and air, where the earth element, as Buddhaghosa explains it, can be perceived as
the formation of one’s body or other physical things in the world. For the ordinary
person, the distortive lens of conceptuality alters the experience of earth. According
to the sutta, the ordinary person

cognizes earth as earth. Having cognized earth as earth, he conceives

(himself as) earth; he conceives (himself ) in earth; he conceives (himself )

72.  Ps i.16:  Evamimesu catūsu nikkhepesu imassa suttassa aṭṭhuppattiko nikkhepo.

Aṭṭhuppattiyañhi idaṃ bhagavatā nikkhittaṃ. Katarāya aṭṭhuppattiyā? Pariyattiṃ nissāya
uppanne mane.
73.  Ps i.16. For a translation of this sutta, and (most of ) its two layers of commentary,
see Bodhi, trans., The Discourse on the Root of Existence:  The Mūlapariyāyasutta and Its
Commentaries, 36–​37.
74.  Ps i.17:  Bhagavā tesaṃ taṃ cittacāraṃ ñatvā  .  .  . bhagavā mānabhañjanatthaṃ
sabbadhammamūlapariyāyanti desanaṃ ārabhi.
138 In t erpr et ing t h e T hr ee PiṬak as

apart from earth; he conceives ‘earth is mine’; he delights in earth. What

is the cause? I say that it is because it has not been understood by him.75

As the modern interpreter Bhikkhu Bodhi explains, this describes the subtle ways
conceiving “turns out to be a double process of identification and appropriation”
as the object of experience is refracted and distorted through the self. Since the
self “lacks foundation” the resulting conception is false and confused, bringing
disappointment and suffering in its wake.76 Buddhaghosa explains that the
distortions occurring at this level of cognizing and conceiving are the result of the
“craving, conceit, and wrong views”77 that proliferate in the ordinary person and
inflect our most basic apprehension of the world. The rest of the sutta applies this
formula to the entire field of experience as it is divided into twenty-​four different
bases. And it contrasts this basic distortion in the ordinary person with the expe-
rience of the learner, the arhat, and the Tathāgatha, all of whom, the text claims,
“directly know” (rather than cognize) the object of experience, and the latter two
of whom are free entirely of any distorting tendencies.
What was it about this message that was so dismaying to the five hundred
learned monks? Buddhaghosa says that the sutta was spoken illustrating the
Buddha’s “supremely deep knowledge of omniscience which does not yield a foot-
hold to the wisdom of others.”78 Though the teaching had great “eloquence and
various methods,” and even though the Buddha had fulfilled the perfections for
four immeasurable and a hundred thousand aeons to obtain omniscience pre-
cisely so that he could explain to others the Dhamma in a way they could grasp it,
these monks did not understand this teaching.79 This was due to the conceit that
obstructed the capacity for them to understand a teaching about the roots of their
own experience. Their extensive erudition generated the conceit that blocked

75.  M i.1:  pathaviṃ pathavito sañjānāti; pathaviṃ pathavito saññatvā pathaviṃ maññati,
pathaviyā maññati, pathavito maññati, pathaviṃ meti maññati, pathaviṃ abhinandati. Taṃ
kissa hetu? ‘Apariññātaṃ tassā’ti vadāmi. See also Bodhi, trans., The Discourse on the Root of
Existence, 27 (M i.1). I prefer “cognize” and “cognition” for sañjānāti and saññā, though they
are often translated as “perceiving” and “perception,” because of the recognizing and naming
that goes on in this cognitive process.
76. Bodhi, trans., The Discourse on the Root of Existence, 12.
77.  Ps i.25:  taṇhāmānadiṭṭhipapañcehi. See Bodhi, trans., The Discourse on the Root of
Existence, 47–​48.
78. This is Bhikkhu Bodhi’s translation, The Discourse on the Root of Existence, 82 (Ps i.55).
79. Ps i.56: vicitranayadesanāvilāsayuttampi. . . . Nanu ca bhagavā attanā desitaṃ dhammaṃ
pare ñāpetuṃ kappasatasahassādhikāni cattāri asaṅkhyeyyāni pāramiyo pūretvā sabbaññutaṃ
patto. So kasmā yathā te na jānanti, tathā desesīti.
 Buddhavacana in the Suttanta 139

their capacity to understand, and so the sutta put before them was like “permitted
food placed before someone whose mouth was bound with a thick, wide cloth.”80
The Buddha knew this would happen and that this experience would shatter their
conceit, and so he taught it for this “special incident.” And indeed, the monks are
horrified at their failure to understand, and they slink away. As for us, the readers
(as coached by Buddhaghosa), we are in a position to simultaneously grasp the
philosophical point of the sutta (about the subtle but pervasive distortions—​like
conceit—​that confound our grasp of our own experience) and to see it enacted in
the so-​called learned monks, whose conceit reveals them to be the very ordinary
persons described in the sutta. The frame story performs the sermon, as it were.
Buddhaghosa goes further to explain context. The episode, he says, is widely
discussed among other monks, who in turn prompt the Buddha to describe
“the story of the past” that relates the back story of these monks. It seems that
these five hundred monks were, in a previous life, Brahmin youths who similarly
tripped over their conceit in their inability to solve a riddle posed by their teacher
(who was of course the Bodhisatta in that previous life). Buddhaghosa also tells
a story of the future of these monks, where after having been humbled by this
occasion described in the Mūlapariyāya Sutta, they are at a later time open to the
Buddha’s teaching in a new way and are given the Gotamaka Sutta, a teaching
that pleases them greatly, and they become arhats!81 This sequel sutta describes
the Buddha as teaching via “direct knowledge” and instructs its audience to re-
joice, be delighted, and full of joy.82 In Buddhaghosa’s reading, the Gotamaka
Sutta in this way recalls the Mūlapariyāya Sutta and speaks to it, offering an ulti-
mately successful conclusion to its disquieting ending.
This rather grand edifice of context—​none of it present in the actual sutta
itself—​depicts the sutta as a calculated intervention rather than (just) a decla-
ration of doctrine. It is an intervention located in an expansive narrative of time
stretching from a distant past life of these monks to the future trajectory of their
awakening. The context also includes another sutta, revealing a larger intertextual
“whole” of which this sutta is just one part. Above all, as Buddhaghosa reminds
us throughout his interpretation, it is an illustration of the keen subtlety of the
workings of the Buddha’s omniscience. The genius of the event of the sutta is
that it once again displays the Buddha’s unimpeded knowledge of particulars,

80.  Ps i.56:  etaṃ suttaṃ ghanaputhulena dussapaṭṭena mukhe bandhaṃ katvā purato
ṭhapitamanuññabhojanaṃ viya ahosi.
81. Ps i.59.
82.  A  i.276. See Bodhi, trans. The Numerical Discourses of the Buddha:  A Translation of the
Aṅguttara Nikāya (Boston: Wisdom Publications, 2012): 355–​56.
140 In t erpr et ing t h e T hr ee PiṬak as

the previous lives, the inclinations, the intellectual and emotional needs, and the
future trajectories of the particular individuals whom the Buddha encounters.
The contents of the sutta—​how the intricate roots of craving, confused view, and
most palpably here, conceit, distort experience—​is enacted explicitly by the very
monks to whom it is delivered. The conceit that confounds their understanding
is made apparent to them by a teaching that they think they already understand.
We would of course miss all of this if we did the usual modern practice of noting
only the fractured text in focusing on the sermon apart from its nidāna or its
commentary.
Bhikkhu Bodhi offers a somewhat different interpretation than
Buddhaghosa does, equally fascinating, that in fact it is not that the monks
did not understand the teaching, but that “they understood it too well.” What
displeased them was the way the sutta dismantles the self and shows the roots
of experience as lacking this supposed foundation. He cites a passage from the
Bṛhadāraṇyaka Upaniṣad that offers a striking parallel to this sutta, though of
course locating the foundation of experience in the self.83 Also interpreting
the Mūlapariyāya in terms of the broader intellectual context of ancient
India, Thanissaro Bhikkhu argues that this sutta is meant to respond to the
Sāṃkhya system of metaphysics.84 We need not decide among these various
interpretations. I note only the contrast in which these modern historicist her-
meneutics go outside the Pali textual world for a context to “explain” the sutta,
where Buddhaghosa’s interpretive context is located entirely within the Pali
imaginaire.
The literary imagination that takes in what the sermon does to the five hun-
dred monks to whom it is addressed does something interesting also to the
reader: to appreciate how and why these monks are upset by the sermon requires
a philosophical engagement with the nuances of the teaching, as well as with
these particular people. The narrative reception of a sutta is a “cue” for readerly
response. At one level when understood in its grander narrative frame, the monks
come off as rather ridiculous—​the usual motif of the arrogant Brahmins whose
conceit confounds their understanding of Buddhism’s deeper truths. But the
reader who had to “look again” at the sutta to figure out what about it was offen-
sive to them even with the aid of the larger narrative about the conceit of ordi-
nary persons, realizes that perhaps she, the reader, did not get the subtlety and,
more importantly, the existential significance of the sutta either at first reading.
In so far as the reception of a sutta is a cue for the reader’s experience, then the

83. Bodhi, trans., The Discourse on the Root of Existence, 24–​25.

84. Thanissaro Bhikkhu, “Mulapariyaya Sutta: The Root Sequence.”
 Buddhavacana in the Suttanta 141

disquiet engendered by the sutta, and the story, can prompt a turn inward. Nor
are we permitted, within Buddhaghosa’s narrative of it, to remain self-​satisfied
and smug about these former Brahmin monks with all their superior “learning”
getting humbled, because of the future: these monks are conditioned by this sutta
to attain insight by a future intervention, the Gotamaka Sutta, where their turna-
round is a matter for rejoicing. There is something deeply humane in this literary
reading whereby readers come to see how others’ (and perhaps their own) conceit
can be dismantled in a way that leads eventually to future triumph.

Conclusions
The foregoing has suggested that for Buddhaghosa the nidāna is an interpretative
method for going deeper into the Buddha’s omniscience. We have attended to the
“concentric nests” that A. K. Ramanujan argues house many Indian conceptions,
and can observe with him that a teaching’s “poignancy is partly in its frame.”85
The frame stories demonstrate the particular capacity of the Buddha’s omniscient
mind that speaks from and to individual contexts in their potentially infinite
particulars and singularity. In the Suttanta, philosophy is a dialogical practice
built into the biographical narratives of the Buddha and the cast of characters
that the texts and commentaries portray in rich literary detail.
Buddhaghosa’s attention to the nidāna can show concretely the manner in
which the Dhamma spoken by the Buddha is always “well-​spoken.” The narrative
specificity of the contextual frame performs what it means to say that the Buddha’s
well-​spoken utterances are “visible here and now, timeless, inviting one to come
and see, leading forward, and to be experienced by the wise for themselves.”86 His
appreciation of how suttas are “visible here and now” suggests a literary sensi-
bility in which the specificity of everyday life becomes a method to glimpse the
workings of the Buddha’s omniscient mind. We find in this specificity and singu-
larity particular instances of the unobstructed nature of his knowledge of beings.
Buddhaghosa is interested in the aesthetic power of the narrative frame because
he sees how it can evoke the deeper and richer poignancy of the teaching that
aims to be existentially transformative to the ideal reader.
The suttas considered here give teachings as interventions in dialogical and
pedagogical contexts. Philosophy can be a “way of life” as well as disembodied
treatises; the recent turn initiated by Pierre Hadot in exploring how the ancients

85. Ramanujan, “Is There an Indian Way of Thinking? An Informal Essay,” 51, 49.

86. Vism 213 on M i.37; A iii.285: sandiṭṭhiko akāliko ehipassiko opaneyyiko paccattaṃ veditabbo
viññūhī. See Appendix A.
142 In t erpr et ing t h e T hr ee PiṬak as

in the West saw philosophy as inseparable from actual life makes the relevance
of biography and narrative obvious.87 The frame stories of the suttas are bio-
graphical: they recount episodes in the Buddha’s long teaching career as well as
interventions in particular peoples’ lives. Philosophy is a discursive practice to be
engaged in with others. And the Buddha is revealed as the much acclaimed “good
friend” whose pedagogical friendship is highly extolled throughout the texts.
As we have seen, the Brahmajāla Sutta gives a sermon about the Buddha’s
views of views; in its own terms the content of the sermon demonstrates the
Buddha’s capacity to survey all views that there are and to see how the advancing
of metaphysical positions is conditioned by feelings, and how it leads to further
entrapment in saṃsāra. Adhering to dogmatic views has causes and conditions
and leads to specific consequences in this life and the next. This general insight
is driven home by the story of Suppiya in which the sermon is nested: Suppiya’s
confusion about the Buddha is rooted in a set of conditions that generate a series
of predictable outcomes explained by the sutta and enacted by the frame story.
Both the sermon and the specific instance to which it speaks demonstrate what
it is that the Buddha “knows and sees.” He sees the long arc of experience and the
scope of views that are conditioned by recurrent patterns of feeling. His interven-
tion to his disciples affords them—​and the reader—​a glimpse of the kind of om-
niscient mind that can see both the particular patterns of views and the universal
patterns in how views are formed and held.
The Mūlapariyāya Sutta is, in Buddhaghosa’s reading, perhaps an even more
dramatic intervention, at least in the lives of those to whom it is addressed. It
demonstrates that study is not always adequate to the Buddhist life in that the ar-
rogant Brahmins needed the Buddha’s skillful and highly unpleasant dismantling
of their conceit. Conceit so deeply hidden and so long habituated is obscure
to them and can be confronted and addressed only through a discomfiting en-
counter with the Buddha. We can see here what Richard Shusterman describes
as the pitfalls of solitary introspection where “one’s view of oneself is always par-
tial in both sense of ‘biased’ and ‘incomplete.’ ”88 He elaborates: “the depths of
one’s soul, the complex layers, quirks, and weaknesses of one’s personality are
hardly transparent to one’s own consciousness either because they are implicitly
repressed or because, as part of one’s second nature, they are so close that they
escape attention.” What is needed is to be addressed by others, a philosophical

87. Hadot, Philosophy as a Way of Life. See also Shusterman, “Philosophy as a Way of Life: As
Textual and More Than Textual Practice,” and Heim, “The “Fecundity of Dialogue’ and the
‘Philosophy of Incompletion.’ ”
88. Shusterman, “Philosophy as a Way of Life,” 49.
 Buddhavacana in the Suttanta 143

requirement the Pali tradition refers to as “voice of another” essential to the

disciplined life.89 And since the Buddha’s mind is not partial or incomplete, as
Buddhaghosa tirelessly reminds us, there is no other quite like the Buddha in how
he can reveal those hidden depths and their conditions in previous lives.
These thoughts suggest that the idea of “context” (nidāna also means “or-
igin”) as a hermeneutical principle reveals a, or perhaps the, central teaching of
Buddhism, which is the radical conditionality of experience. One of the things
that I  have learned is that there is always more context that could be added.
Buddhaghosa often notes that he is conveying “in brief ” what could be said “in
detail,” that there is more to say and that he must be selective. This in and of it-
self is an essential philosophical point. The narratives make clear that the condi-
tionality of a particular event goes back an awfully long time—​people are always
caught in the middle of their lives and what happened before their encounter with
the Buddha and what happens next shape how we interpret what the encounter
itself is. What goes into a philosophical moment or a pedagogical encounter is
a complex web of conditions and relationality, which is being performed by the
narratives. The whole of this complex network is what the Buddha’s omniscient
mind—​when he attends to an encounter directly—​knows without obstacle. And
it is what we come to see through textual exegesis that shows this omniscience
in action and thereby, for Buddhaghosa, orients the ideal reader to the Buddha.

89.  For example, the “voice of another” (parato ghosa), along with proper attention (yoniso
manasikāra) is said to be a condition for right view (sammādiṭṭhi) (Mi.294; Ai.87).
 4

Disentangling the Tangle
Abhidhamma as Phenomenological Analysis

Buddhaghosa begins his Visuddhimagga by describing how its three-

fold structure—​morality (sīla), concentration (samādhi), and understanding
(paññā)—​is the key to disentangling a tangle. It seems that a deity once presented
the Buddha with this question:

The inner tangle and the outer tangle, beings are entangled in a tangle.
So I ask Gotama: who disentangles the tangle?1

Buddhaghosa explains that this is referring to the web of craving that entangles
us all, like a network of branches in bamboo thickets. We are ensnared in a net-
work of desire that keeps us trapped. The Buddha’s reply is the inspiration for the
Visuddhimagga:

When a wise man comes to be established in morality

and develops awareness and understanding,
then that active and intelligent monk disentangles the tangle.2

The verse suggests that morality is a foundation for the cultivation and devel-
opment of awareness and understanding, which Buddhaghosa takes to be a

1.  Vism 1 (I.1):  Anto jaṭā bahi jaṭā, jaṭaya jaṭitā pajā, taṃ taṃ gotama pucchāmi, ko imaṃ
vijaṭaye jaṭanti. This is quoting S i.13, which gives the two verses on which the Visuddhimagga
is the commentary.
2. Vism 1 (I.3): Sīle patiṭṭhāya naro sapañño, cittaṃ paññañca bhāvayaṃ; Ātāpī nipako bhikkhu,
so imaṃ vijaṭaye jaṭanti. Buddhaghosa takes “developing awareness” (cittaṃ bhāvayam) to be
samādhi.
 Disentangling the Tangle 145

threefold structure or sequence of practices. His nearly eight-​hundred-​page long

commentary on this verse, the Path of Purification, is offered as the way to disen-
tangle oneself from craving and its conditions: it is a path starting with morality
that develops awareness through concentration techniques, and culminates with
practices for understanding. The “purification” in the title refers to nibbāna, and
the “path” refers to the means of approaching it.3 Put another way, the entire path
leads to the culmination of purification of “knowing and seeing,”4 which is the
final section of the text.
These opening considerations about what the Visuddhimagga is frames the
text within a metaphor of disentangling that I think is helpful for understanding
Buddhaghosa’s approach to Abhidhamma practices of knowledge, practices which
he discusses and extends in the Visuddhimagga. Our experience is entangled, but
the Abhidhamma texts untangle the thickets to disaggregate and reorder the phe-
nomena of experience to help us understand and see the conditional relations
that bind us. This chapter explores Buddhaghosa’s theory of Abhidhamma textu-
ality as he articulates it in both his commentaries on the Abhidhamma texts5 and
the Visuddhimagga.

3.  “Therein purification should be understood as nibbāna, entirely purified and free of
all stains. The Visuddhimagga is the path to that purification. The path is the means of
approaching [it].” Vism 2 (I.5): Tattha visuddhīti sabbamalavirahitaṃ accantaparisuddhaṃ
nibbānaṃ veditabbaṃ. Tassā visuddhiyā maggoti visuddhimaggo. Maggoti adhigamūpāyo
vuccati.
4.  Another schema for the organization of the Visuddhimagga, though not one
that Buddhaghosa discusses explicitly, is the sevenfold purification described in the
Ratnavinīta Sutta (M i.145):  purifying morality (sīlavisuddha), purifying awareness
(cittavisuddha), purifying view (diṭṭhivisuddha), purification by overcoming doubts
(kaṅkhāvitaraṇavisuddha), purification by knowing and seeing what is and is not the path
(maggāmaggañāṇadassanavisuddha), purification of knowing and seeing the practices
(paṭipadāñāṇadassanavisuddha), and purification of knowing and seeing (ñāṇadassanavi
suddha).
5.  There is a lively scholarly question about Buddhaghosa’s direct involvement with the
Abhidhamma commentaries, though each of the three of them (the Atthasālinī, the
Sammohavinodanī, and the Pañcapakaraṇa-​aṭṭhakatha, which comments on the last five
books), mentions Buddhaghosa in the colophon. For the most recent discussions of au-
thorship of the Abhidhamma commentaries, see Cousins, “The Case of the Abhidhamma
Commentary,” and von Hinüber, “Building the Theravāda Commentaries: Buddhaghosa and
Dhammapāla as Authors, Compilers, Redactors, Editors, and Critics.” Even if he did not per-
sonally author or transmit each of the commentaries attributed to him, we can use the name
“Buddhaghosa” to refer, as Lance Cousins does, to the “school of Buddhaghosa,” as he likely
headed a team of scholars working on this material (392).
146 In t erpr et ing t h e T hr ee PiṬak as

Interpreting the Abhidhamma
It has not always been obvious to modern scholars what the formidable corpus of
Abhidhamma texts is trying to do. The seven canonical books of the Abhidhamma
are varied in their nature and purposes. The first book, the Dhammasaṅgaṇī
(Enumeration of Dhammas) consists of many long lists of phenomena classified
variously into groupings (pairs and threes). The second, the Vibhaṅga (Analysis)
is composed of eighteen chapters that provide various analytical techniques of key
technical and modular terms as they occur in specific formulas (the aggregates,
the truths, dependent origination, and so on). Further analytical descriptions and
methods (nayas) of questioning these terms and formulas are developed in the
catechisms, the Dhātukathā (Discourse on Elements), Puggalapaññatti (Describing
Persons), and the Yamaka (Pairs). The Kathāvatthu (Issues for Discussion), a text
attributed by Buddhaghosa to the teacher Moggaliputta Tissa, is a discussion and
refutation of other Buddhist schools. Finally, the “oceanic” Paṭṭhāna (Starting
Points) explores causes and conditions through various algorithmic methods that
can endlessly ramify.
While this description of the Abhidhamma books begins to name their
contents (and other scholars offer more thorough accounts that I  need not
reiterate here6), we can begin by considering several ways that this material has
been interpreted. How are we to understand what these nearly endless lists and
classifications of phenomena are doing? What sort of philosophy is this? Modern
scholars offer competing accounts of the philosophical significance and purpose
of the canonical Abhidhamma texts; considering them can help us to specify
Buddhaghosa’s distinctive thinking about them and how his understanding of
them guided his own textual practice.
One line of thought treats these texts as offering metaphysical or onto-
logical systems. In this view, the Abhidhamma lists are itemizations of the ul-
timate realities in the world. Often little distinction is made between the Pali
Abhidhamma and the north Indian traditions of Abhidharma on this point. To
cite just one typical example, a 2013 article on Buddhist metaphysics asserts that
all Abhidharma traditions (including the Pali Abhidhamma) were united by a
“common core of philosophical principles,” which includes an effort to iden-
tify “primary existent objects” as “ultimately real” and to identify these ultimate
reals as things that have “an intrinsic nature” that is, “properties [they have] inde-
pendent of anything else,” which “exist no matter what, without depending on the

6. Nyanatiloka Mahāthera’s Guide through the Abhidhamma Piṭaka is a useful synopsis of the
Abhidhamma canonical texts.
 Disentangling the Tangle 147

existence of any other dharma or on any conceptualizing mind.”7 Abhidharma

lists are catalogs of real entities, and thus they constitute ontologies. This meta-
physical reading sounds more like the target of the Madhyamaka critique of the
Abhidharma systems than it does many actual Abhidharma texts; certainly, as far
as the Pali Abhidhamma is concerned, as I show in what follows, this account is
antagonistic in every respect to the way Buddhaghosa understood his tradition.
A quite different interpretation of the Pali Abhidhamma has been advanced
by Nyanaponika Thera, who suggests a “phenomenological” reading of the canon-
ical tradition that he contrasts sharply with the notion that Abhidhamma is doing
ontology. In certain respects, Nyanaponika’s interpretation of Abhidhamma can­
onical texts may be seen as similar to Buddhaghosa’s interpretation of them and
to his own practices. Nyanaponika sees the Abhidhamma as a twofold method
or procedure that enacts both “analysis” and “investigating the relations (or the
conditionality) of things,” though he notes that the second is sometimes not fully
understood by modern interpreters.8 He argues that it is “phenomenology” in
the sense of dealing “as the name implies, with ‘phenomena,’ that is, the world
of internal and external experience,” in contrast to ontology, or metaphysics,
which “inquires into the existence and nature of an essence, or ultimate prin-
ciple, underlying the phenomenal world.”9 He shows how the Abhidhamma
methods of reductive analysis (most obvious in the Dhammasaṅgaṇī) and

7. Westerhoff, “Metaphysical Issues in Indian Buddhist Thought,” 130. It is not difficult to find
scholars who share this perspective. Siderits also speaks of a single ahistorical Abhidharma tra-
dition with no acknowledgment of the possibility that the Pali texts might depart from the
general picture provided by Vasubandhu, and he portrays the whole tradition as a reductive
metaphysics (Buddhism as Philosophy, ch. 6). Griffiths equates “abhidharma” with “meta-
physics,” even when discussing the Milindapañha (Religious Reading, 116). Despite his insist-
ence on tracking Indian Buddhist textualities, his book never even mentions the huge Pali
commentarial project as a distinct tradition; it is unclear whether it is included in, adjacent
to, or entirely irrelevant to his treatment of “Indian Buddhism.” Another example of the re-
ceived view: the entry for “Abhidharma” in Buswell and Lopez, eds., The Princeton Dictionary
of Buddhism, 4 (in which Abhidhamma is lumped in and gets no separate treatment): “the
abhidharma provided an objective, impersonal, and highly technical description of the specific
characteristics of reality and the causal processes governing production and cessation.” We look
at the treatments of certain Pali scholars below.
8. Nyanaponika Thera, Abhidhamma Studies, 21.
9.  Nyanaponika Thera, Abhidhamma Studies, 19. There is a rather muted but still long-​
standing scholarly tradition of asserting that the Pali Abhidhamma tradition at the canon-
ical level was phenomenological rather than metaphysical in orientation; both Karunadasa,
The Dhamma Theory:  Philosophical Cornerstone of the Abhidhamma, 8, and Ronkin, Early
Buddhist Metaphysics, ch. 3, argue this, though they suggest that there was a metaphysical shift
at the postcanonical layer. Going back even further historically, Morris cites approvingly Rhys
Davids’s insistence that early Buddhism, including the Abhidhamma texts, was not engaged in
metaphysics (Morris, ed., The Puggala-​Paññatti, viii, citing Rhys Davids’s Hibbert Lectures).
148 In t erpr et ing t h e T hr ee PiṬak as

classificatory explorations of conditionality (most evident in the Paṭṭḥāna) de-

scribe a complex network of conditionality, within which nothing can be seen as
independently real.
Nyanaponika’s account of the Abhidhamma as a “method” accords with
Buddhaghosa’s interpretation of this genre as a method of analytical practice. We
can helpfully say that for Buddhaghosa the Abhidhamma texts can form a kind
of “phenomenological” practice, concerned with describing the phenomena of
felt experience, and are not concerned to argue for ultimate essences or features
of reality. For him, Abhidhamma texts and techniques explore how things in
experience come about, are presented to us, and fade away over time. They elab-
orate the processes of breaking apart experiences into their constitutive elements
in a given context or “occasion” (samaya), as the Dhammasaṅgaṇī puts it, to
make it possible to see how these elements are conditioned by and condition
one another.10 The patterns of conditionality explaining and describing these
relations are the central teachings of dependent origination, name-​and-​form,
the five aggregates, the four truths, and so on. Such teachings are not, in this
reading, final assertions of the way things are, but analytical practices to explore
and describe experience.
Where I  think Buddhaghosa differs from Nyanaponika Thera is that the
latter occasionally slides back into ontology. That is, while still rejecting an idea
of Abhidhamma as a realist ontology, he does think phenomenological practice
serves, ultimately, ontological understanding. For example, he suggests that the
methods and procedures of the Abhidhamma culminate in “a definite and val-
uable contribution to ontological problems, that is, to the search for an abiding
essence in reality.”11 In this reading, the phenomenal methods and formulas of
the Abhidhamma teachings issue in a metaphysical truth: there are no essences
in reality. This is a plausible account of the canonical Abhidhamma; perhaps the
canonical texts are attempting to establish a metaphysics of experience even if
not of a reality beyond it. And while by no means necessary, this slide from phe-
nomenological practice to metaphysics seems difficult to forestall, as indeed the

10.  The Dhammasaṅgaṇī describes each of its many listings of phenomena (dhammas) as
occurring on an occasion, a conjunction of circumstances, which is the moment being analyzed,
as Nyanaponika Thera notes (Abhidhamma Studies, 6–​7).
11.  Nyanaponika Thera, Abhidhamma Studies, 20–​ 21. Bhikkhu Bodhi’s introduction to
Nyanaponika’s book even more forcefully works this phenomenological approach into a met-
aphysical position. In contrast, Gethin resists the metaphysical slide in his reading the formula
of the five khandhas as exploring “how the world is experienced” in a manner “not primarily
as having metaphysical significance” (Gethin, “The Five Khandhas:  Their Treatment in the
Nikāyas and Early Abhidhamma,” 50).
 Disentangling the Tangle 149

Abhidharma and the later Abhidhamma traditions do seem to head, ultimately,

in a metaphysical direction. Over time the Abhidhamma tradition seems some-
times to have been interpreted as an ontology of dhammas, at least insofar as
may be suggested by the way the medieval Abhidhammatthasaṅgaha presents
Abhidhamma material and has been read.
But my interest is centered on Buddhaghosa and his intellectual labors (which
have received surprisingly little systematic attention on this question). His work
allows us to ask whether it might be possible to engage only in phenomenal ana-
lytical exploration without ever making such exploration work to reveal or posit
a metaphysical position. Can we practice phenomenological methods without
assuming that their purpose is to determine how things are in reality? On his own
account, as we have seen, Buddhaghosa took his task as a Vibhajjavādin to be anal-
ysis and categorization to explore relationality, and he did not, at any point, stand
back from the methods that comprise these activities to claim or argue that they
reveal the way things are. Recall that for him, the intellectual work of the three
piṭakas is the radical exploration of the workings of the Buddha’s omniscience,
radical in that, like the Buddha’s knowledge, it continues unobstructed into all
possible phenomena. One’s practice as a Buddhist is to follow the Buddha’s om-
niscience in its workings, which in the case of the Abhidhamma are procedures
of analysis and exploration of conditional relations. Abhidhamma for him is the
intellectual work of breaking down the moments of experience and categorizing
them into further groupings to understand conditionality and relationships in
their temporal aspects. It does not yield, at the end, a definitive description of, ar-
gument for, or view about metaphysical reality; in fact, as an intellectual practice,
at least in theory, it has no final ending. Abhidhamma is seen as a tool for seeing
and understanding to practice the unobstructed probing of phenomena enacted
by the Buddha’s omniscient ken.
I argue that Buddhaghosa read the Abhidhamma as a set of methods or
tools for disaggregating moments of experience to help us explore causal
patterns among the phenomena it begins to identify; it explores the functions
and manifestations of phenomena according to the various ways they can be
considered in different patterns. The Abhidhamma—​and Buddhaghosa’s exten-
sion of its practices in his commentaries and the Visuddhimagga—​are conceived
as methods for disentangling the tangle of experience so that we can begin to cut
through our ensnarement in it. Matrices are the starting points for analysis, and
the classic schemes and patterns so foundational to the doctrines of the tradition,
like the four noble truths, name-​and-​form, the five aggregates, and dependent
origination, are deployed as phenomenological practices—​that is, methods to ex-
plore therapeutically one’s experience—​and not assertions or arguments about
reality.
150 In t erpr et ing t h e T hr ee PiṬak as

A Precedent in the Suttas
At this point it might be helpful to refer to a sutta that I think shares in or even
inspires the methods Buddhaghosa deploys and which can help us see what
the canonical Abhidhamma might be doing. I do not argue here that the phe-
nomenological analysis I  see at work in this sutta is the dominant or the only
kind of knowledge practiced in the Pali suttas. The tradition is just too big and
the methods and genres of texts are just too varied for such grandiose claims
(as I have been suggesting throughout). But I do feel confident in asserting the
presence of an analytical and phenomenological strand of thinking that can be
traced back to the Buddha of the Pali suttas, and it was an approach about which
the Buddha was explicit. Consider as one example the “Many Types of Feeling
Sutta” (Bahuvedanīya Sutta). I give the whole of the relevant passage here.

Thus have I heard. On one occasion the Bhagavan was living at Sāvatthi
in Jetavana in Anāthapiṇḍika’s Park. Then the architect Pañcakaṅga
approached Venerable Udāyī, paid homage to him, sat to one side of him,
and asked:
“Venerable sir, how many kinds of feeling (vedanā) has the Bhagavan
stated?”
“Architect, three feelings have been stated by the Bhagavan. The three
feelings stated by the Bhagavan are pleasurable (sukha) feeling, painful
(dukkha) feeling, and neither-​pleasurable-​nor-​painful (adukkhamasukha)
feeling.”
Thus told, the architect Pañcakaṅga said to Ven. Udāyī: “But Venerable
Udāyī, the Bhagavan did not say there were three feelings. He said there
were two feelings:  pleasurable feeling and painful feeling. This neither-​
pleasurable-​nor-​painful feeling was described by the Bhagavan in terms of
an exalted and peaceful pleasure (sukha).”
Then a second time and a third time Ven. Udāyī told the architect
Pañcakaṅga that the Bhagavan spoke of three feelings, not two. And
a second and third time Pañcakaṅga said that the Bhagavan spoke of
two feelings, not three. Ven. Udāyī was not able to convince the archi-
tect Pañcakaṅga, and the architect Pañcakaṅga was not able to convince
Ven. Udāyī.
Ven. Ᾱnanda heard this discussion of Ven. Udāyī and the architect
Pañcakaṅga, and so he went to the Bhagavan. He approached him, paid
homage to him, and sat to one side of him. Sitting to one side, Ven. Ᾱnanda
told the Bhagavan everything of this discussion between Ven. Udāyī and
the architect Pañcakaṅga. Thus informed, the Bhagavan said this to Ven.
 Disentangling the Tangle 151

Ᾱnanda: “Ᾱnanda, it was actually a true way of teaching (pariyāya) that

Udāyī would not accept from the architect Pañcakaṅga, and a true way
of teaching that the architect Pañcakaṅga would not accept from Ven.
Udāyī. Ᾱnanda, I have stated two types of feeling in one way of teaching.
I  have stated three types of feeling in another way of teaching; I  have
stated five types of feeling in another way of teaching; I have stated six
types of feeling in another way of teaching; I have stated eighteen types of
feeling in another way of teaching; I have stated thirty-​six types of feeling
in another way of teaching; and I have stated one hundred and eight types
of feeling in another way of teaching. This is how the Dhamma is taught
through [various] ways of teaching.
When the Dhamma is taught by me through such ways of teaching
it can be expected that even what is well spoken and well said by some
will not be approved, accepted, or conceded by others, and they will live
quarreling, disagreeing, arguing, and attacking one another with verbal
weapons. But still, Ᾱnanda, the Dhamma is taught by me through such
ways of teaching. When the Dhamma is taught by me through such ways
of teaching it can be expected that what is well spoken and well said by
some will be approved, accepted, and conceded by others, and they will
live in harmony, politely, not arguing, blended like milk and water, and
gazing upon one another with loving eyes. (M i.396–​400; cf. S iv.223–​28,
231–​32)

This discussion, prompted by the intelligent lay follower and architect Pañcakaṅga
and the venerable Udāyī, suggests that they want to take the Buddha’s utterances
about feelings (vedanā, a term that becomes a key category of Abhidhamma
analysis) to deliver final clarity on just how many feelings there are. Are there
two (pleasure and pain) or are there three (pleasure, pain, and indeterminate
feeling)? They want a definitive answer as to which enumeration is the correct
or final description of what feeling is. The Buddha’s answer to them is highly in-
structive because it is a rebuttal of their entire mode of thinking. In fact, both
enumerations are correct ways of teaching (pariyāya). Further, all the Buddha’s
various enumerations of feelings are just that, ways of teaching about experience
to be usefully deployed in different contexts. We can analyze feeling into two
categories, three, five, six, eighteen, and so forth, for different purposes, but one
should not stop with any single enumeration and declare it to be, in the final anal-
ysis, what feeling is. We should resist taking analytical distinctions for ontological
description. The Buddha notes that this approach to teaching—​something “well
spoken” and “well said”—​may not sit well with those who are contentious, who,
presumably, wish to argue for and against fixed positions. But for those who do
152 In t erpr et ing t h e T hr ee PiṬak as

respond to such an approach, the various enumerations of the Dhamma that ex-
plore feeling in a modal fashion can be put to practice in life without causing
friction.
Buddhaghosa does not say much about this exchange, but his few details are
worth noting. About the architect Pañcakaṅga, a “head carpenter,” he notes simply
that Pañcakaṅga’s name means “Five Tools” because he carried five tools, namely,
an adze, chisel, measuring stick, hammer, and a measuring string-​and-​tube.12 This
little detail is an intriguing narrative device. Pañcakaṅga is interested in tools,
and can, presumably, be persuaded that the Buddha’s teachings about feelings are
just that, tools. But architects are also attentive to numbers and may well be prone
to precision in wanting to know the architecture of experience:  are there two
feelings or three? And so he debates with Udāyī, an elder described as a paṇḍita,
a learned scholar. Both are right about the number of feelings but in both being
right both are wrong if they think only one listing is apt. Buddhaghosa glosses
pariyāya, “way of teaching” as kāraṇa (activity, action, performance), furthering
the idea that teaching various lists is really an activity or performance aimed at par-
ticular purposes.13 Enumerations are contextual activities that make possible fur-
ther explications of experience—​they perform or enact instruction; the Buddha
gives various listings of feeling depending on context or purpose. Buddhaghosa
goes on to explain that the Buddha teaches two feelings when speaking of feeling
in the sense of the phenomena of the body and of mental awareness; three when
speaking of pleasure, pain, and indeterminate feeling; five when speaking of the
five [main] faculties; six with reference to the six sense doorways; and so on.14
We can surmise from this that what he is after is not a single and final architec-
tonic of experience, but identifying how particular phenomena can be variously
schematized in a modal and modular fashion.
This line of approach to enumerated lists is helpful to us for interpreting the
strand of teachings that informs Buddhaghosa’s project of reading Abhidhamma
lists and in writing his own treatise, the Visuddhimagga. The canonical lists
variously carve up experience. Their lists are open-​ended and resist closure.
The Dhammasaṅgaṇī does not give complete or exhaustive itemizations of
dhammas:  it is not noticed often enough that each list in this text ends with

12. Spk iii.79: vāsipharasunikhādanadaṇḍamuggarakāḷasuttanāḷisaṅkhātehi vā pañcahi aṅgehi

samannāgatattā so pañcakaṅgoti paññāto. Thapatīti vaḍḍhakījeṭṭhako.
13. Spk iii.79: pariyāyenāti kāraṇena.
14. Spk iii.79; Ps iii.113–​14: Pariyāyanti kāraṇaṃ. Dvepānandāti dvepi, ānanda, pariyāyenāti
kāraṇena. Ettha ca kāyikacetasikavasena dve veditabbā, sukhādivasena tissopi, indriyavasena
sukhindriyādikā pañca, dvāravasena cakkhusamphassajādikā cha.
 Disentangling the Tangle 153

an “et cetera,” as I discuss further in what follows. Read with the Bahuvedanīya
Sutta, we can suggest that its enumerations of lists might be “ways of teaching”
that begin to provide various enumerations in order to develop understanding
through different tools. Buddhaghosa carries forward this method of analysis in
his own work.
To be sure, Buddhaghosa does think that there are better and worse ana-
lytical tools and lists. Not any list goes. He is a devout Buddhist with specific
dogmatic commitments that frame his work and so he tries to show that the
specific lists and formulas given in the Buddha’s teachings and comprising the
key terms and formulas of the Abhidhamma texts—​the four noble truths,
the five aggregates, the twelvefold dependent origination, and so on—​are the
best methods for analysis to achieve the therapeutic and soteriological aims
of Buddhist practice and understanding. In places he seems to give special
weight to distinguishing the processes of name or naming (nāma) and form
or forming (rūpa) in our phenomenology as the central Abhidhamma task.15
What we might say is that for him what makes these methods useful is not that
they argue for a posited metaphysical reality (either of experience or things
“out there”) or because they land, at the end, on a single, final analysis, but
because of their efficacy in bringing about certain experiences or realizations.
That is to say, the truth value of the methods he deploys is not grounded on a
claim that they are or produce accurate accounts of reality; his scholarly work
does not aim at metaphysical argumentation. Rather, he is, as he has told us all
along, an Analyst. He takes the Abhidhamma texts and his own philosophical
task in commenting on them to be the practice of analysis to achieve parti­
cular purposes for understanding and refashioning one’s experience—​that is,
to help one disentangle the tangle.
One of the challenges in making this argument is that we nearly always find
Buddhaghosa in the midst of carrying out this practice rather than standing back
from it and telling us explicitly what he is doing. This of course is fully consistent
with what I have argued he takes to be his task, which is to not argue for a posi-
tion. On his own description he is a Vibhajjavādin, one who, in accordance with
the tradition, can “take up the meaning [of a given matter] and then return again
to that meaning by explaining it with different methods,” while never affirming

15. Sp 22; Sv 1.19; As 21: [the third piṭaka, the Abhidhamma] he characterizes as “this is a teaching
on distinguishing name and form, for here is taught the distinguishing of name and form
that is the opposing of the passions, et cetera.” (rāgādipaṭipakkhabhūto nāmarūpaparicchedo
ettha kathitoti nāmarūpaparicchedakathāti vuccati). See Heim and Ram-​Prasad, “In a Double
Way: Nāma-​Rūpa in Buddhaghosa’s Phenomenology,” for a phenomenological reading of his
work on nāmarūpa and the entire Abhidhamma.
154 In t erpr et ing t h e T hr ee PiṬak as

one’s own view.16 For my part, I  only realized that this is what he is doing by
following him around texts for a very long time watching him do the work of an-
alytical and classificatory practice without ever drawing back and making meta­
physical assertions. In some important sense, I  believe, the best argument for
this understanding of his approach is the Visuddhimagga itself. There is a crucial
sense in which this text must be read as a whole, or at least taken in large chunks,
to see the modular methods of its treatment of ideas. Its methods might not be
apparent to those who dip into it as a reference tool to look up single points of
doctrine, because of the risk of reifying analytical categories by severing them
from the very contexts in which they are being defined.
My arguments to demonstrate Buddhaghosa’s phenomenological approach
will draw on several kinds of evidence. First, I  begin by gathering together,
recapping, and furthering what we have seen Buddhaghosa say in previous
chapters about what it is that the Abhidhamma does. I  have argued and will
continue to demonstrate that his commentarial framing of the Abhidhamma
offers essential clues for his theories of texts and practice. There is much to be
said for taking him at his word about what he thinks texts are and what he says
he is doing as an interpreter. Second, I consider several types of methods that he
discusses in the Abhidhamma commentaries focusing on the Dhammasaṅgaṇī,
the Vibhaṅga, and the Paṭṭhāna as well as his Visuddhimagga. I argue throughout
that he understands Abhidhamma textuality to initiate a series of methods and
practices whereby the practitioner can practice the movements of the Buddha’s
omniscience to yield transformative vision and understanding.

Buddhaghosa’s Oceanic
and Immeasurable Abhidhamma
We have seen that Buddhaghosa thinks that it is only in the Abhidhamma, spe-
cifically the Paṭṭhāna, that the omniscience of the Buddha can begin to find
room. It was in the fourth week of awakening upon contemplating the algo-
rithmic relationships of this text that dazzling rays emitted from his body cast
forth a brilliance that shot out through billions of world systems.17 In ­chapter 1
we explored the spatial and temporal metaphors Buddhaghosa proposes to
interpret what the Abhidhamma is—​it is a great ocean of methods whose

16. Vism 522 and Vibh-​a 130. See ­chapter 2.

17. As 13–​14. Endo discusses the Buddha’s rays at this event, noting the Atthasālinī’s emphasis
on this event as portraying the Buddha’s omniscience (“The Buddha’s Fathom Long Halo
[Byāmappabhā] and Rays [Raṃsi],” 99).
 Disentangling the Tangle 155

unfathomable depths carry those contemplating it to sublime joy and bliss.

Recall the Venerable Mahāgatigamiyatissa, who, on a boat from Lanka to India,
likened the powerful and endlessly vast ocean to the methods of the Paṭṭhāna.
This analogy helped him to see that the “border to the complete Paṭṭhāna cannot
be known,” a realization prompting his arhatship. In another simile we explored
earlier, Buddhaghosa considers the classificatory groupings of phenomena in var-
ious listings to be similar to the arranging of the stars in the heavens in infinite
constellations. The Buddha brought order and arrangement (in the formulas of
the Abhidhamma) to apparently measureless and disordered phenomena. And
the temporal aspects of the Abhidhamma evoke great puzzlement:  how could
its “hundreds and thousands of methods,” a knowledge “endless and immeasur-
able when worked out with the mind” come to be communicated in a mere three
months’ time to Buddha’s mother in heaven and then to Sāriputta on earth? And
how is the endlessness of its methods to be captured in complete recitations and,
ultimately, in seven finite volumes? We recall that the way he was able to teach
the Abhidhamma to Sāriputta was through teaching him “a hundred methods, a
thousand methods, a hundred thousand methods,” much like a man standing at
the shore merely gesturing to the vast ocean.18 Abhidhamma in these discussions
is an infinitely vast ocean of methods.
Buddhaghosa is very explicit about how he interprets the Abhidhamma both
in contiguity with and in contrast to Sutta and Vinaya genres of thought, and
a brief review of our findings in c­ hapter 2 may be helpful. A dominant theme
for him is the Abhidhamma’s inherently expansive nature. In glossing the term
“Abhidhamma,” he claims that the particle “abhi” means that it “exceeds and is
distinguished from the Dhamma”19 of the Suttas in that it classifies teachings
in an unlimited way. For example and as we have seen, when presented in the
Suttanta, the five aggregates are “detailed only with regard to one instance in
the Suttanta, not unrestrictedly” as they are in Abhidhamma analysis.20 By not
being restricted to a single place or instance, they offer a sort of “view from no-
where.” And again as we have already seen, he sees all three genres of the tipiṭaka
as different methods (naya) that the Buddha taught;21 they are also practices of
study (pariyatti), not just “baskets” (piṭaka) or collections of teachings.22 They are

18. As 16; see ­chapter 1 and Appendix C.

19. As 2: Dhammātirekadhammavisesaṭṭhena. Atirekavisesatthadīpako hettha ‘abhi’-​saddo.
20. As 2: Suttantañhi patvā pañcakkhandhā ekadeseneva vibhattā, na nippadesena.
21. For example, As 19.
22. As 20.
156 In t erpr et ing t h e T hr ee PiṬak as

trainings (sikkhā) at which one could be better or worse trained, which refer not
to memorizing doctrine but engaging in styles of thinking. A monk ill-​trained in
the Abhidhamma, for example, can “overrun thinking on dhammas,” and think
about imponderables (like speculative ponderings about the world).23 Conversely,
Buddhaghosa ventures to say, one well trained in Abhidhamma is the true teacher
of the Dhamma; unlike one trained only in the Suttas, the Abhidhamma expert
never gets muddled (na āloḷenti) and can answer every question.24 He knows the
Dhamma from every side by his mastery of matrices and how to expand them to
every instance.
Buddhaghosa develops further his approach to the Abhidhamma in his
nidāna to the Abhidhamma piṭaka in the Atthasālinī. As much as he suggests that
Abhidhamma is a decontextualized or abstracted form of teaching (nippariyāya),
as we saw in ­chapter 2, Buddhaghosa is quite concerned to give it a context. In
fact, that the Abhidhamma texts themselves apparently lack a nidāna becomes a
problem for him because it gives fodder to those who would deny their canon-
icity.25 He deals with this by providing several possible options for its nidāna,
options that suggest what he takes it to be.
He observes that one scholar, the elder Tissabhūti, considers the
Abhidhamma nidāna to be the Mahābodhi tree itself. In this view, the Buddha
is said to describe his dwelling at the place of awakening as a time for “pene-
trating feeling (vedanā) in an unlimited way (nippadesa)” via ten different places
or instances (padesa):  according to the aggregates, the bases, the elements, the
truths, the faculties, the conditions, the foundations of mindfulness, the jhānas,
name (nāma), and according to dhammas. Tissabhūti argues that the Buddha’s
awakening at the Bodhi Tree was the initial location where he learned to ex-
plore phenomena in an unlimited fashion; on a much later occasion, he returned
to this site and “dwelled” in it when on retreat, and emerged from it to teach
a sutta that Buddhaghosa calls the Padesavihārasutta.26 So the Bodhi Tree was
the setting (nidāna) for this sutta. But Buddhaghosa reports that another elder,

23.  As 24; Sp i.26:  “One poorly accomplished in Abhidhamma overruns thinking on

dhammas and thinks about imponderables, and from this becomes deranged” (Abhidhamme
duppaṭipanno dhammacintaṃ atidhāvanto acinteyyānipi cinteti, tato cittakkhepaṃ pāpuṇāti).
This passage goes on to mention an Aṅguttara passage on four unthinkable things, one of
which is ponderings on the world (lokacintā) (A ii.80).
24. As 29.
25. As 29.
26. As 29–​30. The sutta to which this is referring seems to be S v.12–​13. In this sutta the Buddha
claims to have returned, for a fortnight, to the place he occupied shortly after his awakening,
when he contemplated feeling and its conditions.
 Disentangling the Tangle 157

Sumanadeva, finds this account to be unlikely and offers his own claim that the
nidāna—​the context and place of the original teaching—​is at the throne of the
god Sakka at the root of the Pāricchattaka tree in Tāvatiṃsa heaven, where the
Buddha taught the Abhidhamma to the deities.27
Buddhaghosa’s own account is much more elaborate. He says that unlike the
suttas, which have a single nidāna describing when and where they were taught,
the Abhidhamma has two nidānas: one which is the origin and context for the
Buddha’s attainment (adhigamanidāna) of the Abhidhamma, and one for his
teaching of it (desanānidāna). The nidāna for his attainment or discovery of it
is the very long period of starting off as Sumedha, who makes a vow under the
Buddha Dipaṅkara, becomes a bodhisatta, and then spends one hundred thou-
sand eons and four incalculable periods developing the Ten Perfections prerequi-
site to his attainment of awakening under the Bodhi Tree. This nidāna continues
up until he penetrated omniscience in the third watch of the night:

In the first watch of the night he attained knowledge of his previous

lives, in the middle watch he attained knowledge of disappearance and
reappearance (that is, transmigration), and dwelling in the last watch
of the night he comprehended the knowledge of omniscience adorned
with the qualities of all the buddhas such as the Ten Powers and the Four
Confidences, and attained the ocean of the Abhidhamma methods. This
should be understood as the nidāna of the attainment of it.28

This sequence describes a very long process of living innumerable previous lives
that culminated in the last part of the night, when he arrives finally at the abstract
and general methods that articulate the processes structuring all those previous
lives. In fact, because the Abhidhamma was conditioned by these biograph-
ical events, Buddhaghosa goes on to include the long story of Sumedha of the
Jātakanidāna, using this text to serve as the Abhidhamma’s nidāna as well. This is
how we come to find, rather unexpectedly, a very long excursus in the Atthasālinī,
where the introduction of the Jātaka commentaries works also as the introduc-
tion to the Abhidhamma Piṭaka.29 The entire four incalculable eons and 100,000

27. As 30–​33.
28.  As 35:  paṭhamayāme pubbenivāsañāṇaṃ, majjhimayāme cutūpapātañāṇaṃ
patvā, pacchimayāmāvasāne dasabalacatuvesārajjādisabbabuddhaguṇapaṭimaṇḍitaṃ
sabbaññutaññāṇaṃ paṭivijjhantoyeva imaṃ abhidhammanayasamuddaṃ adhigañchi.
Evamassa adhigamanidānaṃ veditabbaṃ.
29.  As 32; Pe Maung Tin does not translate this large section of the text in The Expositor,
41. For a translation of this section of the text in the Nidānakathā of the Jātakaṭṭhakathā,
158 In t erpr et ing t h e T hr ee PiṬak as

kappas of the Bodhisatta’s preparation for his discovery of the Abhidhamma is

the maximal context or origin conceivable.
The second nidāna, that is, the origin and context for the teaching of the
Abhidhamma, is the period right after this, as the Buddha sat contemplating the
teaching, particularly the fourth week, when he probed the Abhidhamma.

The Abhidhamma having been attained in this way, [the Buddha] then
spent seven days seated on a single throne, seven days not blinking his eyes,
and seven days walking up and down. In the fourth week, having examined
the Abhidhamma which had been attained by discovering it with his
own self-​existent knowledge, he passed the next three weeks among the
Goatherds’ Nigrodha tree, the Mucalinda tree, and the Kingstead tree.30

The initial discovery of Abhidhamma had him riveted, unblinking and un-
moving. But it unfolds over time, to a point where he can move and then examine
it further in a full week. After four more weeks he decides to teach (after some
hesitation and being beseeched by the great deity Brahmā), and eventually sets
out to teach “the group of five at the Deer Park of Isipatana.” This nidāna goes
up until the first sermon at the Deer Park at Isipatana, the moment known as the
Turning of the Wheel of the Dhamma.

And turning the Wheel of the Dhamma he made them, who were headed
by the elder Aññāsikoṇḍañña, as well as eighteen koṭis of brahma deities,
drink the ambrosia of the Deathless. Thus it should be understood that
the nidāna of the teaching is [what occurred] up to the Turning of the
Wheel of the Dhamma.31

This fascinating reading locates the teaching of the Abhidhamma not as a dis-
crete teaching apart from the teaching of the suttas, but as commencing pre-
cisely when they do, at the first sermon. This suggests, as I take up later, that the
Abhidhamma is, at least in some important sense, the Suttanta Dhamma: they
are taught simultaneously.

see Jayawickrama, The Story of Gotama Buddha, 3–​61. The durenidāna as given in the
Jātakaṭṭhakathā is a slightly longer account, however.
30.  As 35:  Evaṃ adhigatābhidhammo ekapallaṅkena nisinnasattāhaṃ animisasattāhaṃ
caṅkamanasattāhañca atikkamitvā, catutthe sattāhe sayambhūñāṇādhigamena adhigataṃ
abhidhammaṃ vicinitvā aparānipi ajapālamucalindarājāyatanesu tīṇi sattāhāni vītināmetvā.
31. As 35: dhammacakkaṃ pavattento aññāsikoṇḍaññattherappamukhā aṭṭhārasa brahmakoṭiyo
amatapānaṃ pāyesi. Evaṃ yāva dhammacakkappavattanā desanānidānaṃ veditabbaṃ.
 Disentangling the Tangle 159

But first, let us look in more detail how he treats the Abhidhamma nidānas.
To specify the particulars of the nidāna of the Abhidhamma, Buddhaghosa asks
and answers an important listing of questions about the Abhidhamma:

By what was it fostered? It was fostered by faith in the aspiration for

Awakening.
Where did it mature (lit. become “cooked”)? In the five hundred and fifty
jātakas.
Where was it acquired? At the root of the Bodhi tree.
When was it acquired? The full moon of Visākha.
By whom was it acquired? By the Omniscient Buddha.
Where was it examined? At the ground of the Bodhi tree.
When was it examined? During the seven days at the Jewel House.
By whom was it examined? By the Omniscient Buddha.
Where was it taught? Among the deities in Tāvatiṃsa heaven.
For whose sake was it taught? The deities.
For what purpose was it taught? For the sake of release from the four floods.
By whom was it received? By the deities.
Who is learning it? Trainers and ordinary good people.
Whose learning is mastered? Arhats free of taints.
Who knows it by heart? Those who practice it know it by heart.
Whose word is it? The word of the Buddha, the Perfectly Awakened Arhat.
By whom has it been conveyed? By the lineage of teachers.32

This catechetical format echoes the style of asking about the nidāna that we
see for the Sutta and Vinaya recitations, where Ᾱnanda is quizzed on the con-
textual particulars of the suttas and Upāli on the vinaya rules. The acquisition,
study, teaching, reception, impact, and future transmission of this knowledge are
each located and specified. Perhaps most intriguingly, the Abhidhamma is said
to have been “cooked” in the jātaka stories that comprise the many lifetimes of
the Bodhisatta as he worked for the awakening experience under the Bodhi tree.
This account situates the Abhidhamma in the Buddha’s life story, as all nidānas

32.  As 31–​32:  kena pabhāvitoti bodhiabhinīhārasaddhāya pabhāvito. Kattha paripācitoti

aḍḍhachakkesu jātakasatesu. Kattha adhigatoti bodhimūle. Kadā adhigatoti visākhāpuṇṇ
amāsiyaṃ. Kenādhigatoti sabbaññubuddhena. Kattha vicitoti bodhimaṇḍe. Kadā vicitoti
ratanagharasattāhe. Kena vicitoti sabbaññubuddhena. Kattha desitoti devesu tāvatiṃsesu.
Kassatthāya desitoti devatānaṃ. Kimatthaṃ desitoti caturoghaniddharaṇatthaṃ. Kehi
paṭiggahitoti devehi. Ke sikkhantīti sekkhā ca puthujjanakalyāṇā ca. Ke sikkhitasikkhāti arahanto
khīṇāsavā. Ke dhārentīti yesaṃ vattati te dhārenti. Kassa vacananti bhagavato vacanaṃ, arahato
sammāsambuddhassa. Kenābhatoti ācariyaparamparāya.
160 In t erpr et ing t h e T hr ee PiṬak as

must; philosophy is, as we have seen, biographical. But this is a prologue like no
other. The entire preawakening biographical tradition of incalculable previous
lives culminating in the moment under the Bodhi Tree becomes the runup to
the discovery of the Abhidhamma method. When the Jātakas become the con-
text for the Abhidhamma we discern the maximal scope and scale of what the
Abhidhamma is. The Bodhisatta’s entire spiritual journey is the origin and con-
text for it.
Further, by situating the discovery of the Abhidhamma as the culmination of
the Perfections developed over all the many previous lives, this account makes the
additional point that the Abhidhamma is the abstraction of knowledge earned
from living many lifetimes. The raw experience of all of those lives becomes, fi-
nally, “cooked”—​refined, completed—​during the awakening. The knowledge
comes from the particular lives the Bodhisatta lived in all their complex singu-
larity, which prepared him to acquire in a refined manner the abstract and general
explanatory formulas that constitute its methods.
Also noteworthy is that this passage recalls the tradition of the Buddha
teaching the Abhidhamma as a separate three-​ month-​long sermon to the
Tāvatiṃsa deities among whom was his mother (which we considered in
­chapter 1). Earlier in the Atthasālinī, speaking of the Abhidhamma, he says:

At a future time (that is, after the week of awakening) the Tathāgata sat
in the middle of the deities of the ten thousand world systems at the
Paṇḍukambala stone at the root of the Pāricchattaka tree in Tāvatiṃsa
heaven and, making his mother become a “witness of the body” (an ad-
vanced contemplative attainment), he taught it, passing from one dhamma
to another, teaching the divisions of kusaladhammas, akusaladhammas,
and indeterminate dhammas.33

This locates the teaching of the piṭaka, starting with the Dhammasaṅgaṇī
(which is in fact structured on the threefold kusala, akusala, and indeterminate
dhammas), at some unspecified time subsequent to his awakening. Teaching the
Abhidhamma Piṭaka to the deities in heaven works as a narrative device to suggest
its “divine eye” quality of seeing the general patterns and structures of phenomena
across particular lives. The dwellers in heaven are perhaps best suited to under-
stand it in its pure, entirely abstracted, form. Still, the deities need a buddha to

33. As 15: Aparabhāgepi hi tathāgato tāvatiṃsabhavane pāricchattakamūle paṇḍukambalasilāyaṃ

dasasahassacakkavāḷadevatānaṃ majjhe nisinno mātaraṃ kāyasakkhiṃ katvā kusalā dhammā,
akusalā dhammā, abyākatā dhammāti dhammaṃ desento satabhāgena sahassabhāgena
satasahassabhāgena dhammantarā dhammantaraṃ saṅkamitvā saṅkamitvāva desesi.
 Disentangling the Tangle 161

give them this divine lens for seeing how phenomena work. The mention of his
mother is intriguing. He helps his mother, as a featured member of this heavenly
audience, attain the capacity to become a “witness of the body”; she had died
shortly after his birth and here, over the Abhidhamma, they are reconnected.
Thus we acquire two different ideas about the context of the teaching of the
Abhidhamma; in the first, the nidāna to teaching the Abhidhamma is everything
leading up to the first sermon, where he then commences to teach the Dhamma
(which includes the Abhidhamma). In the second, the Abhidhamma is a separate
body of texts, the piṭaka, realized in the fourth week of Awakening, taught to his
mother and also conveyed to Sāriputta, and (somehow) later captured in seven
volumes. But the first idea, where the Abhidhamma is part of and taught simultane-
ously with the teaching of the Suttas, bears further consideration.
As we have seen, the Abhidhamma is the “higher Dhamma” composed of the
formulas and methods given throughout the suttas, unmediated by their particular
contexts. The suttas put contextual flesh on the methods of the Dhamma categories
and patterns to locate them in particular stories, but they contain the Abhidhamma
methods, which are, of course, the Dhamma itself. We can see this in practice, by
way of example, in the two suttas we started to understand from the last chapter. For
instance, when the Buddha taught the Brahmajāla Sutta he was teaching his monks
about the problematic style of thinking behind the positing of the sixty-​two meta-
physical views in a narrative context precipitated by the occasion with Suppiya. But
at the same time, the teaching offers Abhidhammic analysis at its kernel, particu-
larly in the teaching that the construction of metaphysical positions is conditioned
by vedanā, feeling. Vedanā is an analytic category of Abhidhamma, and describing
its conditioned and conditioning relationality is Abhidhamma practice. Recall
the repeated formula relevant to all the cases in the Brahmajāla’s treatment of the
sixty-​two views:

Monks, the Tathāgata understands this: these viewpoints, grasped in this

way, held in this way, lead in this way to certain future conditions. And
the Tathāgata understands this: he understands what transcends this, but
yet he does not hold on to this understanding, and because of not holding
on to it he alone on his own has found peace. Monks, the Tathāgata is
free of all attachment, having truly known the arising and disappearing of
feelings, and the satisfactions, dangers, and release from them.34

34.  D i.16:  Tayidaṃ, bhikkhave, tathāgato pajānāti—​ ‘ime diṭṭhiṭṭhānā evaṃgahitā

evaṃparāmaṭṭhā evaṃgatikā bhavanti evaṃabhisamparāyā’ti, tañca tathāgato pajānāti, tato
ca uttaritaraṃ pajānāti; tañca pajānanaṃna parāmasati, aparāmasato cassa paccattaññeva
nibbuti viditā. Vedanānaṃ samudayañca atthaṅgamañca assādañca ādīnavañca nissaraṇañca
162 In t erpr et ing t h e T hr ee PiṬak as

We recall that identifying the way that feeling prompts the grasping and promoting
of views is the key insight that the Buddha offers in the sutta; his grasp of the
general pattern underlying and structuring the process of all sixty-​two types of
views is the abstract Abhidhamma formula this particular narrative conveys to
its audience.
We might also consider the Mūlapariyāya Sutta, the second sutta we
considered at length in the preceding chapter; this sutta too has a highly tech-
nical, Abhidhammic teaching at its core that constitutes the content of teaching
the five hundred conceited monks. In this case the content is not about feeling
(vedanā) but about cognizing (saññā), another basic Abhidhamma category, and
the sutta ramifies through other technical Abhidhamma terms and formulas,
such as the twenty-​four bases. These methods are taught in this sutta in order to
have an impact on the five hundred monks.
In sum, what we can say is that while the Abhidhamma methods can be
found in a body of seven treatises in the Abhidhamma piṭaka, they can also be
said to be embedded in the contextually situated knowledge of the suttas. The
Abhidhamma methods are the very methods of the Dhamma; when these are
collected in the piṭaka taught for three months in Tāvatiṃsa, they are worked
out systematically and “unrestricted to any instance” and thus “exceed and are
distinguished from” the Dhamma of the Suttanta. But when the methods of the
Dhamma occur in a particular sutta they speak to their instance (padesa), where
they can teach us how they work within that context. For our sake, as learners, the
same Dhamma teachings are available in these two forms—​as unlimited modular
ramifications unconstrained by particular narrative contexts, and, conversely, as
narratively situated practices that illumine the very contexts in which they are
embedded.
In addition to explaining how the teaching of Abhidhamma is presented in
these two ways, Buddhaghosa has also shed light on how the Abhidhamma terms
and patterns were discovered by the Buddha. In this sense, the Abhidhamma
methods are nippariyāya, abstracted knowledge, because they are the knowledge
that was abstracted from the many previous lives of the Bodhisatta as he learned
about experience through living it. His extraordinary biography provided the raw
experience from which he could understand the general patterns operative in all
experience. His night of Awakening and its immediate aftermath comprise the full
discovery of the Abhidhamma formulas that can be used to explore all experience

yathābhūtaṃ viditvā anupādāvimutto, bhikkhave, tathāgato (this formula is repeated at D i.22,

24, 29, 39). See ­chapter 3, p. 127ff.
 Disentangling the Tangle 163

in a completely unobstructed manner—​precisely what the commentators meant

by his omniscience.

Abhidhamma Methods: Reductive and

Classificatory Analysis in a Modular System
We have seen Buddhaghosa claim that the Buddha discovered the “ocean of
Abhidhamma methods.” But what specifically are its methods and how do
they work to convey the Buddha’s unobstructed and endless omniscience? The
Abhidhamma offers, according to Buddhaghosa, two chief methods (naya). The
first is reductive analysis that breaks down moments of awareness (citta) into the
smaller bits that constitute it (such as sensory contact, cognizing, feeling, inten-
tion, and so on). Reductive analysis and the many lists it generates offer very fine
parsings of moments of experience, with a striking degree of granularity and pre-
cision. But the lists of phenomena do not end there with reduction as their final
end; rather further listings regrouping and recategorizing phenomena destabi-
lize them, and we are led to the second important method that Buddhaghosa
discusses and employs. This is the method of “comprehension by groups,” that is,
assembling the modular data of experience in different sets and groupings to un-
derstand how phenomena are open to and shaped by other members in particular
sets, and how they are parts of systems of causes and conditions that operate in
certain specifiable ways.35 These groupings are various; they constitute the ubiq-
uitous lists and formulas found in both Abhidhamma and Sutta texts (which
will be familiar to all scholars of Buddhism, such as name-​and-​form, the five
aggregates, the twelvefold dependent origination, and so on). For Buddhaghosa
these groupings are methods for analyzing experience, not ontological assertions.
The phenomena yielded by reductive analysis can be further analyzed modally
by how they occur in groupings with other phenomena in experience and across
time in the systems of causal relations in which they are experienced. Listing

35. While this classificatory method is in evidence throughout Abhidhamma and his interpreta-
tion of it, we can point to a quite explicit discussion of Buddhaghosa treating “comprehension
by groups” (kalāpasammasana) in chapter XX of the Visuddhimagga. He gives an example of
the way phenomena grouped together under the category of form (rūpa) can be understood by
the three general characteristics, which are impermanence, suffering, and lacking essence. And
form itself, of course, constitutes part of a larger grouping, the five aggregates (which also in-
clude sensory contact, feeling, perception, and constructions) that make up human experience
and by reference to which form can be understood. Chapter XX concerns the kind of knowl­
edge one gains when one attends to the groupings particular phenomena belong to, the char-
acteristics they share with other members of the grouping, and the causal connections among
items within the groupings as a whole in their interactions and placements in other groupings.
164 In t erpr et ing t h e T hr ee PiṬak as

phenomena and exploring how they work together constitute both the canon-
ical methods and Buddhaghosa’s exposition; his Visuddhimagga also defines and
analyzes experience by breaking it down and then regrouping its components var-
iously to explore their causes and conditions.
The Dhammasaṅgaṇī exemplifies the first method, that of reductive analysis,
though it is not limited to it and also includes the work of groupings as I dem-
onstrate later. The text is an extension and elaboration of the matrices (mātikās)
of the suttas, taking up various moments of experience in the many twofold
and threefold ways in which they may be classified, and then elaborating them.
One threefold distinction that structures much of the text classifies phenomena
(dhamma) into good (kusala), bad (akusala), and indeterminate (abyākata) expe-
rience. The Dhammasaṅgaṇī opens with a long list identifying the constituents in
one particular occasion or moment of good awareness (kusalaṃ cittaṃ):

What are the good phenomena? On whatever occasion there is the

arising of a good awareness of the sensory realm accompanied by
joy and knowledge, whether its object is an object of (visual) form,
sound, smell, taste, touch, or phenomenon, then at that occasion there
are: contact, feeling, cognizing, intention, awareness, initial thinking,
sustained thinking, joy, pleasure, oneness of mind, faculty of faith, fac-
ulty of energy, faculty of mindfulness, faculty of concentration, faculty
of wisdom, mental faculty, faculty of happiness, faculty of vitality, right
view, right thought, right effort, right mindfulness, right concentra-
tion, power of faith, power of energy, power of mindfulness, power of
concentration, power of wisdom, power of shame, power of apprehen-
sion, nongreed, nonhatred, nondelusion, noncovetousness, nonmalice,
right view, shame, apprehension, tranquility of body, tranquility of
mind, lightness of body, lightness of mind, softness of body, softness
of mind, workableness of body, workableness of mind, proficiency of
body, proficiency of mind, uprightness of body, uprightness of mind,
mindfulness, meta-​attention, calmness, insight, exertion, balance, and
whatever other dependently-​arisen formless dhammas occur on that
occasion—​these are good phenomena.36

36. Dhs 9: Katame dhammā kusalā? Yasmiṃ samaye kāmāvacaraṃ kusalaṃ cittaṃ uppannaṃ
hoti somanassasahagataṃ ñāṇasampayuttaṃ rūpārammaṇaṃ vā saddārammaṇaṃ vā
gandhārammaṇaṃ vā rasārammaṇaṃ vā phoṭṭhabbārammaṇaṃ vā dhammārammaṇaṃ vā
yaṃ yaṃ vā panārabbha, tasmiṃ samaye phasso hoti, vedanā hoti, saññā hoti, cetanā hoti, cittaṃ
hoti vitakko hoti, vicāro hoti, pīti hoti, sukhaṃ hoti, cittassekaggatā hoti, saddhindriyaṃ hoti,
 Disentangling the Tangle 165

This is a listing of fifty-​six dhammas or constituents (cetasikas) that can be pre-

sent in a certain moment of “good” awareness (though not every one of them
will be present in every such kind of awareness). A particular moment of con-
scious experience is here broken down and dissected, as it were, into its smaller
constituents, and the text goes on from here to provide similar listings for other
kinds of moments of awareness (bad, indeterminate, et cetera). A fuller discus-
sion of this precise list of fifty-​six cetasikas is not important for our purposes and
has been given elsewhere.37 But several points about this listing are important for
our notice.
First, many of the items repeat themselves. For example, sati, mindful-
ness, occurs once on its own, and also as a faculty, a path factor (right mind-
fulness), and a power. These variant repetitions get at different aspects and
functions of these phenomena in a modal and modular system; as powers they
are intensifying factors, as faculties they have a governing role in awareness, and
as path factors they are eightfold path practices, all of which indicate different
ways they operate and can be experienced. Buddhaghosa says that items can be
repeated because they do different work (kicca) in their different capacities, as
a single craftsman can do different work in different guilds.38 A dhamma can
be classified in multiple different ways—​the classifying of phenomena into
groupings being the second main method of Abhidhamma—​according to the
different work it does. Against an opponent’s view that the repetitions in the
lists are evidence of the texts’ confusion and lack of application he vehemently
objects:

The teaching of buddhas is not lacking application, but is [capable of ]

application. And nothing said [by them] is unknown, for everything
that is said is known. For the Perfectly Awakened Buddha knows the

vīriyindriyaṃ hoti, satindriyaṃ hoti, samādhindriyaṃ hoti, paññindriyaṃ hoti, manindriyaṃ

hoti, somanassindriyaṃ hoti, jīvitindriyaṃ hoti, sammādiṭṭhi hoti, sammāsaṅkappo hoti,
sammāvāyāmo hoti, sammāsati hoti, sammāsamādhi hoti, saddhābalaṃ hoti, vīriyabalaṃ hoti,
satibalaṃ hoti, samādhibalaṃ hoti, paññābalaṃ hoti, hiribalaṃ hoti, ottappabalaṃ hoti, alobho
hoti, adoso hoti, amoho hoti, anabhijjhā hoti, abyāpādo hoti, sammādiṭṭhi hoti, hirī hoti, ottappaṃ
hoti, kāyapassaddhi hoti, cittapassaddhi hoti, kāyalahutā hoti, cittalahutā hoti, kāyamudutā
hoti, cittamudutā hoti, kāyakammaññatā hoti, cittakammaññatā hoti, kāyapāguññatā hoti,
cittapāguññatā hoti, kāyujukatā hoti, cittujukatā hoti, sati hoti, sampajaññaṃ hoti, samatho
hoti, vipassanā hoti, paggāho hoti, avikkhepo hoti; ye vā pana tasmiṃ samaye aññepi atthi
paṭiccasamuppannā arūpino dhammā—​ime dhammā kusalā.
37. Nyanaponika, Abhidhamma Studies, ch. 3; Heim, The Forerunner of All Things, 91–​102.
38. As 135–​36.
166 In t erpr et ing t h e T hr ee PiṬak as

work (kicca) of each and every dhamma, and knowing that, he shows the
classifications according to their work. Knowing the eighteen single-​work
dhammas, he showed their classification in one place for each. Knowing
the seven two-​work dhammas, he showed their classifications in both
places. Knowing that feeling has three different kinds of work, he showed
its classification in three places. Knowing that energy and mindfulness
each have four different kinds of work, he showed their classifications in
four places each.39

This is to say that eighteen of the dhammas have only one job and so are
demonstrated in just one classificatory analysis. But feeling has three kinds of
work and so gets classified in three different places. Mindfulness has four jobs,
as we have seen. And so on. This is a highly functionalist definition where each
item is known and described depending on the work it does. The craftsman who
is skilled in plumbing, carpentry, and roofing is to be known by his various crafts,
the actual work that he does. In the workshop of experience, phenomena are
known and investigated according to the tasks that they do, and their tasks differ
in different contexts.
A further illustration of the modal variability of the dhammas on this list
is that they are here considered kusala (which I translate simply and broadly as
“good” because for Buddhaghosa kusala can cover five normative values: salutary,
blameless, proficient, free of distress, and conducive to pleasant results40). Many
of these phenomena are kusala only when they occur together with other phe-
nomena in a kusala citta, a moment of good awareness. But should the “same”
phenomenon occur in a moment of bad or indeterminate awareness, as many of
them can and do, they become in that context bad or indeterminate. In other
words, phenomena like contact, feeling, cognizing, intention, awareness, and so
on, take on their normative valence depending on the larger relational system in
which they occur, and they are inflected by other phenomena in it.
These considerations make it difficult to see dhammas as intrinsic, self-​
contained, or independent essences because they change according to classificatory

39. As 135: ‘buddhānaṃ desanā ananusandhikā nāma natthi, sānusandhikā va hoti. Ajānitvā

kathitāpi natthi, sabbā jānitvā kathitāyeva.Sammāsambuddho hi tesaṃ tesaṃ dhammānaṃ
kiccaṃ jānāti, taṃ ñatvā kiccavasena vibhattiṃ āropento aṭṭhārasa dhammā ekekakiccāti ñatvā
ekekasmiṃ ṭhāne vibhattiṃ āropesi. Satta dhammā dvedvekiccāti ñatvā dvīsu dvīsu ṭhānesu
vibhattiṃ āropesi. Vedanā tikiccāti ñatvā tīsu ṭhānesu vibhattiṃ āropesi. Vīriyasatīnaṃ cattāri
cattāri kiccānīti ñatvā catūsu catūsu ṭhānesu vibhattiṃ āropesi.
40.  Tattha ārogyaṭṭhena, anavajjaṭṭhena, kosallasambhūtaṭṭhena, niddarathaṭṭhena,
sukhavipākaṭṭhenāti pañcadhā kusalaṃ veditabbaṃ (Sv iii.883–​84).
 Disentangling the Tangle 167

schemes that define them as doing different kinds of work and having different
valences. Nyanaponika Thera underscores this point, arguing, “there is no justi-
fication for believing in any unchangeable ‘bearers’ of definite qualities,” for it is
“impossible to speak of a thing as the bearer of a single quality in a strict sense, if
the functions of the respective factor, its direction of movement, its intensity, and
its kammic quality are variable, in accordance to the relational system to which
that factor belongs.”41 The changing work and manifestation of dhammas suggests
just how dynamic and unfixed canonical Abhidhamma analysis is. Nyanaponika
is right to urge the tremendous methodological importance of these features: a
composite thing cannot be said to be sufficiently described just by enumerating its
single parts, because we must know also how those parts change in relationship to
how they are arranged in various schemas. Abhidhamma methods try to convey
how even the smallest bits of experience change depending on how and with
what else they are arranged. They change jobs, intensities, and valence depending
on the relational system they help constitute in any particular moment of experi-
ence. Like the king involved in the practical task of assembling and assigning jobs
to his craftsmen, the observer of experience is interested in the capacities, skills,
and work of each—​and not, we might surmise, in assigning them a putative un-
changing essence.
Further, the passage ends its list with the open-​ended “and whatever other
dhammas” may occur. This big “et cetera” indicates that this is not a closed or
totalizing list or a final architectonic of experience, but an initial attempt at
describing experience that can be added to. And in fact, Buddhaghosa does add
to it, adding nine additional cetasikas to this particular list.42 This “et cetera”
occurs at the end of every list of dhammas in the Dhammasaṅgaṇī.43 That the
lists may be extended conveys the idea that this type of analysis, far from breaking
down experience into final or exhaustive listings of irreducible bits of reality, is
a process of analytical reduction that can be broken down further or differently
through additional scrutiny of experience and the intercausal relationships that
constitute it.
This open-​endedness evident in the Dhammasaṅgaṇī can be contrasted with
the Abhidhammatthasaṅgaha, Anuruddha’s compendium (perhaps eleventh cen-
tury), which catalogs a tidier and apparently complete list of fifty-​two cetasikas,

41. Nyanaponika, Abhidhamma Studies, 40–​41.

42. As 131.
43. That is, each list ends with “and whatever other formless, dependently-​arisen phenomena
that also occur at that occasion” (ye vā pana tasmiṃ samaye aññepi atthi paṭiccasamuppannā
arūpino dhammā) (Dhs 8, 75, 87, for example).
168 In t erpr et ing t h e T hr ee PiṬak as

with no repetitions and no suggestion that the list might be extended. This me-
dieval text more readily lends itself to an interpretation of it as an ontological
system naming in a final and complete way the ultimate constituents of expe-
rience. (It is often the sole or chief text on Abhidhamma consulted by modern
scholars.) I think too that it is noteworthy that Anuruddha’s text is a compendium
(saṅgaha), a genre quite different from Buddhaghosa’s path (magga), in its effort
to synthesize and summarize a system that presents itself, at least at the canonical
level and in Buddhaghosa’s pragmatic purposes, as an open and unexhausted set
of practices.
To sum up and look ahead: in Buddhaghosa’s hands, the first method of reduc-
tive analysis—​breaking up a moment of experience (a citta) into the phenomena
it comprises (understood as phenomenological terminuses rather than ontolog-
ical ultimates)—​allows for the second method of classificatory analysis whereby
those phenomena come to be known modally according to the groupings in
which they can be seen to operate in experience. We continue below looking
at Buddhaghosa’s definitional practices—​primarily, what it means to identify
the characteristic (lakkhaṇa), particularity (sabhāva), and function (rasa) of
these phenomena—​in order to see how they refine analysis without arriving at
essences. All of these practices of exegetical work of analysis enact the under-
standing (paññā) that is the practice and the goal of both the Abhidhamma as
Buddhaghosa reads it, and his own Visuddhimagga.

Definitional Practices
The phenomena (dhammas) yielded by reductive analysis of a moment of aware-
ness (citta) can be discerned and defined by specific definitional practices devel-
oped by the Abhidhamma commentaries. Such formal practices sought to define
the particular characteristics of dhammas, using terms like sabhāva (particularity)
and lakkhaṇa (characteristic).44 Sabhāva is of course a key term in light of how
the Sanskrit term “svabhāva” was taken by some of the Indian traditions (most
pointedly in the mature Sarvāstivāda Abhidharma and the Madhyamaka critique
of it) to refer to an ontological category, a “real existent” that cannot be further

44.  Unlike the Abhidharma system described by Vasubandhu, which defines dharmas in
terms of dravya, substances, often taken to refer to things-​in-​themselves, irreducible entities,
or ontological reals, we never encounter the Pali equivalent to the Sanskrit term dravya or
the substance ontology that seems to go with it. See Cox, “From Category to Ontology: The
Changing Role of Dharma in Sarvāstivāda Abhidharma,” for a careful charting of the develop-
ment of an ontological reading of dharmas in the Sarvāstivāda system. Cox is careful to distin-
guish this system from that of the Pali.
 Disentangling the Tangle 169

reduced or analyzed. In contrast, sabhāva is used by Buddhaghosa to define the

characteristics of dhammas, but it carries no ontological sense; sabhāva need be
nothing more than the distinctive character, particularity, or “own way of being”
of a dhamma or other category of analysis (as is evident in a passage we discuss
later45). When fifty-​six dhammas are listed in the passage of the Dhammasaṅganī,
they are thought to be distinguishable one from the other according to their
characteristic (lakkhaṇa) or way of being (sabhāva), terms sometimes related to
each other.
Specifying a category’s characteristic (lakkhaṇa) is part of a standard four-
fold definitional procedure in Pali scholasticism:  the others include specifying
its function (rasa), manifestation (paccupaṭṭḥāna), and proximate cause
(padaṭṭhāna). The characteristic can be a “dhamma’s particularity (sabhāva)
or a generality.”46 Something can be characterized by what it shares with other
things or by its particular way of being (apples have characteristics they share with
other fruit as well as their particular characteristics). Lakkhaṇa operates much as
English “characteristic” does—​things have characteristics, both general and par-
ticular, by which they can be distinguished from one another. There need be no
metaphysical claims associated with it.
We can observe Buddhaghosa’s idea of sabhāva by taking up an example of his
usage in a definition (one of any that we could use for this). When considering
vedanā, feeling, he offers many ways of analyzing it. We saw previously that there
are at least ten different ways of analyzing vedanā: according to the aggregates,
the bases, the elements, the truths, the faculties, the conditions (in the twelvefold
formula of dependent origination), the foundations of mindfulness, the jhānas,

45. As 39.
46. As 63: tesaṃ dhammānaṃ sabhāvo vā sāmaññaṃ vā lakkhaṇaṃ nāma. In the same passage
Buddhaghosa defines the other three definitional practices thus: rasa is the function (kicca)
or accomplishment (sampatti); paccupaṭṭhāna is the way it manifests (upaṭṭhānākāra) or the
effect (phala), and padaṭṭhāna is its near cause (āsannakāraṇa). For rasa as kicca or sampatti, see
also Vism 8 (I.21). Much mischief can be caused by translations that translate rasa in these texts
as “essential property” (as, for example, Pe Maung Tin does in The Expositor, 84) or “essence.”
Another misleading translation related to both lakkhaṇa and rasa is Ñāṇamoli’s translation
of yāthāvasarasato; in an important passage about purificatory knowledge we are told about
a certain incomplete knowledge that cannot observe the three characteristics [of saṃsāra] “in
their true nature” in Ñāṇamoli’s rendering (Vism 639 [XXI.2]: yāthāvasarasato tilakkhaṇaṃ
sallakkhetuṃ nāsakkhi). But the phrase yāthāvasarasato is less metaphysically charged than this
makes it sound and can be better put thus: it is not possible [for incomplete knowledge] to
observe the three characteristics “according to their exact function.” If we avoid metaphysical
translations like “nature,” “being,” and “essence” (which would have to be argued for rather
than assumed, and about which Ñāṇamoli himself is uneasy even as he uses them [see ch. VIII,
n 68]), different possibilities for interpreting the entire philosophical project at work here can
open up.
170 In t erpr et ing t h e T hr ee PiṬak as

nāma, and according to dhammas; and of course the Bahuvedanīya Sutta, as we

have seen, suggests its own enumerations going up to 108. This claim in and of it-
self makes it clear that vedanā is not a final, irreducible existent arrived at through
reductive analysis, because all of these schemas are themselves further analyses
of it. We can take two of these different schemas to see the role of sabhāva in
defining feeling: feeling analyzed as aggregate and feeling analyzed as dhamma.
Feeling is, according to one analysis, a cetasika, that is, one of the dhammas, or
phenomena that occur on the lists breaking down moments of experience, but it
is also an aggregate or composite entity (khandha) of phenomena that together
with the other four aggregates can be used to describe human experience without
reference to the idea of a self (atta).
As a composite entity it is, by definition, further reducible. Buddhaghosa says
that as one of the five aggregates, feeling can be defined variously:

But though it is singlefold according to its particularity (sabhāva) be-

cause of its characteristic of being felt, it is also threefold by its type: good,
bad, and indeterminate. . . . And it is fivefold by dividing its particularity
(sabhāva) thus: pleasure, pain, joy, sadness, and equanimity.47

This kind of classificatory matrix should by now be familiar. Sabhāva does not
refer to an irreducible existent arrived at through analysis but merely refers to
the particularity that distinguishes feeling, which is here the phenomenological
experience of being felt (the sabhāva of a phenomenon is usually the verbal form
of it). But note that the ways of being felt, its sabhāva, can be divided further into
three, or again, five. If sabhāva were a primary existent arrived at through final
analysis why is it here further reducible?
So much for vedanā as an aggregate, but how does Buddhaghosa define
vedanā when it is considered as a dhamma, or cetasika? In this case, his defini-
tional practice does not refer to sabhāva at all and he gives instead a functionalist
definition of feeling to get at how it is experienced. If sabhāva were the essential,
irreducible essence of a dhamma for him, it is difficult to explain why he would
fail to mention it when he goes about the business of defining each of the fifty-​
six dhammas. For him, an item is defined by its work, its conditions, and how it

47.  Vism 460–​61:  Sā pana vedayitalakkhaṇena sabhāvato ekavidhā pi jātivasena tividhā

hoti: kusalā akusalā abyākatā cā ti. . . . Sā sabhāvabhedato pañcavidhā hoti: sukhaṃ dukkhaṃ
somanassaṃ domanassaṃ upekkhā ti.
 Disentangling the Tangle 171

manifests, rather than by assigning a final “essence” to it.48 When defining specific
dhammas, he deploys the definitional device of defining a category by its charac-
teristic, function, manifestation, and proximate cause. For example, he defines
vedanā (feeling) thus:

Vedanā is what is felt. Its characteristic is what is felt, its function is expe-
rience or its function is enjoying the manner in which [something] can be
wanted, its manifestation is tasting [the other] cetasikas, and its proximate
cause is calmness.49

To define a phenomenon is to describe what it does (vedanā is felt), how it

functions (it is the experience or enjoyment of something), and how it manifests
by interacting with other cetasikas—​it “tastes” them, and is conditioned by them
(that is, by calmness, passaddhi, another cetasika). The “definition” of vedanā
describes how it is known experientially and how it is related to other phe-
nomena. At no point does Buddhaghosa make an ontological leap here to iden-
tify a primary, irreducible, self-​contained real existent somehow independent of
the phenomenological analysis that defines it.
And yet, several scholars have suggested that even if the canonical
Abhidhamma might be seen to be a purely phenomenological system, the
same cannot be said for Buddhaghosa and the aṭṭhakathā tradition he
represents. Noa Ronkin argues that at the canonical level sabhāva and
salakkhaṇa are just “the epistemological and linguistic determinant(s) of
the dhammas as individuals, rather than an ontological determinant of the
dhammas as primary existents,” but that by the time of the aṭṭhakathā an
ontological model creeps in.50 She bases her case largely on her reading of a
single passage in the Atthasālinī that Rupert Gethin also discusses. This is the
commentary on the key passage we considered earlier, where Buddhaghosa

48.  That Buddhaghosa’s definitional practices do not define dhammas in terms of sabhāva
contravenes Ronkin’s notion that sabhāva is “a synonym for a dhamma” and the commentaries
define “any dhamma by virtue of its sabhāva” (Early Buddhist Metaphysics, 114).
49. As 109: Vedayatīti vedanā. Sā vedayitalakkhaṇā, anubhavanarasā iṭṭhākārasambhogarasā vā,
cetasikaassādapaccupaṭṭhānā, passaddhipadaṭṭhānā. Note that this is a kind of feeling that is joyful.
50.  Ronkin, Early Buddhist Metaphysics, 117. Williams says the “position on the role of the
svabhāva as positing primary existents is common to both the Sarvāstivāda and the Theravāda,”
and cites Vism 484 in support of his assertion that both are agreed on an ontological under-
standing of svabhāva/​sabhāva (Williams, “On the Abhidharma Ontology,” 242). But Vism
484 is just an analysis of bases (āyatana) and says merely that before arising they have no in-
dividuality (or way of being), nor do they have one when they cease. Nothing in this commits
Buddhaghosa to an ontological position.
172 In t erpr et ing t h e T hr ee PiṬak as

describes what the opening passage of the Dhammasaṅgaṇī is doing in saying

“on whatever occasion there is the arising of a kusalacitta” when it begins to
enumerate the fifty-​six cetasikas present in that particular kind of awareness.
Buddhaghosa says:

That is why the meaning should be said this way: on an occasion when the first
great kusala awareness of the sensory realm arises, on that occasion there arise,
by virtue of being the constituents of awareness, more than fifty dhammas
which are dhammas just in the sense of their particularity (sabhāva). There is
nothing else, neither creature (satta), being (bhāva), human (posa), or person
(puggala).51

The question is: what does dhamma mean when it refers to sabhāva? Ronkin (taking
her cues from the Mahāṭīkā), takes this defining of dhamma with reference to
sabhāva to be an ontological move. She says:

the sabhāva indicates that its respective dhamma does not depend on any
other item for its existence. Dhammas are self-​existents; this is the meaning
of their upholding their sabhāva, that is, they uphold their own nature and
thus their self-​existence.  .  .  . Unlike the earlier occurrences of sabhāva as
essence in the sense of a dhamma’s individuator, here it acquires an ontolog-
ical significance.52

But is such an ontological leap required here? And how would such a leap co-
here with Buddhaghosa’s other definitional practices? This passage is perfectly
congruent with the other ways Buddhaghosa has used “sabhāva”:  sabhāva is a
characteristic that distinguishes a dhamma from other dhammas. All sorts of
things are said to have sabhāvas or particularities (including, for example, the
various types of decaying corpses used in meditation practice), which identify

51.  As 155:  Tasmā evamettha attho veditabbo—​yasmiṃ samaye kāmāvacaraṃ paṭhamaṃ

mahākusalacittaṃ uppajjati, tasmiṃ samaye cittaṅgavasena uppannā atirekapaṇṇāsadhammā
sabhāvaṭṭhena dhammā eva honti. Na añño koci satto vā bhāvo vā poso vā puggalo vā hotīti
(As 155).
52.  Ronkin, Early Buddhist Metaphysics, 118. Similarly, Gethin asserts that the statement
focuses “on the fact that there is no being or person apart from dhammas; dhammas are what
exist” (The Buddhist Path to Awakening, 150). On the other hand, see Gethin’s critique of
Ronkin’s “over interpretation” of sabhāva and dhamma (“On the Nature of Dhammas:  A
Review Article”).
 Disentangling the Tangle 173

them without attributing an ontological status to their particularities.53 If we

avoid question-​begging translations of dhamma as “primary existent,” and of
sabhāva as “individual essence” and “primary existent” it is easier to refrain from
ontological leaps.54 Nothing Buddhaghosa says here or elsewhere indicates that
he thinks they are “self-​existents.” Indeed, nothing could be more foreign to his
entire methodology.
The passage says that when they occur in a moment of conscious experience,
dhammas occur in particular ways that make it possible for us to distinguish them
from other dhammas and thus arrive at a list of fifty-​six phenomena. The contrast
with the other possible categories of analysis—​creature, being, human, person—​
suggests that these categories are not useful for the purposes at hand, as we would
indeed expect of paramattha analysis, where “person” and “being” are not the
terms arrived at by further analysis.55 Dhamma analysis does not break down per-
sons or beings; it breaks down moments of awareness, cittas. (Buddhaghosa is
explicit on this point, as we see in what follows).
Elsewhere Buddhaghosa does define dhamma in terms of sabhāva, referring
again to its particular characteristic individuality.56 This occurs in a passage
embedded in a contemplative practice of recollection in the Visuddhimagga, and
the context tells us little about what this could mean. In the Atthasālinī, however,
he gives a more developed definition of dhamma in terms of sabhāva:

Dhammas are that which have their own particularity (sabhāva); alterna-
tively, dhammas are what are had (or borne) by conditions, or dhammas
are what are had (or borne) just by their particularity.57

53.  Vism 183 (VI.35) identifies the distinctive qualities (sabhāva) of ten particular types of
corpses as a contemplative exercise, which, curiously, Ronkin herself notes (Early Buddhist
Metaphysics, 116–​17).
54. Sabhāva is translated as “individual essence” by Ñāṇamoli (The Path of Purification, many
places). Ronkin translates dhammas as “primary existents” and sabhāva as “own nature” (Early
Buddhist Metaphysics, 118). Williams translates sabhāva as “self-​essence” (“On the Abhidharma
Ontology,” 242). That said, the ontological leap is not entirely a fiction of modern scholars,
as it does become increasingly evident at the ṭīkā layer, including the commentary on the
Visuddhimagga, as Ronkin notes. See Karunadasa on this point (The Dhamma Theory, 13–​16).
55. Although, as discussed in ­chapter 2, the official avoidance of the category “person” in the
Abhidhamma and paramattha register is contravened by the fifth book the Abhidhamma
corpus, the Describing Persons (Puggalapaññatti).
56. Vism 293: dhammā ti sabhāvā.
57. As 39: Attano pana sabhāvaṃ dhārentīti dhammā. Dhāriyanti vā paccayehi, dhārīyanti vā
yathāsabhāvatoti dhammā. It is not easy to translate dhārenti and dhārīyanti here. These verbs
mean to wear, have, possess, bear.
174 In t erpr et ing t h e T hr ee PiṬak as

Ronkin sees this as an ontological move. She says, “does not the very use of the
term sabhāva overstress the reality of the dhammas and imply that a dhamma is
a discrete entity, a ‘thing’ existing in its own right?”58 But if dhammas “exist by
their own right” how can they also (or alternatively)—​as is stated in this very
passage—​have conditions? If this is the passage where Buddhaghosa is staking
the claim that dhammas are self-​existent independent essences it is surprising that
he would simultaneously state that they can be borne by conditions. The passage
may more easily and coherently be taken to suggest that a dhamma distinguishes
one thing from another by “particularizing”—​by having a particular character-
istic, which is consistent with his definitional practice of identifying the charac-
teristics of phenomena that distinguish them from others.
When we watch Buddhaghosa in practice with this term, we find that
sabhāva need not entail a notion of essence or self-​existent that is the target
of Madhyamaka critics. In fact, when any Abhidhamma category—​whether a
dhamma or an aggregate or any other paramattha term—​is said to possess its own
particularity it can at the same time be located within an analytically dynamic
modal and modular system that continues to break it down into further analysis
to understand, through the formulas of conditionality, how it works within a rad-
ically conditioned network of phenomena.

Analysis from All Sides
The two other large Abhidhamma texts, the Vibhaṅga and Paṭṭhāna, are both
substantive elaborations and reworkings of the Dhammasaṅgaṇī’s matrices, even
while they take different approaches to analysis. We consider how Buddhaghosa
understands their methods briefly here. The title of the Vibhaṅga means “anal-
ysis,” indicating analytical practices associated with the verbal form vi-​bhāj. Bhāj
and vi-​bhāj mean to distinguish, divide, classify, and develop the details of some-
thing (and give us Vibhaṅga, Vibhajjavāda, vibhajana). The Vibhaṅga does this
by taking up eighteen of the classic formulas or groupings used to describe ex-
perience (aggregates, bases, elements, the four truths, faculties, dependent orig-
ination, et cetera) and subjecting each to a threefold method of analysis, what it

58. Ronkin, Early Buddhist Metaphysics, 112. She translates the passage with more substantive
language than I do: “dhammas bear their own particular natures. Alternatively, dhammas are
borne by conditions, or according to their particular natures” (112). Ronkin’s interpretation
of this as a commentarial assertion of the ontological reality of dhammas has been influen-
tial for other scholarship, as for example, Skilton, citing Ronkin, says this: “crucially also the
commentaries work up a definition of dhamma (ultimate constituent) as “that which bears its
own-​nature” (Skilton, “Theravāda,” 81).
 Disentangling the Tangle 175

calls the Abhidhamma Analysis (abhidhammabhājanīya), the Suttanta Analysis

(suttantabhājanīya), and Asking Questions (pañhāpucchaka). This methodology
is particularly important for us because it is a structure that Buddhaghosa notices
and discusses as the varieties of pedagogical methods the Buddha uses.
The differences between Suttanta Analysis and Abhidhamma Analysis as they
are enacted in the Vibhaṅga, and as Buddhaghosa describes them in his com-
mentary on this text, are several: first, Abhidhamma Analysis is a more detailed
classification than what one sees in the Suttanta.59 He says, “in reference to the
Abhidhamma, the Tathāgata never fails to give a method in a place where it is ap-
propriate to give a method.”60 For example, he says that the four Noble Truths can
be examined in “an unlimited way” or “not restricted to one side” (nippadesato),
and so, in effect, from all sides in the Abhidhamma method.61 This results in a po-
tentially quite lengthy analysis of the four truths. I do not trouble the reader with
the math here but content myself merely to note that Buddhaghosa ends this
particular description of the Abhidhamma treatment of the four Truths with this
assertion: “in this way the Abhidhamma Analysis [of this] should be understood
as an explanation with three main sections, fifteen divisions, and embellished
with sixty thousand methods.”62 Fortunately he only gives some of the headings
of these, and so keeps things brief. But the point is made: such detailed exposi-
tion and analysis from all sides rather than partially is in keeping with the distinc-
tion we saw in ­chapter 2, of the nippariyāya discourse often associated with the
Abhidhamma. Nippariyāya is a use of language to serve meaning in all ways rather
than by single instance (pariyāya).
Another difference is that the Abhidhamma Analysis of a phenomenon is
to break down moments of experience, rather than understanding phenomena
over larger periods of time such as across lives.63 This is perhaps most salient in
the Vibhaṅga and its commentary, the Sammohavinodanī, in their treatment of
dependent origination. Buddhaghosa says that the Suttanta Analysis treats de-
pendent origination in terms of how it explains experience across rebirths, while

59. As for example, Vibh-​a 35: Idāni abhidhammabhājanīyaṃ hoti. Tattha rūpakkhandhaniddeso

heṭṭhā rūpakaṇḍe vitthāritanayeneva veditabbo. Cf As 2.
60.  Vibh-​a 37:  Abhidhammañhi patvā tathāgatena nayaṃ dātuṃ yuttaṭṭhāne nayo adinno
nāma natthi.
61. Vibh-​a 122: Idāni abhidhammabhājanīyaṃ hoti. Tattha ‘‘ariyasaccānī’’ti avatvā nippadesato
paccayasaṅkhātaṃ samudayaṃ dassetuṃ ‘‘cattāri saccānī’’ti vuttaṃ
a 124:  evamidaṃ tividhamahāvāraṃ pañcadasakoṭṭhāsaṃ saṭṭhinayasahassapa­
62. Vibh-​
ṭimaṇḍitaṃ abhidhammabhājanīyaṃ nāma niddiṭṭhanti veditabbaṃ.
63. Vibh-​a 6: Abhidhammaniddese pana khaṇena paricchinnaṃ.
176 In t erpr et ing t h e T hr ee PiṬak as

the Abhidhamma analysis focuses on dependent origination’s workings in a single

awareness (citta) that occurs in just one moment (khaṇa):

The Teacher, who has unobstructed knowledge of all dhammas, taught

the disentangled and unknotted mode of conditions through many kinds
of awareness in the Suttanta Analysis as though covering the great earth
and expanding out in space. But since the mode of conditions is relevant
not only to many kinds of awareness, but also concerns a single [moment
of ] awareness, he set out the matrix by means of the method of the
Abhidhamma Analysis to teach in various ways the mode of conditions
in a single moment of awareness, by saying “constructions are conditioned
by ignorance, et cetera.”64

The teaching of the twelvefold conditions goes both wide (through earth and
sky) and deep (into the moment). The brilliance of dependent origination is how
it explains both the conditionality of experience over time (the whole of a human
life and beyond) and within the smallest units of phenomena that we can observe
in the tiniest fragments of time and experience that we can get at.
Since we are on the topic of dependent origination, I cannot resist pausing
to observe that Buddhaghosa finds that explaining the meaning of dependent
origination is, perhaps particularly, difficult by its very nature, and that its ex-
position should be left to the Vibhajjavādins, described as we have seen, as
those who do not launch into their own view and who return again to the
meaning with various methods.65 He says elsewhere that the Buddha’s own
teaching of it is a thing of beauty:  “because he has achieved elegance in
teaching he teaches the Dhamma by various methods.”66 And for his own part,
perhaps in an allusion to Ᾱnanda’s rash notion that he thought he understood
dependent origination, Buddhaghosa pauses in hesitation and humility at

64.  Vibh-​a 199–​200:  Evaṃ mahāpathaviṃ pattharanto viya ākāsaṃ vitthārayanto viya ca
sabbadhammesu appaṭihatañāṇo satthā suttantabhājanīye niggaṇṭhiṃ nijjaṭaṃ paccayākāraṃ
nānācittavasena dassetvā idāni yasmā na kevalaṃ ayaṃ paccayākāro nānācittesuyeva hoti,
ekacittepi hotiyeva, tasmā abhidhammabhājanīyavasena ekacittakkhaṇikaṃ paccayākāraṃ
nānappakārato dassetuṃ avijjāpaccayā saṅkhārotiādinā nayena mātikaṃ tāva ṭhapesi.
65. Vibh-​a 129–​30; see Ñāñamoli, trans., The Dispeller of Delusion, vol. 1, 161–​62. Cf. Vism 522
[XVII.25].
66.  Vism 524:  So desanāvilāsappattattā nānānayeheva dhammaṃ deseti. This is in a larger
passage about how beneficial dependent origination is and about the beauty of the Buddha’s
ways of teaching it from various starting points, and important theme in the Visuddhimagga’s
treatment of it (ch. XVII).
 Disentangling the Tangle 177

the task of commenting on it. He says, “today I want to teach the commen-
tary on the mode of conditions, but I find no footing, as though I have been
plunged into the ocean.”67 But he rallies, noting that the “dispensation as it is
laid down is embellished with many methods of teaching.”68 Its many starting
points and practices make it possible to find somewhere to begin. Properly
understood, the only response to the Buddha’s omniscient enactment of the
endlessly ramifying exploration of conditions is to find oneself “unfooted,”
even while one can eventually find traction with a toehold in the methods that
start to explore it.
What is notable about these distinctions between Abhidhamma Analysis
and Suttanta Analysis is that they do not mention, nor do they practice, the
distinction between conventional and ultimate language—​ terms associated
with paramattha language are used in both in each of the Vibhaṅga’s sections.
Moreover, the Vibhaṅga is an Abhidhamma text, but it is using throughout, very
extensively, the Suttanta Analysis as one of its practices, illustrating again the
porousness of these methods and distinctions of genre.
The third method of analysis is Asking Questions. In the Vibhaṅga, this is
practiced as a rather stylized kind of catechetical matrix that continues with the
twofold and threefold analysis begun in the Dhammasaṅgaṇī. For example, and as
we have seen, a major structure of the Dhammasaṅgaṇī is classifying phenomena
according to their being good (kusala), bad, and indeterminate. Thus the Asking
Questions section of the Vibhaṅga asks what, for example, are the good, bad,
and indeterminate aggregates (and so on for its eighteen categories). As a rather
fixed and formulaic style of analysis in the Vibhaṅga and its commentary, the use
of a series of questions as a matrix or starting place for exposition, is not enor-
mously productive. But it may have inspired a method that Buddhaghosa deploys
in the Visuddhimagga, which is to frequently structure his discussions of a topic
by a series of questions that get at the varieties, functions, ways of being known,
and practical concerns for the purification of phenomena. His opening chapter
on morality (sīla), for example, is structured by a matrix of questions: “what is
morality? In what sense is it morality? What are its characteristic, function, man-
ifestation, and proximate cause? What are the benefits of morality? How many
kinds of morality are there? What is the staining of it? What is the cleansing of

67. Vism 523 (XVII.25); Vibh-​a 130: Vattukāmo ahaṃ ajja, paccayākāravaṇṇanaṃ; Patiṭṭhaṃ

nādhigacchāmi, ajjhogāḷhova sāgaraṃ. Later in this chapter in the Visuddhimagga he mentions
the incident with Ᾱnanda (Vism 583 [XVII.304]).
68. Vism 524 (XVII.25); Vibh-​a 130: Sāsanaṃ panidaṃ nānā-​desanānayamaṇḍitaṃ.
178 In t erpr et ing t h e T hr ee PiṬak as

it?”69 Long answers to each of these follow. Structured analysis through this sort
of questioning is a deeply engrained practice throughout the literature.
The Paṭṭhāna (Starting Points) is the most monumental and daunting of
the Abhidhamma texts. It introduces a list of twenty-​four types of conditions
(paccaya) and applies them to the twenty-​two threefold and the one hundred
twofold classifications of the Dhammasaṅgaṇī. This text takes the methods of
dependent origination beyond their usual account of the manifested conditions
and effects into a deeper treatment of all kinds of conditionality, causes, and
effects at work in experience.70 In the first section of the text we find a formula of
seven questions to be applied to each of the seven main chapters, which can then
allow the further application of the twenty-​four types of conditions to these. U
Nārada helps us understand the algorithmic fecundity of the text:

When the rest of the seven sections are each taken as reference, by turns,
there is a set of 7 × 7 = 49 questions for root condition alone. For the 24
conditions taken singly, therefore, there are 49 × 24  =  1,176 questions.
From this it can be judged that the number of questions for the whole of
the Paṭṭḥāna must be of a very high order. According to the commentary,
the figure is 404,948,533,248 and the subcommentary, 388,778,713,344. In
the Pali text, however, all the questions are not included, but only those
that are necessary for illustrating the types of questions. . . . If all of them
were to be put into print, it would need over 3 crores of books of 400
pages each.71

And this is just the listing of questions; the text goes on to list the answers. And
the text itself is just the “Starting Points” of the analysis of conditionality, rather
than its exhaustive description.72
It is not clear that the text has ever been fully printed out or recited, or that it
would need to be. U Nārada reports, “the Burmese Mahātheras of old estimated
that if the Paṭṭḥāna Text, as expanded, were to be put into print there would be

69.  Vism 6:  Kiṃ sīlaṃ, kenaṭṭhena sīlaṃ, kānassa lakkhaṇarasapaccupaṭṭhānapadaṭṭhānāni,

kimānisaṃsaṃ sīlaṃ, katividhaṃ cetaṃ sīlaṃ, ko cassa saṃkileso, kiṃ vodānanti.
70. U Nārada, trans., Conditional Relations, xi.
71. U Nārada, Conditional Relations, xv.
72. Gethin discusses the meaning of paṭṭhāna, as “point of departure,” and thus “ ‘basis’, ‘origin’,
or ‘cause’, or possibly it might be taken as signifying a ‘course’ or ‘sequence (of conditions).’ ”
The Buddhist Path to Awakening, 30–​31.
 Disentangling the Tangle 179

three cartloads of books.”73 He goes on to note that the classificatory answers the
text produces are not as important as its methods: “the Paṭṭhāna is interesting only
when the methods for arriving at these answers are known.”74 Perhaps all possible
ramifications do not need to be fully articulated once all the methods of classi-
fication are generated or at least signaled. This again suggests that Abhidhamma
texts, here revealed at their most expansive, are classificatory practices aimed not
at a final delivery of ultimate reals, but instead at an enactment of method. This
extraordinary textual performance of enormity may be why Buddhaghosa found
the Paṭṭhāna to be the Buddha’s omniscience at its most oceanic.
The Paṭṭhāna commentary gives several different interpretations of the name
“Paṭṭhāna,” first deriving it from the sounds of its name from pakāra (way) and
ṭhāna (position, stance, condition) to mean something like “the various ways
of conditions.”75 Alternatively it means “analyzing” (vibhajana) following a
sutta that lists several practices: “making understood, establishing, uncovering,
analyzing, showing.”76 Or it means “setting forth” in much the way that cattle
set forth out of their cowpens when let out.77 Buddhaghosa reiterates in this
context his favorite theme of the Paṭṭhāna’s unique expansiveness: “because of
its setting forth and furthering the twenty-​four paṭṭhānas, the Paṭṭhāna alone
achieves the expanded methods in the divisions and classifications of the causes
and conditions, et cetera, such as kusala, et cetera, of the omniscient knowledge
which is in fact not unobstructed when set forth in the Dhammasaṅgaṇī and
other [Abhidhamma treatises].”78 Only in the Paṭṭhāna is the omniscient mind
truly unobstructed, and thus only in it can the practitioner properly begin to
grasp fully the scope of that mind.

73. U Nārada, Conditional Relations, xcix.

74. U Nārada, Conditional Relations, c.
75.  Pañcappakaraṇa-​aṭṭhakathā 342 (in the Burmese pagination):  Kenaṭṭhena paṭṭhānanti?
Nānappakārapaccayaṭṭhena. ‘Pa-​kāro’ hi nānappakāratthaṃ dīpeti, ṭhānasaddo paccayatthaṃ.
76.  Pañcappakaraṇa-​aṭṭhakathā 342 (in the Burmese pagination):  kenaṭṭhena paṭṭhānanti?
Vibhajanaṭṭhena ‘‘Paññāpanā paṭṭhapanā vivaraṇā vibhajanā uttānīkamma’’nti
āgataṭṭhānasmiñhi vibhajanaṭṭhena paṭṭhānaṃ paññāyati. This is quoting the Saccavibhaṅga
Sutta that analyzes each of the four noble truths (M ii.248).
77.  Pañcappakaraṇa-​aṭṭhakathā 343 (in the Burmese pagination):  kenaṭṭhena paṭṭhānanti?
Paṭṭhitatthena. Gamanaṭṭhenāti attho. ‘‘Goṭṭhā paṭṭhitagāvo’’ti āgataṭṭhānasmiñhi yena
paṭṭhānena paṭṭhitagāvoti vutto, taṃ atthato gamanaṃ hoti. This quotes the Cūḷasīhanāda
Sutta (M i.79).
78.  Pañcappakaraṇa-​aṭṭhakathā 343 (in the Burmese pagination):  Iti nātivitthāritanayesu
dhammasaṅgaṇīādīsu anissaṅgagamanassa sabbaññutaññāṇassa hetupaccayādibhedabhinnesu
kusalādīsu vitthāritanayalābhato nissaṅgavasena pavattagamanattā imesu catuvīsatiyā
paṭṭhānesu ekekaṃ paṭṭhānaṃ nāma.
180 In t erpr et ing t h e T hr ee PiṬak as

Finally, Buddhaghosa asserts that however the term is derived, “this treatise
[or performance] called the Paṭṭhāna is to be understood to be for the complete
abolishing of all conditions.”79 The dual sense of “treatise” and “performance” in
the word pakaraṇa captures the enactive quality of this text: it is a textual com-
position and enactment of a practice. And, as always, the point of knowing all
the possible causal and conditioning relationships among phenomena is to break
free of their tangles. These practices are therapeutic and soteriological, aiming at,
Buddhaghosa says, “achieving purification and reaching happiness, the highest
happiness of nirvana, free of grief, unattached.”80

Conclusions: Developing Eyes That See

If we take Buddhaghosa at his word about what he is doing, we do not find a
philosopher arguing for metaphysical views about the nature of reality (either a
metaphysic of the world “out there” or a metaphysic of experience itself ). Instead,
we see a thinker who draws on and extends long-​established analytical practices
to learn to observe—​and thereby reconfigure—​experience. I think the evidence
for this way of reading him is accumulative, and I have begun in this chapter to
assemble the kinds of practices and claims he is making about his project. My
way of reading is to consider passages from his work not in a decontextualized or
isolated fashion but in relationship to the networks of ideas and practices he is
developing throughout to suggest that collectively his theory of the Abhidhamma
texts, his own intellectual practices, and the accounts he gives of his work urge
a conception of his philosophical labors as a kind of phenomenological anal-
ysis that resists ontological leaps and slides. This phenomenological analysis is
an enactment whereby the analytically trained monk, by following the textual
ramifications of the Buddha’s omniscient teaching, begins to approach that om-
niscience himself as his path to purification.81
Buddhaghosa’s theory of Abhidhamma—​what he takes the texts to be—​
is that it enacts phenomenological methods all the way down. To do this, the
Abhidhamma texts rely heavily on “furthest sense” (paramattha) and abstract

79.  Pañcappakaraṇa-​aṭṭhakathā 342–​ 43 (in the Burmese pagination):  Imesaṃ pana

paṭṭhānānaṃ samūhato sabbampetaṃ pakaraṇaṃ paṭṭhānaṃ nāmāti veditabbaṃ. This line is
repeated for each of the three derivations.
80.  Pañcappakaraṇa-​aṭṭhakathā 498 (in the Burmese pagination):  Pāpuṇantu visuddhāya,
sukhāya paṭipattiyā; asokamanupāyāsaṃ, nibbānasukhamuttamaṃ.
81. See Ram-​Prasad’s chapter on Buddhaghosa’s analytic and contemplative practices in Human
Being, Bodily Being, for a close study of the systematic working out of Buddhaghosa’s method
in a specific meditation context.
 Disentangling the Tangle 181

or categorical (nippariyāya) language that, for purposes of understanding and

contemplation, can analyze experience in certain illuminating ways. The kind of
knowledge described as nippariyāya is relevant particularly for seeing what the
Abhidhamma might be: as the abstract categories we can use to interpret expe-
rience, nippariyāya terms are less subject to qualification (than pariyāya terms)
and can thus more directly describe and restructure experience “in every place.”
Abstract or formal categories help us see structures and patterns. The classic
Abhidhamma formulas—​the five aggregates, the twelvefold dependent orig-
ination, the four truths, and so on—​are nothing more (or less!) than abstract
patterns gleaned from the messy particular contexts of living many lives. In this
respect, Buddhaghosa’s insistence that the Abhidhamma was discovered in and
through the biography of the Buddha in his nearly innumerable lifetimes of the
Bodhisatta preparing for the insight of the night of awakening, demonstrate
its inductive quality. But once discovered, the Abhidhamma formulas in turn
become the methods for contemplating and understanding experience that con-
stitute the path.
And so, the Abhidhamma methods enact the Buddha’s omniscience, or at least
begin to get one closer to seeing just how deep and wide it goes. For Buddhaghosa
the Buddha’s omniscient understanding—​to which the practitioner is to be-
come oriented—​is not a matter of arriving at the truth of a set of fixed proposi-
tional claims, but is an ever-​deeper, unobstructed probing into all phenomena, a
“purification of knowing and seeing,” as Buddhaghosa puts it. The understanding
and wisdom his ideal reader is to acquire is not a metaphysical understanding of
reality, but a capacity to learn to see experience in certain ways. The emphasis
on practice, method, and enactment that we have encountered repeatedly is in
keeping with Buddhaghosa’s treatment of understanding (paññā), the aim of the
path that is developed first through morality and concentration in the three-
fold scheme that structures the Visuddhimagga. This is not the place to argue
for a particular conception of vipassanā, nibbāna, or the highest knowing (ñāṇa)
aimed at by the therapeutic and soteriological practices Buddhaghosa describes
in the Visuddhimagga. I  can only submit that Buddhaghosa’s descriptions
of the knowing (ñāṇa) and understanding (paññā) culminating the path, as
articulated in the paññā section of the Visuddhimagga, look a great deal like the
practices described in this chapter. Paññā is defined as a verb, not a set of views
or propositions to which one assents: Buddhaghosa begins his account of paññā
by defining it as “the act of understanding.”82 In other words, in the attainment
of wisdom, one is not presented with a set of metaphysical propositions to which

82. Vism 436 (XIV.435): Kenaṭṭhena paññāti? Pajānanaṭṭhena paññā.

182 In t erpr et ing t h e T hr ee PiṬak as

one is asked to render assent, whether achieved through philosophical argu-

mentation or by dogmatic consent. Rather, one continues in the active work of
learning, questioning, and purifying view via multiple methods, both contempla-
tive and analytic; the practice is the goal.
The dogmas that are sometimes said to be the metaphysical views of Buddhism,
such as the denial of a permanent self (anattā), appear in the Visuddhimagga
as practices, ways of seeing, and contemplations.83 Sometimes scholars have
assumed that the three characteristics of saṃsāra—​nonself (anattā), imperma-
nence (anicca), and suffering (dukkha)—​are, at the end of the day, metaphys-
ical assertions on the true way things are that Buddhist teachings are trying to
establish. As we saw above, Nyanaponika suggests that the analytical methods
of the Abhidhamma produce the ontological insight of anattā. 84 Tricky for this
assumption is the observation that the three characteristics are encountered
only infrequently in the canonical Abhidhamma, and then as specific analyt-
ical devices, among others, to examine phenomena.85 If nonself, suffering, and
impermanence are the ultimate views about reality that the Abhidhamma
methods are meant to establish, then it is surprising that they would be mentioned
rarely, and when mentioned shown simply to be doing analytic work. For its part,
the Visuddhimagga deploys them more often, but always as contemplative and
analytical practices, such as ways of seeing, understanding, and purifying vision.

83.  To cite two typical examples, the three characteristics of saṃsāra involve repeat-
edly seeing constructions (saṅkhāras) as impermanent, painful, and not-​self as a prac-
tice for understanding:  “one again applies the three characteristics to the constructions by
knowl­edge of reflecting and contemplating, and so comprehends” (puna te yeva saṅkhāre
paṭisaṅkhānupassanāñāṇena tilakkhaṇam āropetvā parigaṇhati; Vism 652 [XXI.47], noting
that this needs to be read in terms of the whole passage to see more fully the work of the
meditation practice). There is the contemplation (anupassana) of anattā, among many other
meditation practices in Vism 50 (I.140) and 696 (XXII.117), for instance. Also chapter XX
can serve as a good example of practices that involve defining, contemplating, and applying the
three characteristics to particular phenomena. These “doctrines” are practices.
84. In addition to the treatment of Nyanaponika Thera’s and Bhikkhu Bodhi’s ideas about phe-
nomenology and ontology mentioned earlier (n 11), readers may wish to consider Thanissaro
Bhikkhu’s arguments that the Buddha’s non-​self teachings worked as a type of “strategy,” rather
than as metaphysical tenets (“The Non-​Self Strategy”); but see also Bhikkhu Bodhi’s rebuttal
of this, “Anattā as Strategy and Ontology.” My argument is not about the suttas nor the canon-
ical layer as a whole; rather my focus is Buddhaghosa’s philosophical methods and orientation,
which none of these authors considers systematically.
85. Vibh 70 deploys the three characteristics as a device for analyzing the twelve bases (āyatana),
and the Paṭṭhāna deploys them in several contexts describing those who “observe the aggregates
as lacking self, impermanent, and suffering” (khandhe aniccato dukkhato anattato vipassanti),
a formula also applied to saṅkhāras, sensory experience and its objects, kusala and akusala ex-
perience, and so forth (Myanmar edition of the Paṭṭhāna, I. 135, I.146, I.162). They are ways of
seeing and observing phenomena by the observer.
 Disentangling the Tangle 183

These observations suggest that ways of seeing and understanding,

refined through analytically reductive, definitional, and classification practices
in the discipline of the Abhidhamma, constitute both the practice and the
goal. Buddhaghosa’s ideal reader gets trained in this discipline to encounter the
workings of the Buddha’s omniscient mind which the Buddha prepared for during
his entire bodhisatta career, achieved in the fourth week of his Awakening, and
has been unfolding ever since. The techniques and practices Buddhaghosa finds
in the Abhidhamma and which he elaborates in his own work are the techniques
of seeing, contemplatively and analytically, the phenomena that entangle so that
one might get untangled.
 5

The “Completely Pleasing”

Exegesis on the Vinaya

As we turn to the ways that Buddhaghosa approaches the Vinaya texts,

I argue two main points. First, according to Buddhaghosa, buddhavacana in the
Vinaya Piṭaka expresses the Buddha’s omniscience by demonstrating the Buddha’s
knowl­edge of time. Shifting narratives about time suggest the interconnections
and fluidity of past, present, and future. They also demonstrate the “unfolding”
quality of the Buddha’s omniscience: it is not that the Buddha determined every-
thing important to know on the night of awakening, but that his teaching career
is the spontaneous enactment of a knowledge that is unobstructed as it unfolds
in present and future. It is this enactment that commentary attempts to track.
Second and relatedly, I build on these perspectives on time (and themes suggested
in previous chapters) to show how the textual encounter is for Buddhaghosa an
encounter with the Buddha’s greatness in a way that facilitates personal involve-
ment and transformation for the ideal reader that Buddhaghosa is attempting
to fashion. Buddhaghosa finds the Vinaya immensely pleasing, and he writes
a commentary called “Completely Pleasing” (Samantapāsādikā) to create the
conditions of welcome and gratitude for its reception.
Narratives of time—​being oriented to the distant past of previous buddhas,
studying the distinct occasions that prompt the rules, locating oneself in the
Buddha’s own biographical narrative, noting the transmission of tradition
by Upāli reciting at the First Council, and finally, looking to the future—​are
threaded throughout the Vinaya. We will be following all of these, but we begin
by becoming grounded in the idea of an “occasion.” For Buddhaghosa, the
Buddha’s words are always occasioned, and this feature is hermeneutically signifi-
cant. The occasion (samaya) of a teaching is for the redactors and commentators
a specific moment that becomes, for the monastic reader that this exegesis is
 “Completely Pleasing” Exegesis on the Vinaya 185

training, a vital opportunity for contact with the Buddha’s omniscient ken and
the starting place for the future. Like the origin stories (nidānas) of the suttas,
narrative occasions are the sites where the Buddha’s interactions with his disci-
ples reveal his knowledge of their particular circumstances. The occasion of the
rules is also a moment in which the Buddha’s omniscience is demonstrated in his
capacity to anticipate the future. Incidents become “occasions” when the Buddha
discerns that what has happened in the particular case is significant for the future,
and needs to be developed into a general rule that will govern the entire saṅgha.
For surely, it is an ideal legal code that can address immediate circumstances but
also foresees the future. The Vinaya, in Buddhaghosa’s view, depicts the Buddha
formulating rules that address simultaneously the needs of his immediate saṅgha
even while they serve as general precedent for future cases. In this way they dem-
onstrate the Buddha’s omniscient ken extending to the future.
The canonical Vinaya has a perhaps surprising amount of biographical
material, indicating both the long arc of the Buddha’s awakening as well as his
day-​to-​day episodes with his disciples from which the rules emerge. The Vinaya
does not just give the rules but also records the stories of the rules within the
teaching career of the Buddha and in relationship to the narrative of his awak-
ening. For Buddhaghosa, part of this is because, like the other piṭakas, the
Vinaya is configured to be an encounter with the Buddha’s qualities and story. In
addition, and more specific to this genre on monastic discipline, the biograph-
ical narratives are deemed indispensable to the distinctive type of interactive and
dialogical pedagogy in this genre because the rules are conditioned by the actual
people in his community; it is the community that creates the conditions for
the rules to emerge. Buddhaghosa’s commentarial work on the Vinaya goes even
further than the canonical texts to make the Buddha and his interlocutors avail-
able, in concrete and immediate narrative terms, to the reader. His opening to
the Vinaya offers both contemplative exaltations of the Buddha’s qualities and
narrative development that foster a relationship of personal involvement of the
reader with the Buddha.
The heart of the chapter is a close reading of part of the Samantapāsādikā,
in particular Buddhaghosa’s nidāna to the opening story of the Vinaya text that
begins with the Buddha’s encounter with a Brahmin at a town called Verañjā,
and ends just prior to his laying down the first rule on the occasion of the
first violation of the monastic code of behavior by a monk called Sudinna. It
is through reading this commentary (nearly a hundred pages in the text) that
we learn to approach, with Buddhaghosa, this body of knowledge. But first, we
need to get an overall picture of the Vinaya texts and how they are presented in
various forms.
186 In t erpr et ing t h e T hr ee PiṬak as

The Canonical Vinaya and Textual Embeddedness

The monastic rules—​227 in the case of monks and 311 in the case of nuns—​are
usually taken to constitute the core of the Vinaya. The listing of just these rules is
called the Pātimokkha Uddesa, the ritual recitation occurring at the twice-​monthly
Uposatha gathering; it is also referred to in the Vinaya as the Pātimokkhasutta.
However, while the Vinaya Piṭaka refers to this barebones listing of rules, par-
ticularly in relation to its ritual function, its texts do not at any point include
it in this form. Rather, each rule is embedded within a commentarial narrative,
and the texts are explicit about the necessity of the contextual particulars that
prompted it: the recitation of the Vinaya Piṭaka, for example, must include these
particulars as they were firmly established at the First Council. The first major
book of the Vinaya is the Suttavibhaṅga, literally, the Exposition of the Sutta.
It presents the rules (sutta) embedded in the explanatory context (vibhaṅga) of
the stories that prompted them (here sutta does not mean the Sutta Piṭaka, but
rather the bare monastic rule). The contextual structure in which the rules appear
is specified in the Vinaya itself: Upāli, the reciter of the Vinaya Piṭaka at the First
Council, was queried by Mahākassapa concerning the narrative (vatthu), the
origin story (nidāna), the person (puggala) involved, the rule (paññatti),
corollaries (anupaññatti), offenses (āpatti), and exceptions (anāpatti).1 This
constitutes the “vibhaṅga” (analysis or exposition) of the rule, and, in effect, a ca-
nonical commentary. This style of naming and recording the particulars parallels
Ᾱnanda’s recitation of the Suttanta Piṭaka, where, as we have seen, the contextual
framing is integral to that genre as well.
In addition to the Suttavibhaṅga, which includes the stories of the rules
for both monks and nuns, the Vinaya Piṭaka contains two other books, the
Khandhaka—​itself composed of two parts, the Mahāvagga and the Cūḷavagga—​
and the Parivāra. The two books of the Khandhaka narrate the Buddha’s career
spanning from the night of awakening to the recitation of the Vinaya at the
Second Council a century following his death. It is in the context of his career
that rules called the kammavācā requirements are presented; these concern
procedures for ordination and other legal processes. The Mahāvagga begins with
a narration of the Buddha’s night of awakening and his several weeks’ aftermath
in contemplation, his hesitancy to teach, the deity Brahmā’s plea for him to do so,
his first sermon to his former companions at Isipatana, and the ordination of his
first disciple, Yasa. The book goes on to narrate his teaching career, his growing
influence in attracting lay and monastic followers, his discoursing with kings,

1. Vin ii.286. The Samantapāsādikā echoes this (Sp i.14), and as I describe later, this formula-
tion is the main structure of the Parivāra.
 “Completely Pleasing” Exegesis on the Vinaya 187

and his formulating rules of monastic etiquette and conduct over time, usually
as a response to criticism from laypeople about his monks or a problem arising
from within the community itself. The Cūḷavagga picks up the narrative of the
Buddha’s career and continues through his passing away and the recitation of the
first two councils following his parinibbāna.
Finally, the Parivāra, the last book, returns to a catechetical format of asking
where, to whom, on what subject, and so forth, each rule was established. Its
colophon identifies Dīpa as its redactor (Vin v.226), but little is known about
him. Buddhaghosa insists that the work is, like the rest of the Vinaya corpus,
buddhavacana (though some of this piṭaka concerns events long after the Buddha’s
parinibbāna and is often presented in the words of others).2 The Parivāra extracts
the particulars of the rules from their narrative contexts to provide a type of
question mātikā of the sort we are already familiar with (we saw Buddhaghosa
apply a similar line of questioning to establish when and where the Abhidhamma
was taught, for example). Like the other books of the Vinaya, this is a long text,
despite its clipped, just-​the-​facts style of presentation. Collectively the three main
books of the Vinaya Piṭaka take up six substantial volumes in I. B. Horner’s trans-
lation of them.
Notable from even this brief précis of the Vinaya corpus is just how much
commentary, of different sorts, constitutes these canonical texts. The texts are
mostly stories, even in the case of the Parivāra, which strips down the narratives
to their bare skeletons. Each rule occurs because of an infraction prompting it, a
situation deemed necessary, as we see in what follows, for the Buddha’s prescribing
the rules to begin with, and the narratives often give us considerable detail as
to the circumstances of the Buddha’s laying down a rule. Often there are word
glosses folded into the canonical presentation, particularly in the Suttavibhaṅga
(and in a brief passage in the Khandhaka). And the Parivāra itself is a second-​
order commentarial practice of questioning and contraction.
Perhaps the most striking commentarial material in the Vinaya corpus is the
overarching narrative of the Buddha’s career that constitutes the dominant frame
of the Khandhaka. Previous scholars have noticed this and drawn attention to the
very long, and quite literary biography it presents.3 And the “biography” extends
after the Buddha’s parinibbāna and through the first two councils, presenting a

2. Sp i.28; As 26; Sv i.23.

3.  Finot and Frauwallner suggest an intriguing hypothesis that the Khandhaka narrative
together with the Mahāparinibbāna Sutta at one time constituted a single narrative (see
Frauwallner, The Earliest Vinaya and the Beginnings of Buddhist Literature, ch. 3). In this view,
the Vinaya was derived from an earlier Ur-​text, an independent biography of the Buddha that
followed a “great” and “imposing” plan, and was literary in nature (53). Other scholars have
not been persuaded; Lamotte notes the biographical fragments in the Vinaya but does not
188 In t erpr et ing t h e T hr ee PiṬak as

history of the Vinaya itself to this point. The Cūḷavagga renders an account of
how the three piṭakas were codified and first recited and then the perils of schism
leading to the Second Council at Vesālī that seems to be the basis for all later
commentarial exposition on this point. This material may well have been the seed
of the subsequent Pali chronicle tradition.
As noted, the rules themselves, even with descriptions of the kammavācās, can
be stated quite concisely, but the redactors insisted that they be presented in the
course of the biographical narrative of the Buddha’s teaching career, whether in
the context of the career as described in the discrete episodes of the Suttavibhaṅga
or in the chronological biography of the Khandhaka. This point is highlighted in
the recitation and codification of this piṭaka given at the end of the Cūḷavagga.
When Upāli is asked to recite the Vinaya, he is queried not just for his memori-
zation of each rule, but on where, when, and to whom each rule emerged, just as
Ᾱnanda is for his memory of the suttas. Upāli’s recitation of these particulars for
each rule becomes the Vinaya Piṭaka (Cūḷavagga xi).
Modern scholars have sometimes seen the narratives in which each rule is
embedded as extraneous to the rules themselves; some scholars have suggested
that the bare rules were the “original” text and the commentarial stories grew
up around them.4 Dieter Schlingloff observes that the explanations of the
rules sometimes seem to detract from the meaning, are implausible, and even
des­cend into “a mindless casuistry.”5 In contrast, I. B. Horner argues that the
stories allow for a developed understanding of the exceptions and qualifications
of the rules and the reasoning behind the different grades of penalty for each
rule.6 Jotiya Dhirasekera insists that many of the rules would be unintelligible
without the narrative commentary in which they are embedded, and that at

see a single story here as constituting the core of all subsequent biographies (History of Indian
Buddhism, 176–​78).
4.  See, for example, von Hinüber, A Handbook of Pāli Literature, 13–​14. He notes that the
“introductory story does not always really suit the rule,” and that “from the point of view of
Buddhist law the introductions are unnecessary.” He observes the rules in the Suttavibhaṅga
are not presented “chronologically” in the order in which the violations took place, but much
more systematically (starting with the four defeats and going through each category of offense
in an orderly fashion). Oldenberg has also argued that the Pātimokkha was the earliest kernel
of the Vinaya (as quoted in Dhirasekera, Buddhist Monastic Discipline, 80).
5. That is, “eine geistlose Kasuistik” (Schlingloff, “Zur Interpretaton des Prātimokṣasūtra,” 538).
He gives several examples of what he means. He argues that the stories are not usually impor-
tant (and are sometimes misleading) for interpreting the meaning or reasoning of the rules, but
they can offer interesting narrative and historical sources in their own right.
6. As Horner argues, though she allows that “these groups of stories are apt to be tedious to
Western readers” (The Book of the Discipline, vol. 1, xxxv).
 “Completely Pleasing” Exegesis on the Vinaya 189

least the majority of the stories “serve a useful purpose in the proper under-
standing of the law.”7
I do not have a position on the historical development of these layers of texts,
but rather I focus on what the Vinaya and Buddhaghosa’s commentary on it say
about these questions. I  am interested in exploring the logic of what the texts
themselves assume or insist on and how they present the rules and what is impor-
tant to know about them. For its part, the canonical Vinaya presents each rule
as encircled by a story. It also includes a narrative of the Buddha stating that he
cannot teach the rules without an incident prompting it and in relation to which
it must be understood. In some important way, while the recitation of the bare
Pātimokkha rules has a crucial function and purpose in the Uposatha ritual, and
as a monk Buddhaghosa himself would have known and practiced it in this form,
it is also the case that the Vinaya Piṭaka is given to us as a whole package and
Buddhaghosa is also interested in this “whole.”

Commentarial Vinaya and Embedded Texts

In his commentary the Samantapāsādikā,8 Buddhaghosa insists on the textual
whole and the narrative promptings of individual rules, as revealed by his discus-
sion on a passage in the Cūḷavagga. The canonical passage concerns different kinds
of knowledge monks might bring to deciding legal questions, and suggests that
there may be those who know neither the rule that has been handed down, nor
the analysis of the rule (where analysis, vibhaṅga, refers to the narrative commen-
tary).9 To this point Buddhaghosa says, “one cannot have mastered the Vinaya
without the Suttavibhaṅga” (the analysis of the rule).10 The canonical passage
goes on to consider the question of a monk who knows the Dhamma (that is,
Suttanta knowledge) and the rule, but does not know the analysis of the rule.

7. Dhirasekera, Buddhist Monastic Discipline, 81.

8. The commentaries on the Vinaya attributed to Buddhaghosa are two: the Samantapāsādikā,
which comments on all three books, and the Kaṅkhāvitaraṇī, which comments only on the
Patimokkha rules; I see the Samantapāsādikā as composed by Buddhaghosa’s team or school
(if not by Buddhaghosa himself ), as I discuss in the Introduction. For my purposes, the richer
contextualization and reflection about texts provided by the Samantapāsādikā make it my
focus here; the Kaṅkhāvitaraṇī refers to itself offering commentary in brief and refers to the
Samantapāsādikā for extensive coverage (Kaṅkhāvitaraṇī nāma Mātikaṭṭhakathā, x–​xi).
9.  Vin ii.96:  tassa neva suttam āgatam hoti no suttavibhaṅga. Note that here sutta is the
barebones rules of the Pātimokkha. Buddhaghosa glosses this as:  “the rule was not learned
means the matrix was not learned” (neva suttam āgatam ti no mātikā āgatā); the rules are the
matrix to be expanded (Sp vi.1197).
10. Sp vi.1197: No suttavibhaṅgo ti vinaya na paguṇo.
190 In t erpr et ing t h e T hr ee PiṬak as

Like the person who knows neither the rule nor the analysis of it, such a person
“does not consider the meaning and [so] evades the meaning in the shadow of the
phrasing,” and should for this reason be recused from the legal deliberations.11
Buddhaghosa reads this in strong terms, suggesting that “evades the meaning in
the shadow of the phrasing” means that “having grasped only the mere phrasing,
he prevents the meaning.”12
This suggests that at least in the case of legal exegesis and procedure, essen-
tial meaning comes from the canonical text’s commentarial analysis (vibhaṅga)
in which the rules are embedded.13 It also rejects a purely literal reading of rules
in favor of gleaning meaning from context, that is, the context in which the orig-
inal rule was established. The Vinaya expert (vinayadhara) does not rely on the
phrasing alone, but also on the meaning achieved from considering the anal-
ysis given in the Suttavibhaṅga. Buddhaghosa furnishes as an example the rule
prohibiting monks from handling gold and silver: where Vinaya experts would
be able to see that the prohibition of monks touching gold and silver extends to
all wealth and property connected to gold and silver, those who know only the
wording of the rule might judge in a narrow manner that the rule prohibits hand-
ling only gold and silver.14 We can say that a literal or “plain language” standard
of interpretation is being rejected here, in favor of contextual interpretation.
Elsewhere I have suggested that casuistry or case law is a rational form of legal
reasoning.15
I have also elsewhere explored how Buddhaghosa reads and expands the
contextual details of the first four rules, the most serious infractions in the
Vinaya, which require the offending monk or nun to be disrobed.16 I  have

11.  Vin ii.97:  ‘‘Suṇantu me āyasmantā. Ayaṃ itthannāmo bhikkhu dhammakathiko. Imassa
suttañhi kho āgataṃ hoti, no suttavibhaṅgo. So atthaṃ asallakkhento byañjanacchāyāya atthaṃ
paṭibāhati. Yadāyasmantānaṃ pattakallaṃ, itthannāmaṃ bhikkhuṃ vuṭṭhāpetvā avasesā
imaṃ adhikaraṇaṃ vūpasameyyāmāti.
12.  Sp vi.1197:  Byañjanacchāyāya atthaṃ paṭibāhatīti byañjanamattameva gahetvā atthaṃ
paṭisedheti.
13. This may differ from other uses of language like epithets of the Buddha that deliver their
innate meaning from the sounds of the words themselves, as discussed in ­chapter 2 concerning
the practices of nirutti analysis.
14.  Sp vi.1197:  Jātarūparajatakhettavatthupaṭiggahaṇādīsu vinayadharehi bhikkhūhi āpattiyā
kāriyamāne disvā ‘‘kiṃ ime āpattiyā kāretha, ‘nanu jātarūparajatapaṭiggahaṇā paṭivirato hotī’ti
evaṃ sutte paṭiviratimattameva vuttaṃ, natthi ettha āpattī’’ti vadati.
15. On casuistry, Heim, The Forerunner of All Things, 149, and Huxley’s arguments about this
(“Buddhist Case Law on Theft,” 313–​19).
16. Heim, The Forerunner of All Things, 147–​69.
 “Completely Pleasing” Exegesis on the Vinaya 191

argued that the stories of these rules are richly drawn, both in the canonical
commentary and in the additional narrative details Buddhaghosa supplies in
his commentary. The narratives portray complex people caught in unfortunate
and difficult circumstances. For example, Sudinna, the monk who committed
the first monastic infraction and thus initiated the beginning of the Buddha’s
laying down of all the rules, introduced the offense of sexual intercourse into
the community. The rule itself makes it easy to censure him, and the Buddha
rebukes him harshly; but the commentaries portray a complicated and
unfortunate set of circumstances that provide a morally complex and hu-
mane legal education for the Vinaya expert. Sudinna was hardly a depraved
person succumbing to wanton lust and disgrace. Rather, he was an earnest and
devout monk striving to pursue the monastic life. But his parents, fearing
penury in their old age (as he was their only son and he had no issue), prevailed
upon him to sleep briefly with his former wife to produce an heir. Nor are
his parents particularly craven in this—​their characters are drawn quite sensi-
tively and they are motivated by fear of certain destitution, deep love of their
son, and concern for their daughter-​in-​law. None of this mitigates the harsh
penalty and reprimand he receives—​in this respect the Buddha has an eye on
the future and legislates in no uncertain terms against sexual activity in any
circumstance and for any motivation. But it does present Vinaya law and ped-
agogy in terms of the complicated and situational circumstances in which it
is formulated and in which the redactors and commentators intended it to be
adjudicated.
Here I  wish to go further into the nature of the pedagogy as it is devel-
oped in these texts. I  now see more clearly that these styles of legal reasoning
are connected, at perhaps every juncture, to the Buddha’s biography, and ulti-
mately to the Buddha’s field of knowledge. The stories furnish a complex legal
education conducive to discovering meaning in the rules and applying it to
make legal judgments. They do so because, at least in the interpretative prac-
tice that Buddhaghosa promotes, they give us direct access to the Buddha’s dis-
tinctive pedagogy where a personal and transformative relationship with the
Buddha is being fostered. We turn in the next section to exploring further what
Buddhaghosa thinks this genre is and how his theory of Vinaya shaped his inter-
pretative practice.

Buddhaghosa’s Vinaya
We may recall Buddhaghosa’s claim that on four occasions the greatness of the
Buddha’s knowledge may be glimpsed:
192 In t erpr et ing t h e T hr ee PiṬak as

On four occasions the thundering of buddhas becomes great, their knowl­

edge is approached, their magnitude is understood, and their teaching,
stamped with the three characteristics [of saṃsāra] and connected to
emptiness, becomes deep. What are these? His declaring the Vinaya, the
planes of existence, the workings of conditionality, and the differences of
doctrines.17

I have already discussed how Buddhaghosa takes these to mean several different
genres—​the “planes of existence” refers to the Abhidhamma, and the “differences
of doctrines” refers to the Brahmajāla Sutta, the beginning of the Suttanta.18 Here
we may take a closer look at how he thinks the Vinaya, in particular, gives ex-
pression to buddhas’ greatness. The Vinaya reveals this because in it the Buddha
specifies:

“this is light, this is heavy, this is curable, this is incurable, this is an offense,
this is not an offense, this leads to being cut off, this leads to rehabilitation,
this leads to instruction, this is blamable by the world, this is blamable
because of the rules, in such a case such should be declared.” No one else
is powerful or strong when it comes to the laying down of the cases that
constitute the declaration of the monastic rules. This is the scope of the
Tathāgata alone and not the scope of others.19

The ability to discern the gravity of an action and its significance both for the
individual concerned and the institution of the early and future Buddhist com-
munity is revealed as extraordinary knowledge.
We have also seen that like the other two piṭakas, the Vinaya is not only a basket
that holds a collection of texts, but is a practice in learning or study (pariyatti).

17. Sv i.100: Buddhānañhi cattāri ṭhānāni patvā gajjitaṃ mahantaṃ hoti, ñāṇaṃ anupavisati,
buddhañāṇassa mahantabhāvo paññāyati, desanā gambhīrā hoti, tilakkhaṇāhatā, suññatāpaṭis
aṃyuttā. Katamāni cattāri? Vinayapaññattiṃ, bhūmantaraṃ, paccayākāraṃ, samayantaranti.
See c­ hapter 3, p. 130 for my earlier discussion of this passage.
18. The Suttanta is sometimes represented by its first sutta, the Brahmajāla Sutta: Buddhaghosa
defines the Suttanta Piṭaka as a “teaching on explaining views, for the explaining of views is
taught in it as what opposes the holding of the sixty-​two views” (dvāsaṭṭhidiṭṭhipaṭipakkhabhūtā
diṭṭhiviniveṭhanā ettha kathitāti diṭṭhiviniveṭhanakathā [Sv i.19; As 21; Sp 22]).
19.  Sv i.101:  ‘‘idaṃ lahukaṃ, idaṃ garukaṃ, idaṃ satekicchaṃ, idaṃ atekicchaṃ, ayaṃ
āpatti, ayaṃ anāpatti, ayaṃ chejjagāminī, ayaṃ vuṭṭhānagāminī, ayaṃ desanāgāminī, ayaṃ
lokavajjā, ayaṃ paṇṇattivajjā, imasmiṃ vatthusmiṃ idaṃ paññapetabba’’nti yaṃ evaṃ otiṇṇe
vatthusmiṃ sikkhāpadapaññāpanaṃ nāma, tattha aññesaṃ thāmo vā balaṃ vā natthi; avisayo
esa aññesaṃ, tathāgatasseva visayo.
 “Completely Pleasing” Exegesis on the Vinaya 193

And like the other two piṭakas, it is a kind of training, a kind of avoidance, and has
four kinds of depth. The Vinaya is a training in higher moral precepts, an avoid-
ance of the transgressions since “precepts oppose the transgressions caused by the
defilements,” and it goes deep in four ways (in the Dhamma, in the meaning, in
the teaching, and in comprehension).20 In defining Vinaya vis-​à-​vis the two other
piṭakas, Buddhaghosa says this:

The Vinaya is called “vinaya” by those wise in the meaning of vinaya because
it contains various and distinctive methods and because it disciplines body
and speech.21

In his nirutti analysis the sound “vi” in vinaya delivers vividha (various), visesa
(distinctive), and wise (vidū), and “naya” in vinaya highlights methods (naya).
Specifically, its “methods are various in reference to the fivefold recitation (uddesa) of
the Pātimokkha, which can be divided into the matrix (mātikā) of the seven sections
on offenses beginning with the defeats, and the analysis (vibhaṅga), et cetera.”22 Here
the skeletal matrix is referred to where the Vinaya methods are apparent in the brief
formulation of the recitation. Further, these “have become distinctive because the
methods of supplementary regulations aim to make flexible what is rigid, and it
disciplines body and speech by restraining physical and verbal transgression.”23 These
glosses define and develop what is meant by the Vinaya, its flexibility, its reach, and
its limitations (it does not legislate the disciplining of the mind, for example).
Additionally, he says that “the Vinaya Piṭaka is instruction by authority [or
rule] taught by the Bhagavan who is worthy of authority and whose authority
is abundant.”24 The ultimate ground or basis of the rules is the authority of the
Buddha. One way to understand the pervasive presence of the Buddha and his

20.  As 21–​22; Sp i.22; Sv i.19–​20. See my discussion of these in c­ hapter  1, pp. 50–52, and
Appendix C.
21. Sp 18; As 19; Sv 17: Vividhavisesanayattā, vinayanato ceva kāyavācānaṃ vinayatthavidūhi
ayaṃ, vinayo vinayoti akkhāto.
22. Sp 18; As 19; Sv 17: Vividhā hi ettha pañcavidhapātimokkhuddesapārājikādi sattaāpattikkhandh
amātikāvibhaṅgādippabhedā nayā. The Bhikkhu recitation is divided into five parts: the introduc-
tion (nidāna), the pārājika, the saṅghādisesa, the aniyata, and the vitthāra (which includes the rest of
the rules). The Bhikkhunī recitation has only four parts, excluding the aniyata (Upasak, Dictionary
of Early Buddhist Monastic Terms, 43). The seven kinds of offenses are pārājika, saṅghādisesa,
aniyata, nissaggiya pācittiya, pācittiya, pāṭidesaniya, and sekhiya. There are many descriptions of
these, but see von Hinüber, A Handbook of Pāli Literature, 10–​12, for a brief overview.
23. Sp 19; As 19; Sv 17: visesabhūtā ca daḷhīkammasithilakaraṇappayojanā anupaññattinayā, kā
yikavācasikaajjhācāranisedhanato cesa kāyaṃ vācañca vineti.
24. Sp 21; As 21; Sv 19: vinayapiṭakaṃ āṇārahena bhagavatā āṇābāhullato desitattā āṇādesanā.
194 In t erpr et ing t h e T hr ee PiṬak as

story throughout the Vinaya texts is that it buttresses and displays this very
appeal to authority. John Holt has raised an important question in this regard:

If Buddhist law is traditionally understood to be the result of the Buddha’s

casuistry, on what basis did he render judgments as to rights and wrongs of
bhikkhu behavior? And, on what basis did the bhikkhu community accept
these judgments?25

He goes on to note that the recourse to ultimate authority is well known to

the Western traditions, where traditional legal systems rested ultimately on the
authority of the word of God. Laws are valid because they have been commanded
by an unquestioned authority, whose commands reach us via a line of eminent
teachers. Holt modulates the Buddhist case by some measure, however, to note,
“the Buddha per se is not the source of law, but rather only a figure who makes
judgments and thereby maintains order according to an appeal to the norm
he has recognized as valid.  .  .  . Therefore, the Vinaya rules are legitimate not
because the Buddha sits on the seat of judgment, but because they are applicable
to realizing the Summum Bonum of the religion.”26 He goes on to describe how
the soteriological aims of Buddhism are the ultimate rationale for the rules.
I think this assessment is supported to some degree in the Vinaya’s own
listing of ten reasons for the rules. The Buddha does not set himself up as the
sovereign or arbitrary lawgiver, but specifies that the rules are rationalized by
their pragmatic value. When the Buddha first begins to lay down rules, he asserts
that they are “conditioned by ten reasons.” They (1) promote the excellence of
the Saṅgha, (2)  promote the comfort of the Saṅgha, (3)  subdue bad people,
(4) promote the comfort of well-​behaved monks, (5) restrain the depravities in
this life, (6) ward off the depravities in future lives, (7) gladden those lacking
faith, (8) increase the faith of the faithful, (9) support the Good Dhamma, and
(10) help in discipline.27 It is notable that only one of these concerns the afterlife
and seems directly soteriological. Most of these rationales reflect institutional
concerns of building and maintaining harmonious relationships within the

25. Holt, Discipline: The Canonical Buddhism of the Vinayapiṭaka, 48.

26. Holt, Discipline: The Canonical Buddhism of the Vinayapiṭaka, 52–​53.
27.  Vin iii.21:  tena hi, bhikkhave, bhikkhūnaṃ sikkhāpadaṃ paññapessāmi dasa atthavase
paṭicca—​saṅghasuṭṭhutāya, saṅghaphāsutāya, dummaṅkūnaṃ puggalānaṃ niggahāya,
pesalānaṃ bhikkhūnaṃ phāsuvihārāya, diṭṭhadhammikānaṃ āsavānaṃ saṃvarāya,
samparāyikānaṃ āsavānaṃ paṭighātāya, appasannānaṃ pasādāya, pasannānaṃ bhiyyobhāvāya,
saddhammaṭṭhitiyā, vinayānuggahāya. See my discussion of these in The Forerunner of All
Things, 141–​43.
 “Completely Pleasing” Exegesis on the Vinaya 195

Saṅgha and support for it in the larger society. For example, communities seem
to work better when the well-​behaved are assured that scofflaws will be checked.
The passage suggests that the Buddha’s authority in initiating the rules rests not
on an unquestioned mandate (only), but rather on making available a pragmatic
rationality for how they support a successful community. The rules also make
people happy and faithful.
Still, we have just seen Buddhaghosa make a strong appeal to the authority (āṇā)
of the Buddha, and his frequent reference to the Buddha’s knowledge brings the full
force of that authoritative knowledge to bear in grounding Vinaya law. It is worth
noting, however, that the role that the Buddha’s omniscience plays in authorizing
the Vinaya rules is neither vague nor unsubstantiated. As I have been demonstrating
throughout, Buddhaghosa is very specific about what he means by the Buddha’s
omniscience and its specific and concrete workings. In fact, he sees the interpre-
tative project itself as demonstrating how the texts known as buddhavacana com-
prise the precise movements of an omniscient understanding and pedagogy. The
Buddha’s omniscient mind that authorizes his teachings is accessible to us, at least
partially. The Vinaya texts are the authoritative commands of the Buddha, but those
commands are the very practices of his omniscient understanding of past, present,
and future demonstrable in scripture, which we are learning how to access through
the exegetical project.
We may continue with Buddhaghosa’s general remarks on what the Vinaya is.
The Vinaya is the “dispensation according to offense” in that “beings whose offenses
are many are taught here according to those offenses.”28 This detail is important in
that it suggests that the Vinaya comes into the world necessitated by the commission
of wrong action, and that each rule matches in an appropriate way the transgres-
sion it targets—​features we discuss later. Further, the Vinaya is “a teaching by var-
ious kinds of restraint because in it the kinds of restraint which are the opposing
of transgressions are taught.”29 Ultimately the rules are a part of a larger regime of
restraint that constitutes the monastic life.
Finally, we might here recall that among his general remarks about piṭaka and
genre Buddhaghosa says that two piṭakas, the Abhidhamma and the Vinaya, pro-
duce much pleasure. It seems that “endless joy and happiness arise for sons of good
families abounding in faith, full of serene clarity, and supremely knowledgeable

28.  Sp 22; As 21; Sv 19:  ye te pacurāparādhā sattā te yathāparādhaṃ ettha sāsitāti

yathāparādhasāsanaṃ.
29.  Sp 22; As 21; Sv 19:  ajjhācārapaṭipakkhabhūto saṃvarāsaṃvaro ettha kathitoti
saṃvarāsaṃvarakathā.
196 In t erpr et ing t h e T hr ee PiṬak as

who contemplate these two texts.”30 (As we have noted, this is not always evident
to modern scholars.) In the previous chapter we were introduced to the sublime
happiness of contemplating the Buddha’s omniscient ken through the methods
of the oceanic Abhidhamma. But why would the Vinaya produce such joy and
happiness? He explains: “when monks who are Vinaya experts are contemplating
the Vinaya text, that is, the declaring of the rules according to faults, [they re-
alize] that it is the ken of buddhas alone, not the scope of others [to know] ‘in this
fault, in this transgression there is the declaring of a rule.’ ”31 Buddhas can perceive
in human actions infractions of various degrees of severity and then fashion an
appropriate rule that will come to govern the entire Saṅgha. To know the text
well is to encounter the Buddha, for only his mind could have produced it, and
contemplating the Buddha is itself great happiness for Buddhaghosa. This may be
why he called his commentary “Completely Pleasing.”

The External Nidāna
The Samantapāsādikā begins with what Buddhaghosa calls the “External
Nidāna” (bahirnidāna),32 a substantial portion of text describing events that
happened subsequent to the Buddha’s parinibbāna, and thus “external to” or out-
side of the events described in the Vinaya itself. It gives predictions and their
fulfillments, accounts of the three councils, the succession of teachers, and the
transmission of the teachings from India to Lanka that construct and authorize
the Mahāvihāra historical memory of their tradition. We have considered some
of its material earlier on the general reflections on piṭaka and Vinaya. Though
the Samantapāsādikā as a whole has been little explored by modern scholars, this
portion of the Vinaya commentary is comparatively well studied and translated,
and has been useful for scholars attempting to piece together how the early his-
tory was remembered.33

30.  As 11:  dvepi hi tantiyo paccavekkhantānaṃ saddhāsampannānaṃ pasādabahulānaṃ

ñāṇuttarānaṃ kulaputtānaṃ anantaṃ pītisomanassaṃ uppajjati. This passage is in
Atthasālinī only.
31.  As 11:  Vinayadharabhikkhūnañhi vinayatantiṃ paccavekkhantānaṃ dosānurūpaṃ
sikkhāpadapaññāpanaṃ nāma—​imasmiṃ dose imasmiṃ vītikkame idaṃ nāma hotīti
sikkhāpadapaññāpanaṃ—​aññesaṃ avisayo, buddhānameva visayoti.
32. Sp i.105.
33. This section is translated by Jayawickrama, The Inception of Discipline, and is the only part
of the Pali text translated in English. There is an English rendering of the Chinese version of
the Samantapāsādikā by Bapat and Hirakawa, Shan-​Chien-​P’i-​P’o-​Sha: A Chinese Version by
Saṅghabhadra of Samantapāsādikā.
 “Completely Pleasing” Exegesis on the Vinaya 197

My analysis for the remainder of this chapter focuses not on the external
nidāna but on the opening section of text that happens right after it and
continues for nearly a hundred pages. This is Buddhaghosa’s nidāna on the be-
ginning of the canonical Vinaya, which he calls the “Section on Verañjā,” and his
way of introducing the canonical text (in contrast to the context outside of the
canonical Vinaya regarding its transmission). The Vinaya opens with a story
about the Buddha’s encounter with a Brahmin at Verañjā, and provides several
smaller narratives concerning Mahāmoggalana’s response to a famine in that re-
gion and Sāriputta’s request for the monastic rules. We follow Buddhaghosa’s
commentary on these narratives beginning first of all with what it means to have
an occasion for the Buddha’s teachings (a translated selection from this commen-
tary is provided in Appendix B).

On One Occasion
The opening sentence of the Suttavibhaṅga, the first book of the Vinaya, reads:

On [or because of ] that occasion the Buddha, the Bhagavan, was staying
at Verañjā at the root of the Naleru Pucimanda tree together with a great
community of five hundred monks.34

Buddhaghosa is interested first with what it means that this happened “on” or
“because of ” that occasion. And he notices that the first words of each of the
three piṭakas start with an “occasion.” The Brahmajāla Sutta starting the Suttanta
starts with this sentence:

Thus have I heard: on one occasion the Bhagavan was traveling a long road
from Rājagaha and Nāḷanda with a great community of five hundred
monks.35

And the opening to the Dhammasaṅgaṇī, the first book of the Abhidhamma
opens thus:

34.  Vin iii.1:  Tena samayena buddho bhagavā verañjāyaṃ viharati naḷerupucimandamūle
mahatā bhikkhusaṅghena saddhiṃ pañcamattehi bhikkhusatehi.
35. D i.i: Evaṃ me sutaṃ—​ekaṃ samayaṃ bhagavā antarā ca rājagahaṃ antarā ca nāḷandaṃ
addhānamaggappaṭipanno hoti mahatā bhikkhusaṅghena saddhiṃ pañcamattehi bhikkhusatehi.
198 In t erpr et ing t h e T hr ee PiṬak as

Which are the good phenomena? On whatever occasion that a good

moment of awareness accompanied by joy and associated with knowl­
edge arises [in] the Realm of Desire, having as its object a form, a sound,
a smell, a taste, or a touch, or a dhamma, on that occasion there is contact,
feeling, . . . .36 [what follows is the list of phenomena or dhammas that can
be present, as we saw in the previous chapter].

In each case, the knowledge that follows gets its start on a concrete occasion
(samaya). Buddhaghosa provides a range of possibilities for “occasion”: it can be
a “meeting, a moment, a time, a multitude, a cause (or reason or motivation),
a view, an acquisition, an abandoning, and a comprehension.”37 He goes on to
explain that a teaching can be occasioned by people meeting together, for ex-
ample, or by an opportune moment, or by the Bhagavan having a reason for
initiating it. In the case of the Vinaya passage before him, he says that “here
‘occasion’ means ‘time’—​

Since because at a certain time an idea occurred to Venerable Sāriputta

that became the reason for his asking for the laying down of the Vinaya
rules, so the meaning should be seen here as ‘because of this time.’ ”38

This refers to a story we shall come to shortly that is embedded in the Verañjā
story that has Sāriputta asking the Buddha to give the rules and the Buddha
refusing to do so until they are occasioned by a violation. This becomes the reason
this story is here and it also sets up the Vinaya rules as having been solicited; the
Vinaya is on record as something that was asked for.
But what happens next is important in terms of thinking about the nature
of the three piṭakas and the different ways a teaching can said to be occasioned.
Buddhaghosa notices that in each of the three opening lines of each piṭaka
“occasion” is in a different grammatical case: instrumental in the Vinaya opening
(tena samayena), accusative in the Suttanta (ekaṃ samayaṃ), and locative in the
Abhidhamma (yasmiṃ samaye . . . tasmiṃ samaye). Why is this? He says that the

36. Dhs 8: Katame dhammā kusalā? Yasmiṃ samaye kāmāvacaraṃ kusalaṃ cittaṃ uppannaṃ
hoti somanassasahagataṃ ñāṇasampayuttaṃ rūpārammaṇaṃ vā saddārammaṇaṃ vā
gandhārammaṇaṃ vā rasārammaṇaṃ vā phoṭṭhabbārammaṇaṃ vā dhammārammaṇaṃ vā
yaṃ yaṃ vā panārabbha, tasmiṃ samaye phasso hoti, vedanā hoti.
37. Sp i.107: Samavāye khaṇe kāle, samūhe hetu-​diṭṭhisu; paṭilābhe pahāne ca, paṭivedhe ca.
38.  Sp i.107:  Idha panassa kālo attho. Tasmā yena kālena āyasmato sāriputtassa
vinayapaññattiyācanahetubhūto parivitakko udapādi, tena kālenāti evamettha attho daṭṭhabbo.
 “Completely Pleasing” Exegesis on the Vinaya 199

“production of meaning is different.”39 To summarize his distinctions: he says that

in the Suttanta, the meanings are produced through continuous connections, so
the teachings move from one occasion to the next (and so the accusative case is
appropriate). In the Abhidhamma, there is the sense of a moment as a “locus”
for the possible phenomena (dhamma) that can occur in a particular moment of
awareness. But in the Vinaya, “the occasion for laying down the monastic rules
was difficult to discern even by such ones as Sāriputta, and yet the Bhagavan was
staying in various places not looking for a cause for laying down the rules. And so
the rules came to be declared because of this occasion in the senses of cause and
reason.”40 The instrumental case is used here to suggest that a specific occasion
caused the subsequent laying down of rules.
Moreover, the phrase “because of this occasion” is the prompting of every rule in
the Vinaya, as it is given in the narrative occasions of each rule (“tena samayena” may
be the most frequent phrase in the Suttavibhaṅga). And Buddhaghosa says that what
he says for this first instance of “because of this occasion” should hold for other such
instances as well.41 In other words, the Vinaya as a whole was prompted or caused by
this episode with Sāriputta, and specific Vinaya rules are prompted or caused by the
particular narrative occasions in which they are set.
Now it needs to be noticed that Buddhaghosa specifies right away that this
explanation is not the one offered by the “ancients,” who say that these variants
in case endings are just differences in phrasing and all of them should be under-
stood in the sense of the locative—​“on that occasion.”42 This is an instance of
Buddhaghosa allowing an option to be put forward even while also preserving
the ancient view, without officially deciding for one against the other; he does
not refute or dismiss the possibility of the first view.43 This is one kind of editorial

39. Sp i.107; aññathā atthasambhavato.

40. Sp i.107: Yo hi so sikkhāpadapaññattisamayo sāriputtādīhipi dubbiññeyyo, tena samayena
hetubhūtena karaṇabhūtüena ca sikkhāpadāni paññāpayanto sikkhāpadapaññattihetuñca
apekkhamāno bhagavā tattha tattha vihāsi.
41. Sp i.105: “In other places with something else similar to this but not described, the explana-
tion should be returned to with this description because of having reached the meaning” (Tassa
sarūpena avuttenapi aparabhāge atthato siddhena yenāti iminā vacanena paṭiniddeso kātabbo).
42.  Sp i.108:  Porāṇā pana vaṇṇayanti—​‘ekaṃ samaya’nti vā ‘yasmiṃ samaye’ti vā ‘tena
samayenā’ti vā abhilāpamattabhedo esa, sabbattha bhummameva attho’’ti. Tasmā tesaṃ laddhiyā
‘‘tena samayenā’’ti vuttepi ‘‘tasmiṃ samaye’’ti attho veditabbo.
43.  As noted by Kieffer-​Pülz and von Hinüber, elsewhere in this text Buddhaghosa says
that the view stated last should be seen as “authoritative”: Yathā cettha, evaṃ sabbattha yo yo
aṭṭhakathāvādo vā theravādo vā pacchā vuccati so pamāṇato daṭṭhabbo. “Just as here, so too
everywhere whatever is the view of the aṭṭhakathā, the view of the Theras, or what is stated
last, that is to be seen as authoritative” (Sp i.300). See Kieffer-​Pülz, “Reuse of Text in Pāli
200 In t erpr et ing t h e T hr ee PiṬak as

hand that he exerts—​an alternative view is advanced alongside that of the ancient
commentators (and noted as such).
For my purposes, the creative reading of the case endings for this phrase in
each genre is conceptually important in that it begins to specify how the teachings
constituting rules come to be laid down and because it is another attempt by
Buddhaghosa to consider genre. It should also be noted that this discussion, var-
iously modified, occurs in each of the commentaries on the three piṭakas, and
elsewhere too, as a bit of traveling text that found wide application for setting
out what the text in question was doing and what occasioned its presence in the
world.44 In terms of the Vinaya as a whole, and the particular narrative rules, it
is an explicit statement of the importance of context for the prompting of this
kind of buddhavacana, and the importance of it in the production of meaning.
Specifically, it suggests that Vinaya rules were not given in advance by the Buddha
but only brought into the world when they were caused by actual incidents,
demonstrating the interactive and dialogical nature of the Buddha’s omniscient
engagement with the world (as we see later). We can suggest that where nidāna
was a key category for the contextualization of the Suttanta, in the Vinaya samaya,
with its range of meanings, is a chief category for context and the fashioning of a
certain kind of context in the development of meaning.
The contrast of the instrumental use of “samaya” in the Vinaya may be marked
by a passage on “on one occasion” that occurs only in the Suttanta commentaries
on the suttas, that gives a very different emphasis.

With the expression “on one occasion the Bhagavan” showing that the
Bhagavan is not present at this occasion, the final nirvana of [his] form
body (rūpakāya) is made clear. By it [that is, by this expression] people
only intoxicated with life get stirred up, and it generates urgency for

Legal Commentaries,” 14, and von Hinüber, “Zu einer Göttingere Dissertation über das
Buddhistische Recht,” 107. For me, this statement is most interesting in its use of “or”—​the
view stated last could be authoritative even if it is neither the view of the Theras nor the older
aṭṭhakathā. This seems to suggest some independence from the views of the elders and the
old commentaries. In any case, “authoritative” need not mean the only correct possibility.
Buddhaghosa is not engaging in pūrvapakṣa-​style philosophical argumentation where the
opponent’s view is placed first, refuted, and then the author’s view (the siddhānta) is stated last.
While capable of coming down decisively in favor of one view over another (as for example,
Vism 102–​103 [III.74–​83]), Buddhaghosa often engages in a different, but equally prevalent,
Indic style often seen in śāstric discourse of naming multiple alternatives, without giving his
preferred option.
44. Sv i.32–​33; Ps i.9; A i.12; Pj i.105; Ud i.21–​23, et cetera (the Udāna is ascribed to Dhammapāla).
The Atthasālinī has only an abbreviated version of this discussion, just mentioning the locative
case as relevant to the Abhidhamma passage (As 61).
 “Completely Pleasing” Exegesis on the Vinaya 201

them in the Good Dhamma, for [it conveys that] “the Teacher of a noble
Dhamma such as this, the Bearer of Ten Powers, with a body like a mass of
diamonds, the Bhagavan, has attained final nibbāna. By whom else should
hope for life be generated?”45

This elaboration retrieves a great deal of meaning out of the phrase “on one
occasion the Bhagavan.” The “one occasion” is a marker of a long ago time when
the Buddha was alive and teaching, in contrast to “this occasion”—​that is, now—​
when he is no longer present with us. The distant occasion in the text signals a
difference from our present, where people intoxicated with life and lacking hope
for the life that the Buddha made possible, may become stirred up and newly
inspired. The “one occasion” exposes the gap between Buddha’s time and ours,
even while it closes it: when one reads those words one can participate in that
original moment where the Buddha was present.
But these considerations begin to drift back to the purposes of the Suttanta,
and here we need to press on with the Vinaya text and its commentarial elabora-
tion. Let us now pick up the story of the Brahmin at Verañjā, whose story begins
the Vinaya.

The Beginnings of the Vinaya:

The Brahman at Verañjā
At the First Council as recorded in the Cūḷavagga, Venerable Upāli is asked to
recite where the first offense occurred. It began with the infraction of Sudinna
at Vesālī—​and he thus conveys the idea that the Vinaya starts with this.46 This is
echoed in the Parivāra, which says that the laying down of the rules begins with
Sudinna at Vesālī.47 But in fact, as Buddhaghosa tells us, Upāli starts with the
actual narrative beginning the first book of the Vinaya that begins not at Vesālī,
but at a town at considerable distance from it called Verañjā, and not with the
monk Sudinna, but with a Brahmin householder in an encounter that seems to

45. Sv i.34: Ekaṃ samayaṃ bhagavāti vacanena tasmiṃ samaye bhagavato avijjamānabhāvaṃ

dassento rūpakāyaparinibbānaṃ sādheti. Tena ‘‘evaṃvidhassa nāma ariyadhammassa desako
dasabaladharo vajirasaṅghāta samānakāyo sopi bhagavā parinibbuto, kena aññena jīvite āsā
janetabbā’’ti jīvitamadamattaṃ janaṃ saṃvejeti, saddhamme cassa ussāhaṃ janeti (also Ps i.10;
Spk i.12; Mp i.14; Pj I.110; Ud 25).
46. Vin ii.286.
47. Vin v.49.
202 In t erpr et ing t h e T hr ee PiṬak as

have little if anything to do with monastic law.48 It is with this beginning that
Buddhaghosa starts his commentary on the text and it appears to be quite impor-
tant to him, as his discussion of this first episode takes nearly a hundred pages (in
the Pali Text Society edition) before he gets to Sudinna and the first rule49 (and
so, together with the external nidāna, we are nearly two hundred pages in before
we meet Sudinna).
Reading with Buddhaghosa suggests that there were several reasons why the
redactors may have begun the Vinaya with this story (a story also found in the
Aṅguttara, A  iv.172–​80). The Verañjā story operates as a conversion narrative,
where, like the beginning of the Suttanta, the Buddha encounters a hostile inter­
locutor, and we get to witness his skill in conversion. The Brahmin at Verañjā
accuses the Buddha of many unseemly things but comes to be reoriented in a dra-
matic way through his encounter with him. Buddhaghosa says that the text starts
with this Brahmin to show the Buddha’s help for householders, and reminds us
several times that the Buddha engages this Brahmin out of compassion.50 This
suggests that the account is a demonstration of the Buddha’s effective peda-
gogy in bringing about transformation, a pedagogy that I discuss later. But the
redactors plugged various other bits of text into the Verañjā narrative, including a
ninefold list of praises to the Buddha, and two nested stories (one of Moggallana
and one of Sāriputta), all of which are also, for Buddhaghosa, essential for under-
standing Vinaya pedagogy and bringing about the desired transformation in his
ideal reader.

First, Some Praises

Before we are introduced to the Brahmin at Verañjā, the canonical account starts
with praise of the Buddha in a widely cited praise formulation of nine epithets
(called the “itipiso” formula for its first words, iti pi so), followed by a praise of his
ability to teach, giving the stock formulation of how he “teaches the Dhamma
beautiful at the beginning, et cetera” that we have so often encountered (Vin iii.1).
As the dialogue progresses, the Buddha describes to the Brahmin the night of
his awakening, and thereby converts him. Two styles of developing the Buddha’s

48. Sp i.30–​31; Sp i.106 on Vin iii.1–​10.

49. Sp i.105–​202.
50. Sp i.109: “There, with the report of Verañjā, Venerable Upāli shows the Buddha’s action of
helping householders” (Tattha verañjākittanena āyasmā upālitthero bhagavato gahaṭṭhānugga
hakaraṇaṃ dasseti). Sp i.130 mentions karuṇā, and Sp i.136 mentions anukampa and karuṇā.
 “Completely Pleasing” Exegesis on the Vinaya 203

greatness—​listing his qualities and recounting his biography—​are thus deployed

to situate the reader in relation to the Buddha’s teaching of the rules.
Buddhaghosa lingers on the nine epithets mentioned in the canonical text
that we have already seen in c­hapter  2 when discussing nirutti analysis:  “for
the Bhagavan is a Worthy, Perfectly Awakened, Accomplished in Knowledge
and Conduct, Well-​gone, Knower of Worlds, Highest Coachman of Men to
be Tamed, Teacher of Deities and Humans, the Buddha, the Bhagavan.”51 In
­chapter 2 we considered how Buddhaghosa uses the phonemes of each of these
epithets to develop an extensive tutorial and contemplation of the Buddha’s
qualities, though we had space only to consider the first epithet, “Araha,” in
that example (see also Appendix B). We recall that nirutti analysis of phrasing
(byañjana) gives, through the sound of words, manifold methods—​“a hundred
methods, a thousand methods”—​for developing implicit meaning, and, further,
that this same passage occurs as a “recollection of the Buddha” (buddhānussati)
contemplation found also in the Visuddhimagga. This extended commentarial
analysis of the nine qualities ramifies meaning in a tour de force of exaltation of
the Buddha at the beginning of this piṭaka; it also makes possible the cultivation
(bhāvanā) of a contemplative exercise right as one first encounters the Vinaya.
It is also a useful instance of the self-​awareness and sense of purpose in which
the modular nature of textual reuse—​where large passages are reused, often
to different purposes—​can be demonstrated as part of Buddhaghosa’s craft.
Buddhaghosa here describes the itipiso commentary as a method associated with
Suttanta exegetical methods.

Now I will make a commentary on these words with an extended method

in the beginning of the commentary on the Vinaya for the sake of
delighting the mind with a Dhamma talk connected with the Buddha’s
qualities [and] for the sake of Vinaya experts’ proficiency in the methods
of the Suttanta.52

He gives the large itipiso commentary elaborating the qualities of the Buddha both
to delight the mind and to train Vinaya experts with a method that belongs to

51. Vin 3.1: ‘itipi so bhagavā arahaṃ sammāsambuddho vijjācaraṇasampanno sugato lokavidū

anuttaro purisadammasārathi satthā devamanussānaṃ buddho bhagavā. Note that the Verañjā
story as told in the Aṅguttara does not begin with the itipiso formula; this has been inserted
here by the Vinaya redactors.
52.  Sp i.112:  Idāni vinayadharānaṃ suttantanayakosallatthaṃ vinayasaṃvaṇṇanārambhe
buddhaguṇapaṭisaṃyuttāya dhammiyā kathāya cittasampahaṃsanatthañca etesaṃ padānaṃ
vitthāranayena vaṇṇanaṃ karissāmi.
204 In t erpr et ing t h e T hr ee PiṬak as

another genre. The itipiso commentary is also given in full in the Visuddhimagga,
and partially quoted in many of his commentaries on the Suttanta.53 I note this
as an example of a very deliberate pastiche using and reusing the same and similar
passages across Buddhaghosa’s commentaries; his comment is an editorial one
describing his purposes.54 Here one function is to have Vinaya experts train also
in this style of exegesis regarding the Buddha’s qualities that he here associates
with Suttanta analysis.
The passage is also said to “delight the mind,” and here I  think Charles
Hallisey’s discussion of the itipiso formula is helpful. He notes the “general fas-
cination” the Theravada tradition has with these nine epithets and the signifi-
cant impact of Buddhaghosa’s treatment of them on subsequent medieval Sinhala
literature.55 He demonstrates that in the medieval Sinhala reception of the Pali
commentaries on the itipiso formula, the nine epithets implicate the reader in a
relationship of devotional involvement and worship of the Buddha. They do not
simply record the greatness of the Buddha in a discursive manner, but generate
the reader’s immediate experience of it. The idea of qualities (guṇa) indicates
features of the Buddha that are known by the impact they have on those who
behold him. This seems to be the case for the Pali tradition as well. Buddhaghosa
says so himself in describing the buddhānussati contemplation, a kind of jhāna
practice, in the Visuddhimagga:

Moreover, a monk devoted to this recollection of the Buddha is respectful

and deferential to the Teacher, he acquires fullness of faith, mindfulness,
understanding, and merit, abounds in joy and delight, is capable of for-
bearing suffering and enduring fear and terror, and achieves a perception
of living with the Teacher. And by being inhabited by the recollections
of the Buddha even his body becomes worthy of worship like a shrine
room. [His] awareness inclines to the buddhaland. Should he meet with
an occasion for transgression, shame and apprehension occur as though he
were seeing the Teacher face-​to-​face.56

53. As for example, Sv i.146; Ps i.52; Mp i.112 and ii.287, and in each of these cases he refers the
reader to the Visuddhimagga for the full passage (Vism 198–​213 [VII.2–​65]).
54. On textual reuse, see Cantwell and Freschi, “Introduction: Reuse and Intertextuality in
the Context of Buddhist Texts,” and Kieffer-​Pülz, “Reuse of Text in Pāli Legal Commentaries.”
55. Hallisey, Devotion in the Medieval Literature of Sri Lanka, 121–​22.
56. Vism 213 (VI.69): Imañca pana buddhānussatiṃ anuyutto bhikkhu satthari sagāravo hoti
sappatisso, saddhāvepullaṃ sativepullaṃ paññāvepullaṃ puññavepullañca adhigacchati,
pītipāmojjabahulo hoti, bhayabheravasaho dukkhādhivāsanasamattho, satthārā saṃvāsasaññaṃ
paṭilabhati. Buddhaguṇānussatiyā ajjhāvutthañcassa sarīrampi cetiyagharamiva pūjārahaṃ
 “Completely Pleasing” Exegesis on the Vinaya 205

The analytic and contemplative practice of working through all nine qualities
makes the Teacher present to the monk in an immediate and transformative
way: affective states change; awareness is inclined to a buddhaland; even one’s
body is transformed. There is the suggestion of visualization practice—​the “per-
ception of living with the Teacher”—​that makes one profoundly aware of being
in the presence of the Buddha in a way that can serve as “propadeutic” for the
religious life.57 Inserted here in the nidāna of the Vinaya commentary, the exten-
sive itipiso contemplation serves as preparation for receiving the monastic rules
by restructuring the reader’s experience with the Buddha. The ideal reader is pre-
pared by this contemplation to feel himself to be in the presence of the Buddha
and to receive his teaching of the rules with deference, faith, and delight.
While none of the nine qualities is said to be more important or is emphasized
more than the others, for our purposes, the epithet “Knower of Worlds,” may
be most salient, and in Buddhaghosa’s commentary on this, he touches on
themes now familiar to us about the Buddha’s omniscience. We have space only
to mention some of the highlights here: the Buddha is the “Knower of Worlds
because he knows the world in every way” and “with the endless knowledge of
buddhas he knew, experienced, and comprehended the endless world spheres and
the endless elements of worlds.”58 Moreover, there are three worlds (according
to one of the schemas that he gives): the world of constructions (saṅkhāraloka),
the world of beings (sattaloka), and the world of space (okāsaloka). The “world
of constructions” refers to the infinite ways human experience is fashioned and
the “food,” or intentionality, that feeds constructed awareness; the “world of
space” refers to the cosmology of continents and world-​spheres, including the
heavens and hells, that are here said to be endless (ananta) but yet known by the
Buddha.59 And he knows the “world of beings”:

Because for all beings he knows their inclinations, he knows their latent
tendencies, he knows their doings, he knows their intentions, he knows
beings who have little dust in their eyes and much dust in their eyes, who
have keen faculties and dull faculties, with good attributes and with bad

hoti. Buddhabhūmiyaṃ cittaṃ namati. Vītikkamitabbavatthusamāyoge cassa sammukhā

satthāraṃ passato viya hirottappaṃ paccupaṭṭhāti.
57. Hallisey, Devotion in the Medieval Literature of Sri Lanka, 236. Hallisey notes the possible
links to visualization practice, and even with Pure Land practices, 241–​42.
58.  Sp i.117:  Sabbathā viditalokattā pana lokavidū and Sp i.120:  Evaṃ anantāni cakkavāḷāni
anantā lokadhātuyo bhagavā anantena buddhañāṇena avedi, aññāsi, paṭivijjhi.
59. Sp i.117–​19. This passage is also in Vism 204–​7 (VII.36–​45).
206 In t erpr et ing t h e T hr ee PiṬak as

attributes, teachable and difficult to teach, capable and incapable, and so

this very world of beings is known to him in all ways.60

Again, we see an emphasis on the Buddha’s deep knowledge of the particularities

of beings, the inclinations and dispositions that drive them, and their capacities
to be taught. The reader’s awareness of this knowledge—​among the other quali-
ties of the Buddha—​is cultivated here as preparatory to approaching the Vinaya.
Buddhaghosa also elaborates on the canonical text’s claim that the teaching is
beautiful in the beginning, middle, and end, and that it is perfect in phrasing and
meaning, with passages that we have explored elsewhere, and that constitute the
contemplative jhāna practices called the “Recollections of the Dhamma” in the
Visuddhimagga. Here he is setting up the reader of the Vinaya with a disposition
to find these qualities of teaching in the text that follows.

A Brahmin Sees the Light

Now to the story. The Verañjā story is a conversion narrative that demonstrates
a certain highly valued dimension of the Buddha’s pedagogy. It seems that the
Buddha was once staying under a tree in the outskirts of Verañjā in a place
frequented by recluses, and was approached by the Brahmin Verañja. The char-
acter of the Brahmin is developed by Buddhaghosa as a particular person:  he
is called the Brahmin Verañja because he grew up there, but in fact his name is
Udaya, the name his mother and father gave him. Their conversation takes place
under a certain Pucimanda tree that was named for the yakkha guardian of it,
Naḷeru.61 Through such details we are narratively grounded in the occasion.
In the canonical account, this Brahmin approaches the Buddha and objects
hotly that the Buddha does not pay proper respect to senior Brahmins. A flurry
of accusations follows: the Buddha is not perfect, lacks taste, lacks enjoyment,
teaches nonaction, teaches annihilation, is disgusted, is restrained, practices
austerities, and is not going to another womb.62 The Buddha absorbs these insults
and turns them into praises. In each case he says, “there truly is a way (pariyāya),

60. Sp i.117: Yasmā panesa sabbesampi sattānaṃ āsayaṃ jānāti, anusayaṃ jānāti, caritaṃ jānāti,
adhimuttiṃ jānāti, apparajakkhe mahārajakkhe tikkhindriye mudindriye svākāre dvākāre
suviññāpaye duviññāpaye bhabbe abhabbe satte jānāti, tasmāssa sattalokopi sabbathā vidito.
61. Sp i.111; Sp i.109.
62.  “Is not perfect” (na sampannam), “lacks good taste” (arasarūpo), “lacks enjoyment”
(nibbhogo), “teaches nonaction” (kiriyavādo), “teaches annihilation” (ucchedavādo), “is
disgusted” (jegucchī), “is restrained” (venayiko), “practices austerities” (tapassī), and “not going
to another womb” (apagabbho bhavam) (Vin iii.1–​3).
 “Completely Pleasing” Exegesis on the Vinaya 207

Brahmin, that I could rightly be said to be so,”63 and goes on to suggest that there
is a way to interpret these words to see that they in fact describe good quali-
ties of his achievement. For example, that he “lacks taste” can mean that he has
conquered the taste for sensual pleasures (“taste” has much the same range in
English:  the Brahmin means that he lacks good taste or refinement, but the
Buddha shifts this to mean that he has lost the savor for sensual pleasure). That he
“is disgusted” means not that he is disaffected by the world, but that he is put off
by wrong conduct and evil and bad phenomena. In each case after pointing out
that these words can have these other meanings whereby they cease being insults,
he adds, “but this is not, of course, what you meant to say.”64
The Buddha goes on to describe his experiences of awakening including his
knowledges achieved in the three watches of the night, where he, like a chick
breaking out of its egg shell, broke out of the shell of ignorance. This lesson in
buddhology takes the form, in contrast to the contemplative exegesis of the
itipiso epithets, of narrative biography, relatively brief, but still recounting the
key moments of his awakening, instructing both reader and Brahmin about who
exactly it is that we are dealing with. The Brahmin responds at once to these
teachings and is converted, citing the Buddha’s skill in methods:  “just so, the
Dhamma has been made visible by Gotama by various ways (pariyāya).”65 The
Buddha’s words have “righted what was upside down, revealed what was hidden,
shown the path to the lost, and brought light into darkness so those with eyes
might see forms.”66
Given the salience of pariyāya teachings (those that work modally “from
one side”) in our discussions elsewhere (where we have explored their contrast
with nippariyāya teachings that work “from every side”), it is noteworthy that
a dominant theme in this opening is the Buddha’s capacity to use many sides
of words. Invective turns to commendation when the Brahmin is schooled on
the other sides of his own terms. According to Buddhaghosa, as he turned the
Brahmin’s insults to praises, “the Tathāgata looked straight into the Brahmin’s
eyes with a cooling compassion,” and “dispelling the darkness in the Brahmin’s

63.  Vin iii.1–​3:  Atthi khvesa, brāhmaṇa, pariyāyo yena maṃ pariyāyena sammā vadamāno
vadeyya—​‘arasarūpo samaṇo gotamo’ti. Literally, “there is truly a way, by which way it could
be said of me.”
64. Vin iii.1–​3: no ca kho yaṃ tvaṃ sandhāya vadesī.
65. Vin iii.6: evamevaṃ bhotā gotamena anekapariyāyena dhammo pakāsito.
66.  Vin iii.6:  nikkujjitaṃ vā ukkujjeyya, paṭicchannaṃ vā vivareyya, mūḷhassa vā maggaṃ
ācikkheyya, andhakāre vā telapajjotaṃ dhāreyya—​cakkhumanto rūpāni dakkhantīti. This is a
frequently seen formula of conversion.
208 In t erpr et ing t h e T hr ee PiṬak as

heart he taught just these topics of abuse but with this and that alternative
way (pariyāya).”67 He emphasizes the technique the Buddha deploys here: “the
Bhagavan took up the direct hostility for the purpose of causing a softening of
[the Brahmin’s] heart, and he taught a different sense of the word when he said,
‘there truly is a way, Brahmin.’ ”68 Meeting the Brahmin where he is, the Buddha
can gently turn him around and bring him to understanding. Our unawakened
grasp of things is often partial and our use of language restricted to a single
sense; this one-​sidedness constricts vision and cultivates ignorance, in this case,
the Brahmin’s initial incapacity to discern who the Buddha is. It can be replaced
with the wider and multimodal (and omniscient) perspective that the Buddha
introduces. Indeed, for Buddhaghosa, the Buddha’s omniscience is the framing of
the teaching: when charged with not treating the Brahmin with adequate respect,
the Buddha, “looking with his eye of the knowledge of omniscience,” did not see
anyone worthy of his worship.69
Read in a manner attentive to pariyāya teaching, this narration of the
Brahmin at Verañjā can be seen as a very fitting introduction to the Buddha’s
Vinaya-​style pedagogy. As with the opening of the Suttanta Piṭaka, we start this
genre with dispraise of the Buddha that turns to praise through a gradual lesson
in the agility of his teaching. Here what is admired is not that the Buddha knows
what fuels the holding of all views as it was in the Brahmajāla, but that he knows
the different sides of words and the modal aspects of language. The Buddha can
see various sides of words and can speak to one side of a context while expanding
it to include others. This is effective pedagogy, but we may also consider it skillful
for the formation of legal monastic rules. Rules must be generated out of one
particular context and speak to it, but at the same time, to be binding for the
universal Saṅgha, must speak beyond that context to all situations in the future.
The story of the converted Brahmin ends there for a while as we get two nested
stories (neither one present in the Verañjā story when told in the Aṅguttara), one
on Moggallāna and a second on Sāriputta. When the canonical Vinaya comes
back to the Brahmin we find that he, despite promising to host the Buddha and
his monks during the rains retreat, had forgotten his invitation for the three

67.  tathāgato anukampāya sītaleneva cakkhunā olokento  .  .  . brāhmaṇassa hadayandhakāraṃ

vidhamanto tāniyeva akkosavatthūni tena tena pariyāyena aññathā dassetvā (Sp i.136).
68.  Sp i.132; Mp iv.77:  Athassa bhagavā cittamudubhāvajananatthaṃ ujuvipaccanīkabhāvaṃ
pariharanto aññathā tassa vacanassatthaṃ attani sandassento ‘‘atthi khvesa brāhmaṇa
pariyāyo’’tiādimāha. Buddhaghosa is using much the same commentarial passage in both the
Sp and the Aṅguttara commentary.
69. Sp i.131; Mp iv.76: sabbaññutaññāṇacakkhunā olokentopi.
 “Completely Pleasing” Exegesis on the Vinaya 209

months and only when the Buddha takes leave of him does he remember. This
is rather embarrassing for him, and he prepares a final alms-​giving for them right
before they depart. The Buddha, though he and his monks had been severely
challenged by hunger and famine in that time (as we see in what follows), accepts
this invitation “out of compassion,” according to Buddhaghosa’s reading, and be-
cause he did not want the Brahmin to think the Buddha was angry (kupito) at
him or to report that he was “not omniscient” (asabbaññū).”70

An Eye on the Future: Moggallāna Wants to Turn

the World Upside-​Down!
Despite the Brahmin Verañja taking refuge in the Buddha and promising to host
him and the monks, he forgets them for the rainy season retreat. This is partic-
ularly unfortunate because food in Verañjā had become scarce, and the monks
were reduced to seeking very coarse animal grain from horse-​traders as alms. The
canonical text describes Ānanda taking the coarse grain and pounding it with a
mortar and pestle and the Buddha, though he knows what it is (of course), asks
him about the sound of the pounding (Vin iii.6). This question allows Ānanda
to explain the poor food, and the Buddha to respond, somewhat cryptically, “It
is very good, Ānanda, what has been gained by you good people. Future people
will despise rice and rice boiled in meat.”71 Buddhaghosa explains that because
of what happened here, with Ānanda and his fellow monks eating bad food un-
complainingly, future monks will be restrained about complaining of their fare.72
What happens now has significance for future monks.
The future is also present in an exchange with Venerable Moggallāna, a monk
famous for his magical powers. According to the canonical text (Vin iii.7),
Moggallāna proposes to flip the earth over so that the part underneath which
is fertile could come to surface and provide food. But the Buddha stops him be-
cause this would send beings into serious disarray. Moggallāna offers another
miracle allowing him to hold creatures in his huge hand so that they would be
unharmed, but the Buddha disallows this. Moggallāna proposes instead to spirit
away the monks to Uttarakuru, the northern land of plenty, but the Buddha
prohibits this too.

70. Sp i.199.
71.  Vin iii.6:  Sādhu sādhu, ānanda! Tumhehi, ānanda sappurisehi vijitaṃ. Pacchimā janatā
sālimaṃsodanaṃ atimaññissatī’’ti.
72. Sp i.181.
210 In t erpr et ing t h e T hr ee PiṬak as

Buddhaghosa’s account is richer in detail. Not only does he elaborate on the

upheavals Moggallāna’s plans would cause for beings (which the Buddha’s compas-
sion would not allow him to tolerate), but he reads the whole account with an eye
on the future. If Moggallāna were to do these tricks now then it will create problems
for future monks when facing famine. They will be unlikely to have a magician like
Moggallāna in their ranks and so will be at a loss to bring about a salvific miracle.
People in that future time will then doubt that the monks are “noble persons” be-
cause they lack the marvels of monks in the time of the Buddha, and their sneering
will result in their own demerit and future births in hell.73 In this brief account,
tucked into the original Verañjā story here at the beginning of the Vinaya, we are
again alerted to the importance of time: what the Buddha and his disciples do in
their present will have an impact on people in the future. The Buddha is portrayed
as cognizant of and responsible for the future in which people will compare unfa-
vorably the monks of their time to those of the Buddha’s time, and will suffer karmic
repercussions for this. The Buddha’s knowledge and compassion extend to them.

Sāriputta Makes a Request

The canonical scene (Vin iii.7–​9) now shifts abruptly to a conversation with
Sāriputta that looks back in time in order to interpret the present and the future.
Sāriputta approaches the Buddha with a question he has been pondering: why
is it that the monastic life under three previous buddhas—​Vipassī, Sikhī,
and Vessabhū—​ was short-​ lived, while that of three others—​ Kakusandha,
Konāgamana, and Kassapa—​was long in duration?
The Buddha answers that the three buddhas with the short-​lived dispensations
were “tired of teaching the Dhamma in detail to the monks,”74 and so taught little
of the ninefold teachings, did not lay down the rules, and did not declare the
Pātimokkha.75 And just as flowers not held together by a thread flitter away and
disperse, the various kinds of people who had joined their communities quickly
dissipated away the monastic life. To be sure, those buddhas never wearied of
exhorting their disciples by reading their minds,76 and helping them in this way,
but this practice did nothing for the future of their dispensations. Conversely,

73. Sp i.183.
74. Vin iii.7: kilāsuno ahesuṃ sāvakānaṃ vitthārena dhammaṃ desetuṃ.
75. Vin iii.8: Appakañca nesaṃ ahosi suttaṃ geyyaṃ veyyākaraṇaṃ gāthā udānaṃ itivuttakaṃ
jātakaṃ abbhutadhammaṃ vedallaṃ. Apaññattaṃ sāvakānaṃ sikkhāpadaṃ. Anuddiṭṭhaṃ
pātimokkhaṃ.
76.  Vin iii.8:  “These Bhagavans did not tire of exhorting the monks, encircling mind with
mind” (akilāsuno ca te bhagavanto ahesuṃ sāvake cetasā ceto paricca ovadituṃ).
 “Completely Pleasing” Exegesis on the Vinaya 211

the three buddhas who tirelessly (akilāsu) taught in detail all the scriptures, the
stories of the rules, and the Pātimokkha, had long-​flourishing dispensations. It
does not take Sāriputta long to rise from his seat, arrange his robe on his shoulder,
and respectfully ask that the rules be given: “Bhagavan, it is the time for this, it
is the time for this, Well-​gone Sir! May the Bhagavan lay down the rules for the
monks and declare the Pātimokkha, so that the enduring monastic life will last
long.”77
The Buddha refuses Sāriputta’s request for the rules, instructing him to wait
and that only the Tathāgata knows the time for giving the rules: “Sāriputta, it is not
until certain things occur which are conditions for the oozings (āsava) that that
the Teacher lays down the rules and declares the Pātimokkha to the disciples.”78
Certain conditions develop when the community has been around long enough,
reaches a certain level of development, acquires wealth, and attains a certain level
of learning (Vin iii.10). The “oozings”—​the flowing out of defilements that con-
stitute monastic violations—​are conditioned by the changing circumstances of
the community itself. Only the Buddha can spot when they occur, and they have
not yet done so.
Let us look at Buddhaghosa’s reading of this. Buddhaghosa thinks additional
details and qualifications are required to understand it. First of all, there is the
question of why Sāriputta was not able to determine the answers for himself. He
was able to know that the previous buddhas’ dispensations were long or short,
but not why. But Buddhaghosa gives another possibility, one offered by a cer-
tain Mahāpaduma Thera who argued that he did know the reason, as this is well
within the range of his sixteen arhat knowledges, but that he asked the Buddha
in order to not have it seem that he was putting his knowledge on par with that
of the Buddha.79
Further, it should not be understood that the three buddhas were lazy.
“Though it says there that ‘they were tired,’ they were not tired because of being
indolent, for there is no indolence or weak vitality for buddhas.”80 Furthermore,
the reason those buddhas taught the scriptures only in brief is because beings

77. Vin iii.9: etassa, bhagavā, kālo! Etassa, sugata, kālo! Yaṃ bhagavā sāvakānaṃ sikkhāpadaṃ
paññapeyya uddiseyya pātimokkhaṃ, yathayidaṃ brahmacariyaṃ addhaniyaṃ assa ciraṭṭhitika.
78. Vin iii.9: Na tāva, sāriputta, satthā sāvakānaṃ sikkhāpadaṃ paññapeti uddisati pātimokkhaṃ
yāva na idhekacce āsavaṭṭhānīyā dhammā saṅghe pātubhavanti.
79. Sp i.184. Mahāpaduma Thera was an elder from Lanka said to be expert on the Vinaya whose
opinions Buddhaghosa often quotes (Malalasekera, Dictionary of Pāli Proper Names, 528).
80.  Sp i.185:  Tattha kilāsuno ahesunti na ālasiyakilāsuno, na hi buddhānaṃ ālasiyaṃ vā
osannavīriyatā vā atthi.
212 In t erpr et ing t h e T hr ee PiṬak as

at that time had few defilements and so needed only brief teachings. And the
buddhas did not lay down the rules of the Vinaya because no one ever violated
them. And they did not have the twice-​monthly recitation of the Pātimokkha
until six years into their careers, and then it was a recitation made by the buddhas
called the “Pātimokkha of Exhortation” (ovādapātimokkha), not the “Pātimokkha
of the Rules” (āṇāpātimokkha), which is to be recited by monks fortnightly.81 This
Exhortation is a brief general encouragement to pursue forbearance, abstain from
evil, and purify one’s mind, as found elsewhere in the canon.82 Like all buddhas,
“our Buddha” taught this exhortation at first, but after twenty years he taught
the Pātimokkha of the Rules, whereupon the monks recited it (buddhas them-
selves do not participate in reciting the rules at the Uposatha).83 Furthermore,
Vipassī, Sikhī, and Vessabhū could “mindmeld” with their disciples and exhort
them directly, so they did not need to give them detailed scripture. And these
buddhas lived very long lives and so did their disciples, so that even though their
dispensations did not last long after them, they lasted during the course of their
lives. In the case of the other three buddhas, who were also extremely long-​lived,
their dispensations outlived them a very long time because of their detailed
teachings. Our Buddha, like his disciples, is very short-​lived, but his dispensation
outlives him for a long time.84
As to why the Buddha refused Sāriputta’s request for the rules, Buddhaghosa
points out that the disciples had come from good families and had given up much
wealth and social standing to follow the Buddha. And at this point the least of his
five hundred accompanying monks at Verañjā were Stream-​Enterers. Surely these
fine men would have been offended if the Buddha had presumed they needed
rules of restraint before any evidence of wrongdoing appeared. It would be like
a bad doctor popping a boil before it had reached the surface of the skin, and
thereby only making the patient worse.85

When what is [known by] the professional [doctor] is experienced as

known by oneself, it is received respectfully without blaming [him]. And
so too when what is [known by] the ken of the Omniscient One is expe-
rienced as known by oneself, one receives [the rules] respectfully. And so

81. Sp i.185.
82. Sp i.186, quoting D ii.49–​50, where Buddha Vipassī says these words.
83. Sp i.187.
84. Sp i.191.
85. Sp i.192–​93.
 “Completely Pleasing” Exegesis on the Vinaya 213

the Rules are steadfast and endure according to how they were originally
fixed.86

Only when the disciples are disposed to see for themselves the need for the
rules by a violation will they welcome talk of the rules and their penalties.
Indeed, the story sets up not just a welcome reception of the rules, but a
sense of gratitude for getting them at all; previous dispensations were not so
fortunate.
The story also demonstrates that the rules did not spring fully formed from
the Buddha’s mind and get handed down from on high. Rather they emerged
out of conditions with the Buddha’s particular community in the course of
interacting dialogically with the actual people he knew. The rules are a response
to the people, occasions, and conditions that the Buddha encountered, and it is
in terms of these that they are to be known.
At this point the Vinaya narrative returns to the story of the Brahmin and the
Buddha and his monks’ departure from Verañjā and their setting out for Vesalī
where they will eventually meet with Sudinna, whose unfortunate predicament
and monastic failing I  have already mentioned, and who occasions the laying
down of the first rule. Having been introduced to the Vinaya from this gateway,
we can leave the text and commentary here. But before we do so, we can look at
how Buddhaghosa wraps up what he has been doing. At the conclusion of the
section on Verañjā, Buddhaghosa gives several verses that capture beautifully and
succinctly what he has been trying to do in introducing the Vinaya, and that sum
up what we have learned from this nidāna.

What is “completely pleasing” is in the Samantapāsādikā due to its lin-

eage of teachers, its illuminating distinctions about the narrative and the
original context, its avoiding of the occasions of others, and purifying just
its own occasion, its clearing up of phrasing, its meanings of the words, its
ordered succession of exegesis on the canonical text, its determining the
rules, and its showing the different kinds of analysis and methods.
Since nothing considered (sampassata) here is seen to be displeasing
(apāsādika) to the learned, it is [called] simply “Completely Pleasing”
(Samantapāsādikā).

86. Sp i.195: sake ca ācariyake viditānubhāvo hutvā sakkāraṃ pāpuṇāti; evaṃ na ca upavādādiraho

hoti, sake ca sabbaññuvisaye viditānubhāvo hutvā sakkāraṃ pāpuṇāti. Tañcassa sikkhāpadaṃ
akuppaṃ hoti, yathāṭhāne tiṭṭhatīti.
214 In t erpr et ing t h e T hr ee PiṬak as

The result is [this] commentary on that Vinaya which was spoken by

the Lord of the World with trembling compassion for the world and with
skillfulness in taming those capable of being disciplined.87

The nirutti that explains the name “Completely Pleasing” emphasizes the serenity,
composure, and delight suggested by the affective experience of pasāda, and we
have seen that this faithful pleasure and delight have been cultivated by various
contemplative practices in the text that orient the reader to the Buddha and the
Dhamma. The verses also recapitulate the importance of the commentarial serv-
ices of attending to previous authorities and to the nidāna, narrative, phrasing,
meaning, method, and analysis. The Samantapāsādikā is also noted for how it
speaks to its own samaya—​its time, occasion, or opportunity—​purifying it, and
ignoring those of others. This delightful detail adds a further emphasis on the
highly potent idea of an occasion grounding a teaching. The Samantapāsādikā is
itself prompted by and speaks to its world.

Conclusions
The Vinaya has long been understood as future-​oriented. On his deathbed, the
Buddha is said to have told Ānanda that “when I am dead and gone, the Vinaya
and Dhamma that I have taught will be your teacher.”88 At the First Council, the
monks realize that the future rests on having the Vinaya known: “Mahākassapa,
sir, the Vinaya is the long life of the Buddha’s Dispensation: when the Vinaya is
enduring, the Dispensation is enduring.”89 And the cautionary tale of previous
buddhas’ failure to teach the rules and their resulting short-​lived dispensations
creates much urgency about the teaching of the Vinaya in our time, and a grateful
reception for it when it is finally taught.
Buddhaghosa affirms that the study and teaching of the Vinaya and its com-
mentary will make the Dispensation possible in the future: “even having heard

87.  Sp i.201:  Tatridaṃ samantapāsādikāya samantapāsādikattasmiṃ—​Ācariyaparamparato,

nidānavatthuppabhedadīpanato; Parasamayavivajjanato, sakasamayavisuddhito ceva. /​
Byañjanaparisodhanato, padatthato pāḷiyojanakkamato; Sikkhāpadanicchayato, vibhaṅganaya­
bhedadassanato. /​Sampassataṃ na dissati, kiñci apāsādikaṃ yato ettha; Viññūnamayaṃ tasmā,
samantapāsādikātveva./​ Saṃvaṇṇanā pavattā, vinayassa vineyyadamanakusalena; Vuttassa
lokanāthena, lokamanukampamānenāti.
88. Sp iv.874, quoting D ii.154: yo vo, ānanda, mayā dhammo ca vinayo ca desito paññatto, so vo
mamaccayena satthā’’ti.
89. Sp i.13; Sv i.11: bhante mahākassapa, vinayo nāma buddhasāsanassa āyu, vinaye ṭhite sāsanaṃ
ṭhitaṃ hoti.
 “Completely Pleasing” Exegesis on the Vinaya 215

even a little of my commentary, monks should understand the Vinaya as some-

thing to be taken up and mastered in the presence of Upāli; in this way the
Dispensation will be long-​lasting, enduring for five thousand years.”90 The idea
that a Vinaya expert studying the Vinaya commentary is, in some sense, “in the
presence of Upāli,” suggests that learning the commentary is the conduit from the
past to the present and future. It is the study of the Vinaya that makes what Upāli
knew present now and in times to come.
We also know that the texts remember the Buddha as aware of a future in
which different circumstances will cast the rules in a different light. The well-​
known episode in the Dīgha Nikāya of the Buddha, shortly before his death,
recommending to Ānanda that in future some of the “minor” rules might be
abrogated if the monks desire to do so, indicates a notion of changing times and
a spirit of pragmatic flexibility toward the rules.91 Buddhaghosa’s commentary
on this episode suggests that the Buddha knew (of course) that Mahākassapa,
the authority who presided at the First Council, would refuse to suspend any of
the rules even given the option. In the face of flexibility, Mahākassapa wanted
the Saṅgha to remain vigilant so as to avoid criticism (and to be seen as officially
so doing).92 (Still, Ānanda was formally reprimanded for not having asked the
Buddha which are the minor and which are the major rules, as recounted in the
Cūlavagga, suggesting that the practical matter of not being able to identify the
minor rules may have also influenced Mahākassapa’s decision.93) But our basic
point remains: circumstances and conditions change and some rules are adapt-
able. Looking back in time at previous buddhas underscores this as well: previous
communities did not need the rules at all because general exhortation sufficed.
But times under our own Buddha are different as the violations of Sudinna and
his fellow monks and nuns show. We come to see that rules and the community
they govern exist in a dialogical relationship—​rules come into being to govern a
particular community only if and how they need them. They are generated out
of the circumstances in which they are needed, but they must also—​though only
the Buddha knows fully to what extent—​govern future contexts. We might even

90.  Sp iv.876:  Appeva nāma mama vaṇṇanaṃ sutvāpi bhikkhū upālissa santike vinayaṃ
uggahetabbaṃ pariyāpuṇitabbaṃ maññeyyuṃ, evamidaṃ sāsanaṃ addhaniyaṃ bhavissati,
pañcavassasahassāni pavattissatīti.
91. D ii.154.
92. See An, trans., The Buddha’s Last Days: Buddhaghosa’s Commentary on the Mahāparinibbāna
Sutta, 180–​81.
93. Vin ii.288–​89.
216 In t erpr et ing t h e T hr ee PiṬak as

take the Buddha’s flexibility about rules getting abrogated to suggest an idea that
the Vinaya itself is still unfolding and being shaped by the community it governs.
The curious choice of the episode of the Brahmin at Verañjā to provide
the opening of the Vinaya as it was redacted makes sense when read with
Buddhaghosa. (Ostensibly, the reason given is that it contains the story of
Sāriputta asking for the rules, but the Sāriputta episode was only here injected
into the Verañjā story and is not part of the version we have of it elsewhere).
For Buddhaghosa, one always has to go through the Buddha to get the teaching.
We have seen this in the case of the other piṭakas. The Suttanta opens with the
extraordinary person of the Buddha fully present and visible: in the Brahmajāla
Sutta we find the Buddha’s omniscient ken triumphing over not only the pathetic
rival Suppiya but over all possible dogmatic positions. With Buddhaghosa’s
commentarial additions we encounter the Buddha in terms of both the pro-
saic activities of his day and the extravagantly cosmic dimensions of his being.
In the Abhidhamma, or rather in the nidāna Buddhaghosa constructs for it, we
are treated to the long Jātakaṭṭhakathā biography of the Buddha’s previous lives
culminating in the night of awakening, as the proper entry to the oceanic under-
standing he achieved and taught in this piṭaka. And here, we enter the Vinaya by
way of the Buddha’s compassion for a householder who needs to be reoriented
from his limited vision to a transformative understanding of the Buddha’s proper
epithets and narrative. Notably, the Buddha effects this reorientation in part by
moving the Brahmin away from the one-​sided understanding of his words to give
him a glimpse of the Buddha’s understanding of ideas and words from all sides.
The pedagogical skill of contrasting pariyāya knowledge and nippariyāya know-
ledge is both demonstrated and remarked here. In every case, the nidānas of each
of the three piṭakas require the reader to encounter the Buddha and his extraor-
dinary omniscient mind as it unfolds.
In Buddhaghosa’s Vinaya nidāna (as elsewhere) we find features that Charles
Hallisey has described as “devotional”: “a relentless insistence on bringing formal
self-​involvement to the fore, the downgrading of the fact-​stating (constative)
dimensions of language in comparison with the active, performative dimensions,
and finally, allowing factual language to be continually undermined by figurative
language.”94 Most obviously, the commentary on the nine epithets or qualities
of the Buddha is a contemplative practice that makes possible the ideal reader’s
personal involvement with the Buddha. The itipiso contemplation is not a
description of the Buddha so much as an exercise that makes him immediately
present through visualization and meditative practice. And in fact the whole tale

94. Hallisey, Devotion in the Buddhist Literature of Medieval Sri Lanka, 120.

 “Completely Pleasing” Exegesis on the Vinaya 217

of Verañjā and its substories is performative and enactive. The Buddha performs
a teaching style that demonstrates his expansive and expanding ken—​he knows
what the Brahmin should mean to say and will learn to say even if he cannot say
it yet—​and he brings about the Brahmin’s understanding so that he can see this.
The Buddha casually performs his knowledge of the future in his conversation
with Moggallāna and of the past with Sāriputta, knowledge which bears on the
present occasion by resituating it in a much larger cosmic drama than we might
have suspected from being on the ground under the Pucimanda tree at Verañjā.
If we can speak of its literary qualities, we may note the way the Verañjā
story (in both canon and commentary) grounds us in the singular occasion with
narrative particulars, only to pull back to allow an astonishingly cosmic perspec-
tive and timescape. We are told explicitly by Buddhaghosa of the importance of
the occasion as the prompting of a teaching: the people that the Buddha meets
and his dialogical experiences with them condition the teachings and the active
unfolding of the Buddha’s omniscience. The rules are not handed down ex ca-
thedra but in response to the Buddha’s understanding of events within his par-
ticular community, so these on-​the-​ground events matter. But we are swiftly
reoriented—​or perhaps, disoriented—​with tales of Moggallāna turning the
world upside down and Sāriputta knowing details of previous buddhas many eons
ago. Both are essential—​the immediacy of the concrete occasion that grounds
the teaching locates us in that moment. But that moment is in turn informed by
the cosmic and temporally dazzling perspectives that a day in the life with the
Buddha allows us to glimpse. The literary moves that make this possible in the
telling of the narrative create, ideally, a reader who pivots from the immediacy of
the particular instance to the mind-​blowing possibilities of the infinite.
 Conclusion

It remains only to recapitulate what I think this exercise in reading with

Buddhaghosa has made possible. This project began out of my desire to take
seriously the vision of philology that Sheldon Pollock’s work has opened up, in
particular that part concerned with identifying and learning from commentarial
theories of texts. The sheer enormity of commentarial textual production in the
Indic world—​far and away, most scholarly work in premodern India was done
in commentaries—​beckons us to learn from commentaries themselves how
we might read and interpret them and the root texts on which they comment.
I have tried to do this in the manner of an apprentice looking over Buddhaghosa’s
shoulder as he labors to create readers adequate to read him.
With this in mind, what I have learned from Buddhaghosa about interpreting
Buddhist scripture can be corralled into three general points and briefly
summarized here. First, the more I read of his work the more I become aware of
his insistence that his ideal reader learn to read scripture as an encounter with
the Buddha. His Buddha is both the man on the ground teaching extemporane-
ously to people in his time, and a fabulous reality whose cosmic dimensions and
significance are spectacularly displayed. Either way, Buddhaghosa’s ideal reader
does not get to the Dhamma without first encountering the Buddha as a focus
of contemplative, literary, and philosophical attention. Most salient among
the Buddha’s qualities, at least for the enterprise of interpreting his words, is
his omniscience. Buddhaghosa constantly discovers that a given passage of
buddhavacana is an instance of the Buddha’s omniscient mind unfolding and
enacting an extraordinary intelligence that meets with no obstacle in pene-
trating the minds and experiences of others. To interpret the Buddha’s words is
to encounter the “immeasurable.”
The second key theoretical intervention that Buddhaghosa makes is his insist-
ence on context as a critical idea for interpretative practice. We are guided to a
 Conclusion 219

narrative context given in the canonical texts through a number of his categories—​
nidāna and samaya have been two of the most salient here. Episodes in the
Buddha’s life that reveal his teachings (in the Suttanta) and the monastic rules
(in the Vinaya) always have narrative contexts in which they are embedded and
through which Buddhaghosa urges they be read. This is related to the teachings
on omniscience:  since the Buddha is not just “Awakened,” but is “Awakening,”
his interactions with his community in his forty-​five-​year-​long teaching career
are the active and demonstrative unfolding and expanding of his omniscient
knowledge without obstacle. Buddhaghosa would have his reader be present in
those moments when the Buddha was teaching to see how he was spontaneously
enacting his knowledge of the particular dispositions, inclinations, and needs
of his community; the event becomes the message. Buddhaghosa elaborates
the literary qualities of the Sutta and Vinaya stories to make these narrative
episodes vivid and arresting with regard to their specific details. But as impor-
tant as the narratives of the Buddha’s teaching career are, they are also nearly
always accompanied by a recontextualization to a much grander context where
the Buddha is revealed as a cosmic entity with rays of light streaming from him
into the far corners of the triple-​cosmos, or as the latest buddha in a cosmic
timescape of other buddhas, or a mastermind who knows the future. The cosmic
context resituates the immediate context, of course, which is to be expected
when dealing with an omniscient being. The context and reach of his words and
teachings are always greater than what might be immediately accessible to the dis-
ciple or reader; so the reader must pay attention to not only the specific context
of what is being learnt from him but also the cosmic context of what is yet to be
learned. These contexts demonstrate the enactment of the Buddha’s omniscience
that the interpreter is meant to follow: the Buddha knows the particular and im-
mediate circumstances of his present, even while being aware of a much grander
schema in which they must also be seen.
These are not the contexts salient to modern historicist interpretations of
texts, which often focus on reconstructing a world outside the text (in terms of
social, intellectual, political, and economic history). The historicist paradigm is,
of course, hardly universal, and Buddhaghosa’s choices about how to read scrip-
ture differ so markedly from it that they can cast it and its assumptions into
sharp relief. While just as intensely focused on the past and recreating the distant
contexts of the Buddha’s life and the transmission of the teachings as modern
historians are when reading Buddhist texts, what counts as relevant about past
context is quite different for Buddhaghosa. He does not share modern concerns
for dates, chronology, obsessions with authorship, and the sociological and
intellectual context of ancient Buddhism. Instead, he is fascinated with how the
Buddha engaged pedagogically with the particular people of his day and how this
220 Conc lusion

introduces the reader to the Buddha and his knowledge. The ideal reader is to feel
addressed by this reconstructed past in an immediate, existential, and transform-
ative way. Texts not only convey knowledge but also, more vividly for him, they
transform persons. So the function of context in reading is quite different even
while his notion of context contains within it a certain reading of “history.”
It is not just the episodes of the Buddha’s teachings that have contexts, but
each text, and each piṭaka as a whole, has a context, a nidāna, through which
it is to be entered. The crucial moment of redaction and transmission of these
contexts represented by the idea of the First Council is central to Buddhaghosa’s
thinking about scripture. In many ways, the commentarial enterprise is a con-
tinuation of this transmission and unfolding of the Buddha’s words. We see this
in the creative expansion that commentaries provide on the practice of nidānas
initiated at the First Council, where the introduction and narrative context of
whole texts becomes an indispensable commentarial service.
Buddhaghosa’s nidānas on the three piṭakas have furnished the central textual
focus of this book and they lead to consideration of the third theoretical interven-
tion I wish to highlight here—​Buddhaghosa’s deliberations on genre. Philology
and hermeneutics both attend to genre, because genre sets expectations and the
horizon of possibility for interpretation. By attending to indigenous theories of
genre we can better discern the horizon of possibility set up by the texts them-
selves. This can destabilize genre expectations that are sometimes presumed to be
universal, such as putative distinctions between the literary and the philosoph-
ical, and between reading practice and contemplative practice, to take two sa-
lient examples disrupted in this work. We can dispense with such distinctions
altogether and see what emerges in the texts themselves. I also think that herme-
neutical questions of genre and indigenous theories of texts have implications
for doing Indian philosophy. Investigating the purposes and practices of a tex-
tual tradition should be carried out, if not prior to philosophical investigation, at
least in conversation with it; perhaps in many instances they are not two separate
enterprises at all.
Genre is important not only as a classificatory schema or structure for
organizing scripture (one that is in fact required for helping us see the modu-
larity of smaller textual units that transgress such classifications in the produc-
tion of new meanings), but also as itself a “generative” device, an idea captured in
Buddhaghosa’s idea that the piṭakas are methods and areas of study. His treatment
of each piṭaka’s style of pedagogy yields disciplinary reading practices that in
some instances differ substantially from modern ways of approaching them. In
the case of the Suttanta, reading with Buddhaghosa has shown the interpretative
productivity of the context in which the doctrinal teachings were administered
and remembered. The narrative contexts of the teachings—​the stories of the
 Conclusion 221

interlocutors whom the Buddha taught—​ become the specific instances to

which the more abstract knowledge, as formulated in the Abhidhamma, speaks.
Buddhaghosa has a literary sensibility in appreciating and elaborating these
stories, and he suggests that they can enact, perform, and thus expand the very
doctrinal teachings they frame. In these ways they are performances of the
Buddha’s expansive and engaged knowledge of particulars without obstacle.
For Buddhaghosa, the Abhidhamma is the oceanic and expansive unfolding
of the abstract categories of the Buddha’s knowledge, the knowledge achieved
during the fourth week of awakening, but made possible by his entire Bodhisatta
career of learning the conditions of life. Buddhaghosa’s reading practices treat
Abhidhamma in a manner quite different from that of certain modern scholars.
Innocent of that strand of Buddhist modernism that requires all doctrinal
formulations to yield metaphysical claims, Buddhaghosa sees the Abhidhamma
categories and matrices not as closed or complete lists of ontological and irre-
ducible reals, but as methods and disciplines of analysis that resist closure as they
explore experience in endless ways. To be sure, his is not a presuppositionless phe-
nomenology: he brings to it doctrinal commitments about the specific practices
and tools useful for soteriological purposes. But his phenomenological practice
is not in the service of arguing for or demonstrating a metaphysical argument or
claim; in this sense, it is phenomenological analysis “all the way down” as the very
exercise of understanding that is his path of purification.
Finally, the Vinaya returns the reader to the particular instance, in this case, the
occasion that prompts the handing down of the rules, as the rich imagination of
context through which the redactors and commentators interpreted the Vinaya.
The emphasis on the occasions that generate the rules shows how the Buddha’s
omniscience is enacted, and the Vinaya created, in dialogical encounters with
others, much as it was in the Suttanta. In addition, the future-​looking quality of
the Vinaya demonstrates a noted feature of the Buddha’s omniscience: he could
take a present incident and recognize its importance as an occasion with signifi-
cance for present and future communities. To read this way, the ideal monastic
reader is to see himself as addressed by these occasions.
Consideration of genre is related also to distinctive features of buddhavacana
such as brief and detailed, pariyāya and nippariyāya, neyyattha and nītattha, and
paramattha and sammuti, that are woven throughout Buddhaghosa’s reflection
on the piṭakas. These function less as techne, impersonal general rules, and more
as mētis, rules of thumb learned from following a master in practice. Identifying
these features helps the reader understand the Buddha’s pedagogy (and the
omniscience that pedagogy performs) to set expectations of how a given teaching
is to be received. For example, the Dhamma can wax and wane in detailed or
brief teachings (as the Buddha’s knowledge is itself endlessly expansive yet
222 Conc lusion

sometimes succinct); it can concern the particular instance or the abstraction

from all instances (pariyāya or nippariyāya); it can give spare, analytic and tech-
nical categories in the “furthest sense” (paramattha) that can destabilize conven-
tional (sammuti) categories, even while conventional categories are retained as
pedagogically useful. Some features are aimed at the interpreter and become her-
meneutical protocols: a reader should look for both denotative meaning (attha)
and the suggested meanings implicit in and generated by phrasing (byañjana)
revealed in the phonoaesthetics of the words and other formal properties of tex-
tual passages; the reader should anticipate taking teachings literally (nītattha) or
being required to draw out the meaning (neyyattha).
At the hub of it all stands the Visuddhimagga, Buddhaghosa’s own treatise,
offering a resystematization that transgresses and blurs the disciplinary lines of
the piṭakas to offer a bit of everything in the production of a new whole. To be
sure, the “new” must be couched as the old—​the Visuddhimagga is “merely” the
elaboration of two stanzas spoken by the Buddha, and the expansion of three
categories (sīla, samādhi, and paññā) that are the brief summary of the whole.
Considered in terms of the generative methods that Buddhaghosa has encouraged
us to find in scripture, the Visuddhimagga may be seen as a perfect instance of
the continuing unfolding of buddhavacana. Its highly modular nature shows
the vitality of systematic scholastic practice in producing new knowledge where
passages of scripture and commentary found elsewhere are reworked into disci-
plinary practices (e.g., the itipiso commentary on the Buddha’s qualities becomes
the disciplinary technique called the “Recollection of the Buddha,” deliberations
on scriptural pedagogy are refashioned as “Recollections of the Dhamma,” and
so on). The text itself becomes an exercise in using and reconceiving the practices
of understanding that constitute its goal. And thus, on the mystery and beauty
of how humans inspire one another across time through texts, the “Voice of the
Buddha” continues to speak.
 Ap p e n d i x A

The Recollection of the Dhamma

This extended passage on the nature of the Buddha’s words and the Dhamma occurs
in the Visuddhimagga (Vism 213–​18; VII.68–​88). Part of it can also be found in
Buddhaghosa’s commentary on the Vinaya (Samantapāsādikā i.126–​28), and sections
of it are also in the Nettippakaraṇa and in several commentaries on Khuddaka Nikāya
texts traditionally ascribed to Dhammapāla (such as ItiA ii.85–​86). I translate in what
follows the Visuddhimagga’s version, where it is presented as a kind of meditation prac-
tice, one of the six main “recollection contemplations” treated in chapter VII that calm
and focus attention (the others are Recollections of the Buddha, the Sangha, moral
precepts (sīla), giving (cāga), and the deities (devatā). In the Vinaya commentary
portions of this discussion occur in the context of the opening story of the canonical
Vinaya about the Brahmin Verañja (see Appendix B).
The passage is a commentary on what it means that the Bhagavan’s words are said
to be “well-​spoken,” as asserted in the canonical verse: “The Dhamma is well-​spoken
by the Bhagavan, visible here and now, timeless, inviting one to come and see, leading
forward, and to be experienced by the wise for themselves” (M i.37; A iii.285). Parsing
this sentence leads Buddhaghosa to discuss at length another canonical statement that
then also requires parsing, which asserts that the Buddha “teaches a teaching beautiful
in the beginning, beautiful in the middle, and beautiful in the end with meaning and
phrasing, and makes known the entirely perfect and pure religious life.” This formula
is given many times in the suttas and the Vinaya (as, for example, M i.179; D i.62; and
Vin iii.1, on which our passage in the Samantapāsādikā is commenting), and so the
attention it gets in the commentary has wide application for understanding the quali-
ties of the Buddha’s words.
I have put in italics the words and phrases from the canonical material that are
receiving commentary. I give the Pali term for words that have particular salience in my
224 Appendix A

discussions throughout the book. I have aimed for readability in these appendices to
give the reader a sense of the range and flow of Buddhaghosa’s commentaries.

D i s c uss i o n o n   t h e R ec o l l ec t i o n o f   t h e   D h a m m a
One wishing also to cultivate in solitude and seclusion the Recollection of the Dhamma
should recall the qualities (guṇa) of the ninefold transcendent Dhamma and the
Dhamma of study (pariyattidhamma) in this way: “The Dhamma is well-​spoken by the
Bhagavan, visible here and now, timeless, inviting one to come and see, leading forward,
and to be experienced by the wise for themselves.”
“Well-​spoken”—​in the case of this word the collection that is the Dhamma of study
should be understood also [as well as the transcendent Dhamma], but in the case of
the other [words] just the transcendent Dhamma [is meant]. Now, the Dhamma of
study is indeed well-​spoken because it is beautiful in the beginning, middle, and end with
meaning and phrasing, and because it makes known the entirely perfect and pure religious
life (brahmacariya). For when the Bhagavan teaches even a single verse it is beautiful
in the beginning with its first line of the Dhamma because of being entirely good, it is
beautiful in the middle with its second and third lines, and it is beautiful in the end with
its conclusion. A sutta with a single sequence of meaning is beautiful in the beginning
because of its introduction (nidāna), beautiful in the end because of its conclusion, and
beautiful in the middle because of the rest. A sutta with several sequences is beautiful
in the beginning with its first sequence, beautiful in the end with the last, and beautiful
in the middle with the rest. And further, it is beautiful in the beginning because it has
an introduction and an origin, it is beautiful in the middle because it is furnished with
reasons and examples and because it does not contravene the purposes appropriate to
those it teaches [literally, those who are tractable]. And it is beautiful in the end because
of its conclusion by causing the receiving of faith for the listeners.
The Dhamma of the entire Dispensation is beautiful in the beginning because of
the morality that has become one’s own well-​being, beautiful in the middle because
of calming and insight meditation as path and fruit, and beautiful in the end because
of nibbāna. Alternatively, it is beautiful in the beginning because of morality and con-
centration, beautiful in the middle because of the path to insight, and beautiful at the
end because nibbāna is its fruit. Alternatively, it is beautiful in the beginning because of
the thorough Awakening of the Buddha, beautiful in the middle because the Dhamma
is a thorough teaching, and beautiful in the end because of the thorough attainment of
the Community. Alternatively, having heard it, it is beautiful in the beginning because
of the highest Awakening that is to be discovered by practicing for this purpose, it is
beautiful in the middle because of the awakening of paccekabuddhas, and it is beautiful
in the end because of the awakening of disciples.
It is beautiful in the beginning since one hearing it brings about only what is beau-
tiful just by listening, due to its eliminating the hindrances; it is beautiful in the middle
since one practicing it brings about what is beautiful just by practicing it, due to its
 Appendix A 225

bringing about the happiness of calm and insight meditation; and likewise it is beautiful
in the end since one obtaining it brings about what is beautiful just by the fruit of prac-
tice, due to its bringing about the state similar [to that of the buddhas] when completed.
In this way well-​spoken is due to being beautiful in the beginning, middle, and end.
Now, when teaching the Dhamma, the Bhagavan makes known—​that is, he
illuminates with various methods—​the religious life of the Dispensation and the reli-
gious life of the path, with meaning (attha) appropriate to that because of its excellence
of meaning, and with phrasing (byañjana) because of its excellence of phrasing. With
meaning is from putting together word and sense via showing, making known, opening
up, distinguishing, making clear, and denoting, and with phrasing is by excellence in
syllables, words, phrasing, mode, language (nirutti), and description. With meaning is
because of depth in meaning and depth in penetration, and with phrasing is because
of depth in Dhamma and depth in teaching [it]. With meaning is because its scope is
the analysis (paṭisambhidā) of knowledge (paṭibhāna) and of things (attha), and with
phrasing is because its scope is the analysis of language (nirutti) and of the Teaching
(Dhamma). With meaning means that it generates joyful faith in inquisitive people be-
cause it can be experienced by the learned, and with phrasing means that it generates
joyful faith in ordinary people because it is worthy of confidence. It is entirely perfect
because of being fully complete from lacking anything to be added, and pure because of
being without fault from lacking anything to be subtracted.
Further, it is with meaning due to its particular distinction in understanding by prac-
tice, it is with phrasing due to its particular distinction in scripture by study, entirely per-
fect because of the five compilations of the Dhamma beginning with morality, and pure
due to lacking defilements, occurring for the sake of crossing beyond, and disregarding
worldly benefits. In this way well-​spoken is due to making known the entirely perfect and
pure religious life with meaning and phrasing.
Alternatively, well-​spoken means spoken well from lacking any corruption in
meaning. For the meaning of the teachings of other sects produces corruption since
their teachings speak of obstacles regarding things that are not actually obstacles, and
their teachings speak of salvation about things that are not actually salvific. Because of
this their teachings are badly spoken, but the meaning of the teaching of the Bhagavan
does not produce corruption. For in the case of the teaching spoken in this way, “these
things create obstacles, these things are salvific,” there is no failure in transmitting the
truth. In this way truly the Dhamma of study is well-​spoken.
Now, the transcendent Dhamma is well-​spoken because of its speaking of a prac-
tice appropriate to nibbāna, and of nibbāna appropriate to the practice [of attaining it].
Since he says, “the practice that is the going to nibbāna is properly declared to the dis-
ciples by the Bhagavan, and so practice and nibbāna come together. Just as the water of
the Ganges and the water of the Yamuna come together and meet, so it is too that the
practice that is the going to nibbāna is properly declared by the Bhagavan to the disci-
ples:  practice and nibbāna come together.” And here the Noble Path, because of being
spoken of as the “Middle Path” due to being the middle way that does not approach
226 Appendix A

both extremes, is well-​spoken. And because of being spoken of as “that by which the
defilements are allayed,” since the allaying of the defilements is the fruit of renun-
ciation, it is well-​spoken. And because of being spoken of in terms of its particular
nature of being eternal, et cetera, since nibbāna alone has the particular nature of being
eternal, deathless, a protection, a refuge, and so on, it is well-​spoken. In this way, the tran-
scendent Dhamma is also well-​spoken.
It is “visible here and now” means that here the Noble Path is indeed visible here
and now in that it is to be seen for oneself by a noble person not doing lustful things,
et cetera in their own continuing experience. For [the Buddha] says, “Brahmin, one
colored, overcome, and exhausted by lust, intends out of what is injurious only to one-
self, intends out of what is injurious only to others, or intends out of what is injurious
to both. One experiences painful mental experiences and unhappiness. But when lust
is abandoned one does not intend out of what is injurious to oneself, one does not
intend out of what is injurious to others, and one does not intend out of what is in-
jurious to both, and so does not experience painful mental experiences and unhappi-
ness. And so, Brahmin, the Dhamma is visible here and now” (A i.157). Further, the
ninefold transcendent Dhamma is also visible here and now in that it is seen oneself
with the knowledge that comes from reflection, with whatever way it is understood,
and it is established as something to be approached with faith in another in that way.
Alternatively, a proper view is a view that is praised, and so what is visible here and
now is said to conquer with proper view. Likewise, here the Noble Path conquers the
defilements by the proper view associated with it, the Noble Fruit by its being the cause,
and nibbāna by its being the scope. From this it is like a charioteer said to conquer with
a chariot, and in this way the ninefold transcendent Dhamma is visible here and now
because of proper view.
Alternatively, “what is seen” is called seeing, and what is visible is just what is seen
which has the meaning of seeing. Visible here and now is appropriately interpreted as
well seen. For the transcendent Dhamma is seeing which itself leads one away from the
fear of saṃsāra by the comprehension that comes from cultivation and the comprehen-
sion that comes from realization. From this it is like one to be clothed being appropriate
for [getting] a cloth, and in this way visible here and now is appropriately interpreted
as what is seen.
Timeless means that there is no time between it and the bearing of its fruit. Timeless
just means immediate. For it is said that it bears fruit immediately at its own occurrence,
that is, it gives fruit without a break of any time such as five days or seven days, and so on.
On the other hand, [some things] take time, which means that there is time separating
[a thing’s] occurrence and the bearing of its own fruit. As in what? The good dhammas
that are mundane. But this is timeless, so it is immediate because its fruit follows imme-
diately. This is said in reference only to the Path.
It is inviting one to come and see which means that the occurrence of “come and see
this teaching” makes it appropriate for the motto “come and see.” Why is it appropriate
for this motto? Because it is found and because of its purity. For if someone says there is
 Appendix A 227

gold or money in an empty fist, he cannot also say “come and see.” Why? Because it will
not be found. And of course, if there is excrement or urine to be found, he also cannot
say “come and see” to gladden [people’s] minds by indicating the presence of something
beautiful. On the contrary, this should even be covered up with grass or leaves. Why?
Because of being impure. But this ninefold transcendent Dhamma is found because of
its particular nature, and it is pure like the circle of the full moon in a cloudless sky and
like a genuine precious stone laid out on an orange cloth. From this, inviting one to come
and see is appropriate for the motto “come and see” because it is found and because it
is pure.
Leading forward means that it should be applied. Here is the exegesis:  leading
forward is appropriately interpreted as applying it in one’s own mind by medita-
tion even being indifferent to whether one’s head or clothes are on fire, for applying
means bringing to. Leading forward means that it leads forward. This is appropriate
for the constructed transcendent dhammas. But leading forward is also appropriately
interpreted as applying with one’s own mind to the unconstructed [dhamma]. The
meaning is that it is worthy of being held dear by realizing it.
Alternatively, what brings one to nibbāna is the Noble Path, so it is something to be
applied. What should be applied is what should be realized, so what is to be applied is
the Dhamma of fruit and nibbāna. So leading forward is what is to be applied.
“To be experienced by the wise for themselves” means that it should be experienced
by the wise, beginning with those who come to know something right when it is open
before them, each for oneself: “by me the path is cultivated, the fruit is approached,
cessation is realized.” For the defilements are eliminated for a monastic student by
practicing the path, not by his teacher. Nor does one live well by the attainment of
[one’s teacher’s] fruit, nor does one realize the nibbāna realized by him. From this it is
said that it should be experienced by the wise, for it is to be seen not in the way that an
ornament on someone else’s head is seen, but only by one’s own awareness. This is really
not the scope of fools.
There is a further way that this Dhamma is well-​spoken. How? Because of it being
visible here and now. It is visible here and now because of being timeless. It is timeless
because of inviting one to come and see. And it invites one to come and see, so it is called
leading forward.
On this occasion of recollecting the qualities (guṇa) of the Dhamma consisting of
being well-​spoken, et cetera, one’s awareness will not be obsessed by lust, nor by hatred,
nor will one’s awareness by obsessed by delusion. “On that occasion one’s awareness is
simply straight with reference to the Dhamma” (A iii.285). But because of the depth
of the qualities of the Dhamma or because of applying oneself to recollections of var-
ious kinds, one’s jhāna meditation reaches access, but not absorption. This is called the
Recollection of the Dhamma because of its being produced by recollecting the qualities
(guṇa) of the Dhamma.
Now a monk who has applied himself to the Recollection of the Dhamma thinks
thus: “I have never seen in the past a teacher endowed with this virtue, who taught such
228 Appendix A

a teaching leading forward, nor do I see another one now besides the Bhagavan,” and
he becomes respectful and reverent to the Teacher just because of seeing the qualities
of the Dhamma. With weighty regard for the Dhamma, he attains an abundance of
faith, et cetera, becomes full of joy and delight, defeats fear and terror, is able to for-
bear suffering, [and] comes to have the perception of living with the Dhamma. Even
his body comes to be inhabited by the recollection of the qualities of the Dhamma,
and so it becomes worthy of worship like a shrine. His awareness bends toward the
attainment of the highest Dhamma, and when he comes into contact with an opportu-
nity for transgression, from recollecting that the Dhamma is the good teaching, shame
and apprehension occur. And even if he penetrates no higher he will still be destined
for a good rebirth.
Therefore, a wise person should certainly practice vigilance in this way always with
the recollection of the Dhamma for the sake of great power.
This is the foremost part of the detailed explanation of the Recollection of the
Dhamma.
 Ap p e n d i x B

Commentary on the Section

on Verañja Starting the Vinaya

This selection is Buddhaghosa’s commentary on the opening lines of the Vinaya, right
after the External Introduction (Sp i.105–​15). I  include it here because it includes
passages I discuss at some length at different junctures in the book and because it is a
good example of the range of practices Buddhaghosa offers in his commentaries, even
while focusing on just a few sentences of the root text (I also provide the first paragraph
of the Vinaya on which it is commenting).
In some cases, the commentarial services are quite basic, such as dividing compounds
and glossing words. Buddhaghosa also gives considerable attention to the grammar of
case endings in this passage, where he notices the curious feature that all three piṭakas
start with an occasion (samaya), and that, more curiously, occasion occurs in different
case endings in the opener of each piṭaka. As he suggests in an interpretation that he
acknowledges differs from that of the “ancients” on whom his commentary relies, this
may be important for interpreting the nature of the Buddha’s pedagogy in each of these
genres (I discuss Buddhaghosa’s reasoning through this possibility in c­ hapter 5).
On the particular occasion that prompts the Vinaya, Buddhaghosa embellishes the
story by giving us small but grounding details on the Brahmin Verañja, such as that
his actual name given to him by his parents is Udaya. Such details help us picture and
participate in this occasion. Buddhaghosa uses the presence of the Buddha at appar-
ently two locations (both at the roots of a stately Pucimanda tree and at the city of
Verañjā which is nearby) in a lovely technique that pairs and contrasts the dual nature
and purposes, simultaneously worldly and transcendent, of the Buddha’s teachings.
Buddhaghosa takes up the itipiso epithets in the canonical text and runs with them,
setting up a lengthy commentary on these nine terms that runs for many pages. He says
that he introduces this kind of exegesis to develop proficiency in Suttanta discourse
230 Appendix B

for the Vinaya experts that this commentary is training, as well as to delight the mind
with the Buddha’s qualities. Here, as I argue in the book, the reader is to encounter,
in a transformative way, the Buddha’s qualities (guṇa) prerequisite to encountering
his teachings. The “Suttanta-​style” exegesis that he develops extensively here on the
nine epithets requires close attention to phrasing (byañjana) through linguistic anal-
ysis (niruttipaṭisambhidā) that develops and expands meaning from the sounds of the
phonemes of the canonical words, a generative exegetical practice I discuss at length
in ­chapter 2. To avoid taxing the patience of the reader (niruttis transferred to English
can feel forced, having lost their rich phonoaesthetics), we stop after the first of the
nine qualities. (But for a translation of the whole section, readers may refer to the
Visuddhimagga in Ñānamoḷi’s translation [pp. 192–​209 translating VII.2–​64], where
the commentary on the nine epithets functions as a contemplative practice called the
Recollection of the Buddha, occurring just prior to the Recollection of the Dhamma
provided in Appendix A.) But even focusing on the expansion of meaning from the
sounds in the first epithet “Worthy” (Araha), we come to learn a great deal of the virtues
and triumphs of the Buddha to which the reader is to become oriented.
In the course of his analysis of the Buddha’s quality of being Araha, Buddhaghosa’s
interest in the Suttanta-​style nirutti exegesis gives way to yet another register of dis-
course, one we associate with Abhidhamma, in giving an intricate elaboration of the
causal and modal relationships of the phenomena (dhamma) in dependent origination.
Dependent origination may in fact be the most generative of doctrinal matrices and
here it gets launched from the mention of “the wheel of saṃsāra” that pops up in the
midst of analyzing the word “Araha.” Buddhaghosa seizes the moment to give an anal-
ysis of how dependent origination can be mapped onto the hub, spokes, and rim of
the wheel of saṃsāra, and further, that beginning with the condition of ignorance, the
conditions for the other eleven factors of dependent origination can be mapped onto
experience of the life course within saṃsāra. In this way, we get a sampling of several
highly fecund commentarial practices expanding meaning and possibility.

T h e S ec t i o n o n   V er a ñ ja
First, the root text—​this is the first paragraph of the canonical Vinaya (Vin iii.1):
Because of that occasion (tena samayena, or “on that occasion”) the Buddha, the
Bhagavan was staying at the root of the Naḷeru Pucimanda Tree at Verañjā, together a
great monastic community of five hundred monks. The Brahmin Verañja heard “Sir,
the renouncer Gotama, son of the Sakyas, gone forth from the Sakya clan, is staying at
Verañjā at the root of the Naḷeru Pucimanda Tree together with a great monastic com-
munity of five hundred monks. And a beautiful sound of fame has gone out regarding
Lord Gotama in this way: ‘because he is thus [itipiso] the Bhagavan is Worthy (Araha),
Perfectly Awakened, Accomplished in Knowledge and Conduct, Well-​gone, Knower
of Worlds, Highest Coachman of Men to be Tamed, Teacher of Deities and Humans,
 Appendix B 231

the Buddha, the Bhagavan.’ Having realized with his own higher knowledge this world
with its deities, māras, brahmas, renouncers, Brahmins, creatures, gods, and humans,
he makes it known. He teaches a teaching beautiful in the beginning, beautiful in the
middle, and beautiful in the end with meaning and with phrasing and makes known
the entirely perfect and pure religious life. It is truly good to see an arahat such as this.”

T h e C o m m en ta ry o n   t h e S ec t i o n o n   V e r a ñ ja
This is the beginning of Buddhaghosa’s commentary on this first paragraph:

Now I will make a commentary on the meaning of the Vinaya showing with
various ways the meaning of the passage beginning with “because of that” [or
“on that,” tena].

I will make a commentary on the meaning of “because of that occasion the Buddha, the
Bhagavan . . .” according to how it has been described, as follows. The sentence with the
designation “because of that” is uncertain. A reapplication can be done by establishing
the meaning later with a sentence not mentioned here but similar to it. That is, it is
established below that an idea [occurring to] Venerable Sāriputta became the reason
for his asking for the laying down of the Vinaya. Therefore, the connection should be
understood in this way here: since it was because of a certain occasion when the idea was
generated, it is “because of that occasion that the Buddha, the Bhagavan, while staying at
Verañjā . . . .” And this is appropriate throughout the entire Vinaya, so that whenever
there is “because of that,” whether earlier or later, the reapplication should be done there
by the sentence “because of a certain [occasion].”
Now this is pointing out only the beginning [of the Vinaya] for it is said—​“ because
of that, monks, I will lay down the rules for monks, that is, because Sudinna indulged
in sexual misconduct; since he indulged, I will lay them down.” And indeed the reappli-
cation established in this way via the meaning earlier with the expression “because of a
certain” is appropriate [in this case too]. “Because on a certain occasion when Dhaniya,
the son of Kumbhakāra, took wood belonging to the king, et cetera, because of that
occasion the Buddha, the Bhagavan, was staying at Rājagaha”—​in this case the reappli-
cation with the wording “because of a certain” is established according to the meaning
later as the meaning of the wording “because of that” has been stated.

Now here “because of that occasion” has the word “occasion (samaya)” which
appears in [the sense of ] a meeting together, a moment, a time, a crowd,
a cause [or motive or reason, hetu], a view, an acquisition, an abandoning, a
comprehension.

For instance, [in sentences] such as—​“perhaps we may meet together if the time and
occasion arise”—​there is the meaning of meeting together. “There is but one moment
232 Appendix B

and one occasion, monks, for living the religious life” is [the sense of ] moment. “It is
an occasion of heat, an occasion of fever” is time. “There was a great occasion on the
mountain” is crowd. “You are not comprehending this occasion, Bhaddāli, that the
Bhagavan is staying at Sāvatthi and the Bhagavan will come to know me thus ‘a monk
called Bhaddāli is not fulfilling the precepts of the Teacher’s Dispensation,’ ”—​here it
means cause. “Because of this occasion the recluse Uggahamāno is staying at the Mallika
single-​hall monastery near the Tinduka tree for debating occasions,” here it means
view. In [sentences] such as this it is an acquisition:  “one steadfast is called learned
from comprehending the meaning, whether meaning in relation to visible phenomena
here and now, or meaning in relation to the future life.” And it is an abandoning in
this instance: “from the complete penetration of conceit he made an end of suffering.”
And it has the meaning of comprehension in this:  “the meaning of suffering can be
directly known in that there is suffering in the sense of oppression, in the sense of what
is formed, in the sense of misery, and in the sense of change.”
But here the meaning [of occasion] is time. Since because at a certain time an idea
occurred to Venerable Sāriputta that became the reason for his asking for the laying
down of the Vinaya, so the meaning should be seen here as “because of that time.”
Here someone says: “now why is it that in the Suttanta the expression ‘one occasion’
is made in the accusative case, and in the Abhidhamma ‘on whatever occasion in the
realm of desire’ is in the locative case, but here this is not done and the expression
‘because of that occasion’ is made in the instrumental case?” This is because here the
production of meaning is different. How so? In the Suttanta the meaning arises in a way
continuously connected. For the occasion that the Bhagavan taught suttas such as the
Brahmajāla, he dwelled with the sublime abiding of compassion continuously on that
occasion. Therefore, the accusative case is made there for the sake of illuminating that
meaning. And in the Abhidhamma there is generated the sense of locus and the sense
of what characterizes a state [of being] by another state. “Locus” as occasion can have
the sense of time and the sense of an aggregation, and so here a state is characterized
by a state of an occasion that is constructed by a cause that is the occurring together
in a [single] moment of such described phenomena as contact, et cetera. Therefore,
expecting this or that meaning, occasion is said elsewhere with the accusative and the
locative, but here it is just with the instrumental.
But the ancients explain: “ ‘one occasion,’ ‘on whatever occasion,’ or ‘because of that
occasion’ is a difference only in expression, and the meaning in each instance is simply
locative.” Therefore, by their view, the meaning of “because of that occasion” should be
understood as “on that occasion.”
We will later explain [or praise, vaṇṇayissāma] the meaning of the words “the
Buddha, the Bhagavan.” “He was staying at Verañjā”—​now here, Verañjā is the name
of a certain city, that of the Verañjas, and is in the locative in the sense of nearby. He
“was staying”—​this is an elucidation of a general [word] for a certain [specific kind of ]
staying among the ways of staying that include the postures, the godly abidings, the
divine abidings, and the noble abidings, but here it is the elucidation of the combination
 Appendix B 233

of certain postures among the postures of standing, going, sitting, and lying down; be-
cause of this, he stays should be understood as the standing, going, sitting, and lying
down by the Bhagavan. Cutting off the discomfort of one posture with another posture,
he carries himself and moves about, and so it is said that he “was staying.”
“At the root of the Naḷeru Pucimanda Tree”—​here Naḷeru is the name of a yakkha.
Pucimanda is a neem tree, and root means nearby. This word root is seen in here: “digging
out the roots, even the fine rootlets and fibers,” as root in the [literal sense] of root. [In
statements] such as “greed is the root of what is bad” it is a unique type of cause. “At the
time of midday the shade cast, where the leaves fall, to that extent is the root of a tree,”
which means nearby. And here what is intended is nearby, since the meaning here should
be seen as near the Pucimanda Tree frequented by the yakkha Naḷeru. The Pucimanda,
pleasing, delightful, and presiding like a king over other trees, stood there in the midst
of the coming and going near to the city. Then the Bhagavan, going to Verañjā, staying
at that appropriate place, stayed underneath and nearby that tree. Because of this it is
said, “he was staying at Verañjā at the root of the Naḷeru Pucimanda Tree.”
Now what if [someone were to object that] to the extent that the Bhagavan
was staying at Verañjā then “at the root of the Naḷeru Pucimanda Tree” need not be
mentioned, or that he was staying there and so “Verañjā” need not be mentioned, for it
is not possible to stay in both places on the same occasion at the same time? It should
not be seen this way, for we mentioned “the locative case in the sense of nearby.” That
is why when a herd of cattle goes near the Ganges and the Yamuna they say, “they go by
the Ganges, they go by the Yamuna.” So, here, to the extent that he stays at the root of
the Naḷeru Pucimanda near to Verañjā, [the text can say] “he was staying at Verañjā at
the root of the Naḷeru Pucimanda Tree.” The word Verañjā is for explaining the town and
surrounding pastures. The words root of the Naḷeru Pucimanda Tree is for explaining a
place appropriate for and frequented by renouncers.
Now, with the report of Verañjā, Venerable Elder Upāli [the reciter of the Vinaya]
shows the Bhagavan’s helpful action for householders, and by the report of the root of
the Naḷeru Pucimanda [he shows his] helpful action for renouncers; likewise by the
former [he shows how Verañjā] was helpful for avoiding weariness in getting requisites,
and with the latter [staying at the Tree, he shows] the means of avoiding luxurious living
from giving up the desire for things; and by the former his inclination for teaching the
Dhamma, by the latter his inclination to be alone; and by the former his going toward
[others] with compassion, by the latter retreating out of wisdom; by the former his in-
clination to accomplish the welfare and benefit of beings, by the latter his freedom from
defilement in effecting the welfare and benefit of others; by the former the comfortable
life marked by not giving up Dhammic happiness, and by the latter, that marked by
the application of the Dhamma of superior persons; by the former his great help for
humans, and by the latter, for deities; by the former the condition of being nourished
in the world by one born in the world, and by the latter [the condition of being]
unsmeared by the world; by the former [Upāli shows that] there is the accomplishment
of that aim for which the Bhagavan was born [indicated] by the saying “monks, there is
234 Appendix B

one person who arising, comes into being for the welfare of the many, the happiness of
the many, for compassion for the world, for the welfare, benefit, and happiness of gods
and humans. Who is that one person? The Tathāgata, the Worthy One, the Perfectly
Awakened Buddha”; and by the latter [he shows] a dwelling [place] that is appropriate
for him wherever he is born. This technique of exegesis should be understood here
starting in this way by this method: he shows that [the Buddha] was staying in a grove,
for the Bhagavan is always born in a grove because it is an occasion both worldly and
transcendent, first in Lumbinī, second, at the Bodhi tree.
With a great monastic community—​here “great” means greatness in qualities (guṇa)
and greatness in number, for the monastic community was great with qualities since
the least among them was a stream-​winner. And they were great in number in the sense
that there were five hundred of them. With a monastic community means a commu-
nity of monks, which means with a group of renouncers, a multitude considered to be
renouncers due to their views and moral precepts. With means together. Five hundred
means the quantity of five hundred of them. Quantity indicates how many. Because
of this, when it is said, “one knows the quantity of food,” it is that one knows quantity
in relation to food, which means one knows how much; and in the same way here too
the meaning should be seen as the quantity of five hundred monks is the amount of
five hundred. A hundred monks is a hundred times five monks. Because of this it is
said, “with a great monastic community,” so that the greatness in number of the great
community of monks is shown. But later the greatness of their qualities will become ev-
ident with the saying: “the community of monks, Sāriputta, is stainless, free of danger,
spotless, pure, and established in the essentials. Sāriputta, the least among these five
hundred monks is a stream-​winner.”
“The Brahmin Verañja heard” means he heard, took in, and understood in keeping
with the sound of the words that reached the doorways of his ears. “Kho” is a particle
that gives emphasis or is merely a space-​filler. Here it is just that he heard with a sense of
emphasis, and the meaning should be understood that there was nothing hindering his
hearing of it. But it could just be a space-​filler smoothing out the phrasing (byañjana).
He was Verañja in that he was born in Verañjā, raised in Verañjā, or lived in Verañjā.
But he was called “Udaya” because of the name given to him by his mother and father.
Brahmin means he recites Brahma, in the sense that he repeats mantras. For this is an
expression with a nirutti of the Brahmins as a class. But noble people are also called
“Brahmins” because of having warded off evils.1
Now, that the Brahmin Verañja heard this meaning is made clear when he [Upāli]
started with “the renouncer Gotama.” Here, renouncer should be understood due to his

1.  There are several niruttis to which Buddhaghosa is drawing our notice with the sounds
brā, ma, and bā in the word brāhmaṇa: they recite “Brahma” as a mantra, and their miseries
are warded off (bāhita). He manages to achieve meaning through nirutti analysis that treats
Brahmins as both a class and as a moral category.
 Appendix B 235

condition of having evils quelled. For this is said (in the Dhammapada): “a Brahmin
is one with evils warded off, a renouncer is called one with evils quelled.” And the
Bhagavan is one with evils quelled by the highest path of noble persons, so he is a “re-
nouncer,” namely, one who acquired his qualities in a real way. “Khalu” is a particle
indicating hearsay. “Sir” [“Bho”] is just a vocative for those in the born into the class of
the Brahmins. And this is said: “one is a Brahmin [literally, one who says ‘bho’] if he has
something.” “Gotama” describes the Bhagavan by reference to his family lineage, and
therefore “Sir, the renouncer Gotama” should be seen here as referring to the “Gotama
lineage.” “Son of the Sakyas” illustrates the high rank of the Bhagavan. “Gone forth from
the Sakya clan” is the illustration of his condition of having gone forth out of faith,
that is, he did not renounce his family because it had fallen to ruin as someone over-
come with a great loss, but rather it is said that he renounced out of faith. The meaning
described [will work] also elsewhere. “That [one]” is in the accusative case in the sense of
referring to what he is called, the meaning of it is Gotama. “Beautiful” means endowed
with the quality of being beautiful, said to be “the best.” “Sound of fame” is just fame,
or the voice of praise.
Now this is the exegesis of that beginning with “because he is thus [itipiso] the
Bhagavan . . .”: because he is the Bhagavan, because he is Worthy (Araha), because he is
Perfectly Awakened, et cetera” said to be so because of this and that reason. Now here
in the beginning of the commentary on the Vinaya I will make a commentary on these
words with an extended method for the sake of delighting the mind with a Dhamma
talk connected with the Buddha’s qualities [and] for the sake of Vinaya experts’ profi-
ciency in the methods of the Suttanta.
Therefore, when it is said that “because the Bhagavan is Worthy”—​here the
Bhagavan should be understood as “Araha” because of these reasons: because of his con-
dition of being aloof (ārakattā), because of enemies (arīnam), because of his destroying
the spokes (arānañca hatattā), because of worthiness (arahattā) of requisites, et cetera,
and because of the lack of secret (raha) evil doing. “Being aloof ” (āraka) means being
established far and removed from all defilements (kilesas), [that is] the condition of
having destroyed the defilements together with their traces with the Path, and so [he is]
Araha because of being aloof; and by this Path these enemies that are the defilements are
destroyed (hatā), and so he is Araha due to the condition of having destroyed (hatattā)
his enemies (arī). Now he is Araha also because of his condition of destroying enemies,
in that all enemies are destroyed, his having wielded with the hand of faith the hatchet
of knowledge that puts an end to kamma, standing firm on the ground of morality,
with feet of heroism on the grounds of the Bodhi Tree, having destroyed the wheel of
saṃsāra with its spokes (ara) that are the constructions such as merit, et cetera, with its
hub consisting of ignorance and craving for becoming, with its axle consisting of the
origins of the oozings (āsava), with its rim of old age and death, joined to the chariot of
the three forms of becoming, revolving from beginningless time.
Alternatively, the wheel of saṃsāra is called the revolving of beginningless saṃsāra,
and its hub is ignorance because of it being a root. Its rim is old age and death because
236 Appendix B

of their being the conclusion; and the remaining ten phenomena [of dependent origi-
nation] are the spokes because of their being conditioned by ignorance and concluding
with old age and death. Here ignorance is failing to understand suffering, et cetera,
and ignorance in the realm of desire is a condition for the intentional constructions
[saṅkhāra] in the realm of desire. Ignorance in the realm of form is the condition
for intentional constructions in the realm of form. Ignorance in the formless realm
is the condition for intentional constructions in the formless realm. Intentional
constructions in the realm of desire are the conditions for rebirth consciousness in
the realm of desire. This method [goes for] the rest. The rebirth consciousness in the
realm of desire is the condition for name-​and-​form in the realm of desire, and likewise
in the realm of form. And it is the condition for name in the formless realm. Name-​
and-​form in the realm of desire is the condition for the six sense bases in the realm
of desire. Name-​and-​form in the realm of form is the condition for the three bases
[that is, mind, sight, and hearing] in the realm of form. Name in the formless realm is
the condition for the single sense base [that is, the mind base] in the formless realm.
The six sense bases in the realm of desire are the conditions for sixfold contact in the
realm of desire. Three sense bases in the realm of form are [the condition for] three-
fold contact in the realm of form; one sense base in the formless realm is the condition
for one kind of contact in the formless realm. Six kinds of contact in the realm of
desire are the conditions for six feelings. Likewise, three in the realm of form are [the
conditions] for three; and one in the formless realm is the condition for one type of
feeling. Six feelings in the realm of desire are the conditions for six groups of craving in
the realm of desire; likewise, three in the realm of form are for three; and one feeling
in the formless realm is the condition for one group of craving. In each case each one
is the condition for clinging of this one or that one. And clinging is the condition for
becoming.
How? Here a certain person thinks “I will enjoy sense desires” and from the condi-
tion of clinging to sense desires, practices bad conduct with the body [or] practices bad
conduct with speech [or] mind. Because of accomplishing bad conduct, he is reborn
in a lower state. The kamma that is the cause of his rebirth there is the becoming that
is kamma, the aggregates produced from that kamma are the becoming that is rebirth,
birth is the production of the aggregates, old age is [their] aging, and death is [their]
breaking up.
Another, [thinking] “I will enjoy the magnificence of heaven,” practices good con-
duct and attains heaven because of accomplishing good conduct. The kamma that is
the cause of his rebirth there is the becoming that is kamma. The rest is by the [same]
method.
Another, [thinking] “I will enjoy the magnificence of the Brahma world,” cultivates
loving kindness as a condition for rebirth in the realm of desire, or he cultivates com-
passion, sympathetic joy, equanimity, and is reborn in the Brahma world having accom-
plished meditative cultivation. The kamma that is the cause of his rebirth there is the
becoming that is kamma. The rest is by the [same] method.
 Appendix B 237

Another, [thinking] “I will enjoy the magnificence of formless becoming,” cultivates

the attainments beginning with the base of boundless space, and by the accomplish-
ment of this meditation is reborn there. The kamma that is the cause of his rebirth
there is the becoming that is kamma. The aggregates produced from the kamma are the
becoming that is rebirth, birth is the production of the aggregates, old age is [their]
aging, and death is [their] breaking up. This is the method [to be used] in the exegesis
in the remaining roots of clinging.
All the words should be expanded by this method in this way: “the knowledge of
the relationships of dhammas is the understanding [that comes when one] takes up
conditions, such as ‘ignorance is the cause, intentional constructions are generated by
a cause, and both of these are generated by a cause’.” And also in the future and the
past: “the knowledge of the relationships of dhammas is the understanding [that comes
when one] takes up conditions, such as ‘ignorance is the cause, intentional constructions
are generated by a cause, and both of these are generated by a cause.’ ” Here ignorance and
intentional constructions are one brief summary (saṅkhepa); consciousness, name-​and-​
form, the six sense bases, contact, and feeling are one [brief summary]; craving, clinging,
and becoming are one; birth, old age, and death are one. Here the first summary occurs
in the past, the two middle are in the present, and birth, old age and death are in the
future. Having mentioned craving, clinging, and becoming by mentioning ignorance and
constructions, these five dhammas are the kammic round in the past; the five dhammas
beginning with consciousness are the round of their results. Having mentioned igno-
rance and constructions by mentioning craving, clinging, and becoming, these five
dhammas are the kammic round; because of expressing those beginning with con-
sciousness by the words birth, old age, and death, these five dhammas are the round of
their results in the future. These are twenty by way of mode (ākāra). And here there is
one link between constructions and consciousness, one between feeling and craving,
one between becoming and birth. Thus the Bhavagan knows, sees, understands, and
comprehends dependent origination in every mode, that is, the four summaries, the
three times, the twenty modes, and the three links in this way. That is knowledge in the
sense of what is known, understanding (paññā) in the sense of the act of understanding
(pajānanā). Because of that it is said: “the knowledge of the relationships of dhammas
is the understanding [that comes when one] takes up conditions.” The Bhagavan, with
the knowledge of the relationships of dhammas, knowing the dhammas as they are
produced, became disenchanted with them, detached, and freed, and so he destroyed,
slayed, and cut the spokes of the wheel of saṃsāra. And so in this way he is Araha because
of having cut the spokes.
Because of being foremost of those deserving alms, he is worthy (arahati) of
requisites such as robes and distinction in worship. And because of this, when the
Tathāgata has arisen, eminent deities and humans do not worship anyone else. Indeed,
Brahmā Sahampati worshiped the Tathāgata with a jewel garland the size of Mt.
Sineru, and other deities and humans such as the King of Kosala and King Bimbisāra
did likewise according to their means. And the great King Asoka established 84,000
238 Appendix B

monasteries throughout Jambudīpa, having dispersed ninety-​six koṭis of wealth dedi-

cated to the Bhagavan who had attained final nibbāna. Why speak of the distinction of
worshipping anyone else? In this way he is Araha because of being worthy of requisites,
et cetera. Moreover, in the world there are fools proud of their cleverness who do evil in
secret (raho) with fear of blame; he never does this, and so is worthy because of the lack
of secret evil-​doing. And so there is this:

Because of being aloof (āraka), and because of having destroyed (hata) the
enemies (ari) that are the defilements, this sage, who cut off (hata) the wheel of
saṃsāra, is worthy (araha) of requisites, et cetera; he does not do evil in secret
(raho), and because of this he called Araha.
 A p p en d i x  C

Four Oceans and Three Piṭakas

The first part of this selection expands Buddhaghosa’s characterization of both the
scriptural tradition and the omniscient knowledge of the Buddha as oceanic in a dis-
cussion unique to the Atthasālinī (As 10–​18). The second part, on defining the three
piṭakas, is also found (with slight differences) in the nidānas to the Vinaya and the
Suttanta piṭakas (As 18–​24; Sp i.18–​28; Sv i.16–​22). I discuss various passages from these
pages on many occasions throughout the book.
In this commentary on the Abhidhamma, Buddhaghosa develops an understanding
of it as an “ocean of methods,” an ocean that can be comprehended in its vastness only
by the oceanic ken of the Buddha. Buddhaghosa traces its discovery to the fourth week
of the Buddha’s awakening under the Mahābodhi tree. As he recreates this scene, we
are treated to a spectacular light display as the Buddha contemplates the seventh book
of the Abhidhamma, the Paṭṭhāna. It is only when contemplating this book that the
Buddha’s omniscience can begin to “find room,” and it is at this moment that six fabu-
lous rays of colored light spring out from his body to pervade the furthest reaches of the
three thousand thousandfold cosmos.
But as fantastic as this is, Buddhaghosa never drifts too far away from immediate
practicalities:  how can such immeasurable knowledge be taught in any amount of
time? And to whom? And how can it be grasped by those of lesser understanding? It is
here that we learn how the Abhidhamma was taught for three months to the Buddha’s
mother in Tāvatiṃsa heaven, and then to the elder Sāriputta on earth. By fashioning
a clone of himself, the Buddha can simultaneously convey without interruption these
sublime teachings to the heavenly beings and return to the earth to brush his teeth, get
a meal, take a nap, and teach Sāriputta. But how to teach even this most excellent dis-
ciple, famed for his capacities for analysis? The Abhidhamma is given to him via a “gift
of methods,” a technique that allows the elder to generate it in its fullness.
240 Appendix C

The second half of the selection is an extended treatment of the piṭakas, where we
see Buddhaghosa reflect on genre and the types of teaching and learning that constitute
both the study and the compilations of the Vinaya, Suttanta, and the Abhidhamma.
Readers developing a fondness for nirutti analysis will get a further taste of this partic-
ular method of generating meaning from the phonemes in these words in the verses and
commentary developed here.
As in the other appendices, I have italicized words that have particular salience in
my discussion in the book, as well as words that are receiving commentarial exposition.

At t h a s ā l i n ī   10 –​2 4
Now, in order to comprehend the depth of the Abhidhamma, four oceans should be
understood: the ocean of saṃsāra, the ocean of water, the ocean of methods (naya), and
the ocean of knowledge. Now, as for the ocean of saṃsāra:
The succession of aggregates, elements, and bases is the unbroken process referred
to as “saṃsāra.”
The round of saṃsāra is spoken of in this way: a first point of birth of these beings
is not discerned since beings are born going back a hundred years, a thousand years, a
hundred thousand years, a hundred aeons, a thousand aeons, or a hundred thousand
aeons; nor did they not exist previously, but have long been born in a time of such and
such a king, born in a time of such and such a buddha, and there is no end point in
which they previously did not exist. The beginning and end of the ocean of saṃsāra are
inconceivable according to what has been shown by this: “monks, a first point of igno-
rance is not discerned whereby [it might be said that] ignorance did not exist before
and came into being later.”
The ocean of water should be understood as the great ocean. It is 84,000 yojanas
deep. Here, there is no measure of the water, whether a hundred measuring cups or a
thousand measuring cups, or a hundred thousand measuring cups. So it is reckoned as
incalculable, immeasurable, just a great mass of water. This is called the ocean of water.
Which is the ocean of method? The Buddha’s words that are the three piṭakas. For
endless joy and happiness arise for those from a good family, possessing faith, abounding
in serene delight, whose knowledge is excellent, who reflect on two of the scriptures.
Which two? Vinaya and Abhidhamma. When monks who are Vinaya experts are
contemplating the Vinaya text, that is, the declaring of the rules according to faults,
[they realize] that it is the ken of buddhas alone, not the scope of others [to know]
“in this fault, in this transgression there is the declaring of a rule”; and endless joy and
happiness arise for those reflecting on the repeated [or abridged] formulas regarding
superhuman attainments, [colors like] blue, and acting as a go-​between.1 And endless

1. I am grateful for Margaret Cone’s help with this passage, though we are still unclear on what
is meant that Vinaya experts reflect on abridged formulas on blue.
 Appendix C 241

joy and happiness arise for those monks who are Abhidhammikas [that is, specialists]
reflecting on the Abhidhamma scripture thus: “the Teacher taught us [by] analyzing
the name-​and-​form dhammas, making this or that grouping, this or that division, as
though arranging the stars in the sky, which is an abstruse and subtle teaching that
has classified into what is form and what is formless the manifold2 aggregates, bases,
elements, faculties, powers, limbs of awakening, kamma and its results.”
A story of this should be understood with this incident: the Elder Mahāgatigamiyatissa
went across the sea [thinking] “I will worship the Mahābodhi Tree,” and sitting on the
upper deck of the ship, he beheld the great ocean. On that occasion (tasmiṃ samaye)
he could not see the far shore nor the near shore, and saw only the great ocean covered
with foam from the fast-​breaking waves like a silver cloth spread out and strewn with
jasmine blossoms. He began to wonder which is more powerful: the speeding waves of
the great ocean or the basis of the method in the entire Paṭṭhāna with its twenty-​four
divisions. He knew that the great ocean is bordered by the great earth below, by the sky
above, by the ring of mountains on one side, and by shores on the other; yet a border to
the entire Paṭṭhāna cannot be known. And so the abstruse and subtle Dhamma appears
as more powerful by one so considering. Even while seated there he was overcome with
joy, increased his insight, destroyed the defilements, and attained the highest fruition
which is arhatship. And he made this cry of ecstasy:

Now having understood for oneself the very difficult and deep wisdom
With [all] causes and origins in succession
taught in its entirety by the Great Sage,
one sees it as though it had assumed a form.

This is called the ocean of methods.

Which is the ocean of knowledge? The ocean of knowledge is the knowledge of
omniscience. The ocean of knowledge is the knowledge of omniscience that makes it pos-
sible only by means of the knowledge of omniscience to know that “this is the ocean of
saṃsāra,” “this is the ocean of water,” “this is the ocean of methods,” which cannot be
known by others. Among the four oceans what is intended here [in this commentary on
the Abhidhamma] is the ocean of methods. For only omniscient buddhas comprehend it.
The Bhagavan, seated at the root of the Bodhi Tree, sat on a single throne for seven
days contemplating the Dhamma he had penetrated thus:  “this Dhamma has been
penetrated by me seeking and striving for more than four incalculables and a hundred
thousand aeons, and by me, seated on this throne, a thousand five hundred defilements
have been eradicated. This Dhamma is penetrated!” Then, rising from this throne he

2.  For interpreting antara at the end of khandhantara, and so forth, I  am aided by the
Mūlaṭīkā: “khandhantaranti khandhanānattaṃ khandhameva vā”: “khandantara” is the man-
ifold [diversity of ] aggregates or just the aggregates.”
242 Appendix C

stood gazing at the throne with unblinking eyes for seven days [thinking] “on this
very throne I penetrated the knowledge of omniscience.” Then a thought arose for the
deities: “surely there is something still to be done by Siddhattha today for he has not
eliminated attachment to that throne.” So the Teacher, knowing that thought of the
deities, rose up into the sky to remove their impression and displayed the Twin Miracle.
The Twin Miracle performed on the Mahābodhi throne, the Miracle performed at the
assembly of his relatives, and the Miracle performed at the assembly at Pāṭiyaputta were
all similar to the Twin Miracle performed at the roots of the Kaṇḍamba tree. Having
performed the Twin Miracle in this way, he descended from the sky and walked for
seven days between the throne and the place where he had stood. And at no time on
any day during the next twenty-​one days did rays of light extend from the body of the
Teacher.
But in the fourth week he sat in the Jewel House facing the northwestern direc-
tion. Note that the jewel house is not a house made of seven jewels but rather, what
should be understood is a house of jewels for contemplating the seven treatises [of the
Abhidhamma]. And here rays of light did not extend from his body while contemplating
the Dhammasaṅgaṇī. Nor did rays of light extend from his body while contemplating
the treatises of the Vibhaṅga, Dhātukatha, Puggalapaññatti, Kathāvatthu, or Yamaka
either. But when he dived into the Great Treatise, beginning to grasp “the condition
of cause, the condition of basis, the condition of nondisappearance, et cetera” then his
omniscient knowledge found space in the Great Treatise as he grasped completely the
entire Paṭṭhāna of twenty-​four [relations]. For just as the leviathan Timirapiṅgala finds
space only in the great ocean 84,000 yojanas deep, so too the knowledge of omniscience
finds room completely only in the Great Treatise.
With the space achieved by the knowledge of omniscience of the Teacher who had
grasped the abstruse and subtle Dhamma easily, six rays of light—​blue-​black, golden,
red, white, copper, and shimmery—​cast forth from his body. The blue-​black rays cast
forth from his hair and from the blue-​black places in his eyes; and by them the surface
of the sky appeared to be dusted with collyrium powder, covered with blue lotuses and
blue flax, like a jeweled fan waving to and fro, and like a dark blue cloth stretched out.
The golden rays cast forth from the golden places of his eyes and skin, and by them
the directions shone brilliantly as though streaming from a shower of golden liquid,
spread with a golden cloth, and strewn with saffron powder and the blossoms of the
Kanak Champa tree.
The red rays cast forth from the red places in his eyes and from his flesh and blood,
and by them the directions shone brilliantly as though dyed with powder from red lead,
sprinkled with the liquid from molten lac, covered with a red blanket, and strewn with
Scarlet Pentapetes, Pāribhaddaka, and Bandhujīvaka flowers.
The white rays cast forth from the white places in his eyes and from his bones and
teeth, and by them the directions shone brilliantly as though issuing from a stream of
milk flowing from a silver bowl, like a canopy of a spread-​out silver cloth, like a silver
 Appendix C 243

fan waving to and fro, and like strewn blossoms of Kunda jasmine, white water lilies, the
Nirgunda shrub, Sumana jasmine, Mallika jasmine, et cetera.
The coppery and shimmery rays cast forth from this and that place in his body.
These six rays, having cast out, caught the massive and great earth.
The great earth, more than two hundred and forty thousand yojanas thick, became
like a ball of polished gold. And having pierced the great earth, they [the rays] caught the
water below. The water that holds back the earth, more than four hundred and eighty
thousand yojanas deep, became like refined gold sprinkled from golden cups. Having
penetrated the water, they caught the atmosphere. The atmosphere, more than nine
hundred and sixty thousand yojanas in breadth, became like a mass of compounded
gold. Having penetrated the atmosphere, they sprang out further into open space.
And having ascended to the upper regions, the rays caught the [worlds] of the Four
Great Kings. Having penetrated these, they penetrated the Tāvatiṃsa, Yama, Tusita,
Nammānaratī and Paranimmitavasavattī worlds, then the nine Brahmā worlds, the
Vehapphala, and then having penetrated the five Pure Abodes, they caught the four
formless realms. And having penetrated the four formless realms, they sprang out into
open space.
They sprang out horizontally into endless world-​systems. In so many of these places
there is no moonlight in the moon, no sunlight in the sun, no starlight in the stars,
and nowhere is there light in the pleasure gardens, palaces, and wishing trees of the
deities, in their bodies or their ornaments. Even the Great Brahmā, able to suffuse the
three thousand thousandfold world systems with light became like a firefly at sunrise.
There was known only the mere borders of the moon, sun, stars, divine pleasure parks,
palaces, and wishing trees. The Buddha’s rays overwhelmed so much space. But this was
not the Buddha’s magical power of resolve, nor the magical power created by medita-
tion. Rather, the blood of the Lord of the World became bright as he contemplated
the abstruse and subtle Dhamma, and his physical form became bright, and the com-
plexion of his skin became bright. The actual material form that arises from such aware-
ness completely and steadily established itself in a place measuring eighty cubits. He
contemplated in this manner for seven days.
How large is the Dhamma contemplated for seven days and nights? It is endless and
immeasurable. This is the extent of the Teaching [occurring] in the mind. But it is not
to be said that the Teacher, [although] having put into speech the Dhamma thought out
by the mind for seven days, cannot teach and arrive at the end of it even in a hundred
years, a thousand years, or a hundred thousand years. For subsequently the Tathāgata
taught it, seated in the middle of the deities of the ten thousand world systems on the
stone throne of Sakka at the root of the heavenly Coral Tree in the Tāvatiṃsa realm,
making his mother grasp the truth of the body, teaching the Dhamma thus: “these are
the good dhammas, these are the bad dhammas, these are the indeterminate dhammas,”
passing on from one dhamma to the next dhamma through a hundred divisions, a
thousand divisions, a hundred thousand divisions. This teaching set forth without
244 Appendix C

interruption for three months was endless and immeasurable as it flowed rapidly like
the heavenly Ganges and like water ushering from an upside-​down pot.
For a teaching of buddhas making a blessing at the time of blessing a meal, if ex-
panded just a little, is the length of the Dīgha and Majjhima Nikāyas. And the teaching
of those teaching the Dhamma to an assembly gathered at the end of a meal is the length
of the two large nikāyas, the Saṃyutta and Aṅguttara together. How is this possible?
The life spans of buddhas are fleeting, and [so] their teeth touch well, their mouths
flow smoothly, their tongues are supple, their sounds, sweet, and their words, quick.
Therefore, so much Dhamma can be taught in a moment. However, the Dhamma
taught in the three months is still endless and immeasurable.
For even the elder Ānanda, [honored as] “He Who Heard Much,” a master of the
three piṭakas, learned, recited, or taught 1,500 verses and 60,000 words standing there
as easily as if he were plucking flowers from vines. And that much was just one of this
elder’s courses of exposition. But while giving the exposition continuously, the elder
would not be able to give another one, or even learn it. Only the Perfectly Awakened
One could cause [him] to learn it. A disciple of such extraordinary memory, extraor-
dinary understanding, and extraordinary resolution is not able to reach the end of
the teaching taught in this manner by the Teacher in the three-​month period even if
studying for a hundred or a thousand years.
But while teaching without interruption for three months in this way, how did the
Tathāgata feed his body which comes from [past] clinging and is dependent on mate-
rial food? Only by eating. For buddhas every period of time is well appointed, well di-
vided, and easily perceived. Therefore, the Bhagavan, even while teaching the Dhamma,
observed time in the human world. He noted that it was time for the almsround and
so created a fabricated buddha, determining that “this one should teach the Dhamma
up to a certain extent while holding the robe, the bowl, the cup, and assuming a guise
of likeness,” while he, taking his own bowl and robe, went to Anotatta Lake. There the
deities gave him a tooth stick made of nāga creepers. Having chewed this and attended
to his body in Anotatta Lake, he stood at the platform, got dressed in his red-​dyed cloth
and donned his robe, took his crystal bowl offered by the Four Great Kings, and went
to Uttarakuru. Then he gathered almsfood, sat at the banks of Anotatta Lake, enjoyed
it, and headed to a sandalwood grove for the midday rest.
The elder Sāriputta, General of the Dhamma, went there, performed the duties
for the Perfectly Awakened One, and sat to one side. Then the Teacher gave him the
method (naya). He showed him, “Sāriputta, so much Dhamma is taught by me.” When
the Perfectly Awakened One was giving the method in this way to the chief disciple ac-
complished in analysis (paṭisambhidā), there was a gift of the method (nayadāna) much
like one pointing at what is seen by stretching out one’s hand while standing at the
seashore. And to the elder also the Dhamma, taught by the Bhagavan with a hundred
methods, a thousand methods, a hundred thousand methods, became clear.
Having sat in that divine abode, at what time did the Teacher go to teach the
Dhamma? There is a time for teaching the Dhamma to the inhabitants of Sāvatthi,
 Appendix C 245

sons of good families, who have arrived, and it was at that time that he went. And
[while the fabricated Buddha in Tāvatiṃsa] was teaching the Dhamma, who
[among the deities] recognized the one who had gone or the one returning, and
who did not? The most eminent deities knew, while the deities of lesser eminence
did not know. Why did they not know? Because there was no difference between
the Perfectly Awakened One and the fabricated Buddha with regard to their rays
of light, et cetera. For there is no difference in the rays, the sound, or the words of
both of them.
Now, the elder Sāriputta took the Dhamma taught in this way by the Teacher and
taught five hundred of his fellow monks in his monastery. There is a connection to the
past for these [monks]: they had been born at the time of the Ten-​Powered Kassapa
Buddha as little bats and were hanging in a mountain cave when there was the sound of
two Abhidhammika monks reciting the Abhidhamma. They took it as a portent even
though they did not know the dark side from the light side, and just because they took
it as a portent, when they died they were reborn in the realm of the deities. Having
lived in the realm of the deities for one interval between buddhas, they were born in
the human realm at this time and, being delighted by the Twin Miracle, ordained as
monks in the presence of the elder. The elder took the Dhamma taught in this way
by the Teacher and taught it to them. And at the conclusion of the teaching of the
Abhidhamma of the Perfectly Awakened One, these monks grasped the seven treatises
immediately.
The way of reciting in the Abhidhamma originated with the elder Sāriputta. And
the process of enumeration in the Great Treatise was established by the elder. For the
elder, without effacing the distinctiveness of the Dhamma in this manner, established a
process of enumeration in order to make it easier to grasp, retain, master, and say. This
being so, was he then the first Abhidhammika? No, he was not. The first Abhidhammika
is the Perfectly Awakened One alone. Sitting on the throne at the Mahābodhi Tree, he
penetrated it and became the Buddha. And seated at the single throne for seven days he
uttered this cry of ecstasy:

Indeed, when dhammas become visible

to the ardent and meditating Brahmin,
then all his doubts disappear,
since he understands that a dhamma has a cause.
Indeed, when dhammas become visible
to the ardent and meditating Brahmin,
then all his doubts disappear,
since he understands that a condition ends.
Indeed, when dhammas become visible
to the ardent and meditating Brahmin,
he stands with the armies of Māra scattered,
like the sun alone shining in the sky.
246 Appendix C

This was the first utterance of the Buddha. But the reciters of the Dhammapada
differ, and say that this was the first utterance of the Buddha:

I traversed saṃsāra through many births not finding,

though searching, for the maker of the house. Birth is suffering, again and again.
But now you are seen, maker of the house! You will not build the house again!
All the rafters are broken, the ridge-​piece, destroyed.
Awareness is deconstructed, the elimination of craving attained.
(Dhammapada 153–​54)

At the occasion of the final nibbāna while lying between the pair of Sāl trees the
final utterance of the Buddha was spoken thus: “Come now, monks, I tell you, condi-
tioned things are subject to decay. Strive with vigilance.” Now, the Good Dhamma is
the words of the Buddha that illuminate the Deathless that were taught by the Buddha
between these two events for forty-​five years as though he were connecting a garland of
flowers and stringing together a line of jewels.
All of this gathered together comprises the three piṭakas by way of piṭaka, the five
nikāyas by way of nikāya, ninefold by way of parts, and 84,000 divisions by way of
divisions of the Dhamma. How exactly? Everything is divided into just three by way of
piṭaka: the Vinaya Piṭaka, the Suttanta Piṭaka and the Abhidhamma Piṭaka. Here, both
Pātimokkhas, two Vibhaṅghas, twenty-​two Khandhakas, and sixteen Parivāras are the
Vinaya Piṭaka. The Dīgha Nikāya is the collection of thirty-​four suttas starting with
the Brahmajāla. The Majjhima Nikāya is the collection of 152 suttas beginning with the
Mūlapariyāya. The Saṃyutta Nikāya is the collection of 7,762 suttas beginning with the
Oghataraṇa. The Aṅguttara Nikāya is the collection of 9,557 suttas beginning with the
Cittapariyādāna. And the Suttanta Piṭaka also includes the Khuddaka Nikāya divided
into fifteen by way of these texts: the Khuddakapāṭha, the Dhammapada, the Udāna,
the Itivuttaka, the Suttanipāta, the Vimānavatthu, the Petavatthu, the Theragāthā, the
Therīgāthā, the Jātaka, the Niddesa, the Paṭisambhidā, the Apadāna, the Buddhavaṃsa,
and the Cariyāpiṭaka. And the seven treatises starting with the Dhammasaṅgaṇī are the
Abhidhamma Piṭaka.

Now,
The Vinaya is called “vinaya” by those wise (vidū) in the meaning of vinaya be-
cause it contains various (vividha) and distinctive (visesa) methods (naya) and
because it disciplines (vinaya) body and speech.

For here, its methods are various in reference to the fivefold recitation (uddesa) of the
Pātimokkha, which can be divided into the matrix (mātikā) of the seven sections on
offences beginning with the defeats, and the analysis (vibhaṅga), et cetera, and they are
distinctive because the methods of supplementary regulations aim to make flexible what
 Appendix C 247

is rigid. It disciplines body and speech by restraining physical and verbal transgression.
Therefore, because of having various methods and because of its distinctive methods the
Vinaya is called “vinaya” because of disciplining body and speech. Because of this the
following is said for the sake of proficiency in the meaning of the utterance:

The Vinaya is called “vinaya” by those wise in the meaning of vinaya because it
contains various and distinctive methods and because it disciplines body and
speech.
And further:
[The Suttanta] is called “sutta” because it is indicating (sūcana) meanings
(attha), because it is well-​spoken (suvutta), because it is flowing (savana), be-
cause it is yielding (sūdana), because of its stringing together (suttāṇa), and be-
cause of a sutta being shared in common (sabhāga).3

It indicates this in reference to meaning divided into meaning for oneself and meaning
for others, et cetera. It is well-​spoken in that the meanings are spoken in accordance with
the inclinations of those being taught right here. It flows is said [to indicate] that it bears
fruit in meaning, like crops. It yields is said [to indicate] that it discharges as a cow does
milk. And it is good (suṭṭhu) is said [to indicate] that it protects and guards them. And
there is the sharing in common of a sutta. And just as a thread is a measure for carpenters,
in this way [a thread/​sutta] is for the wise. And just as flowers strung by a thread do not
scatter and disperse, so too by it meanings are gathered. Because of this the following is
said for the sake of proficiency in the meaning of the utterance:

[The Suttanta] is called “sutta” because it indicates meanings, because it is well

spoken, because it is flowing, because it is yielding, because of its stringing to-
gether, and because of a sutta being shared in common.

The meaning of the word Abhidhamma has already been mentioned.4 Here is an-
other method:

[The Abhidhamma] is called abhidhamma because by it phenomena (dhammas)

[can be] grown, have characteristics, are honored, are defined, and exceed what
has been said.

3. This ancient verse has words that are not in The Pali Text Society’s Pali-​English Dictionary and
are to be understood from the Sanskrit: sūdana (Skt syandana, “dropping, oozing, trickling”
and suttāṇā (from Skt. sūtraṇa, “the act of stringing together”), as well as Buddhaghosa’s exe-
gesis of them (Monier-​Williams, Sanskrit-​English Dictionary, 1273, 1242).
4. In an earlier part of the Atthasālinī, not translated here.
248 Appendix C

Here the word “abhi” is shown in reference to [all of these]—​how [the dhammas] can
be grown, characterized, honored, defined, and can exceed. Therefore, they can be grown
is understood in such [canonical passages as]: “my severe and painful feelings increase
(abhikkamanti) without relenting.” In terms of characteristics:  “such nights as these
characterized (abhiññāta) and distinguished (abhilakkhita),” et cetera. In terms of
being honored [we have similar epithets such as]: “mighty king of kings, ruler of men,” et
cetera. They are defined in this way where Dhamma and Vinaya are not to be confused
with one another: “one able to instruct in the higher Dhamma (abhidhamma) and in
the higher Vinaya (abhivinaya).” And they exceed is “because of surpassing beauty,” et
cetera.
Now here, phenomena (dhamma) are said to be grown also by [another]
method:  “one has cultivated the Path by giving rise to the [realm of ] form, and so,
having pervaded one direction with thoughts accompanied by loving-​kindness, one
dwells.” They have characteristics (salakkhaṇa) because of their being characterized
by their object, et cetera, by a method such as:  “it has as its object a visual form or
it has as its object a sound.” They are honored means they are worthy of reverence by
this method:  “trainer phenomena, adept phenomena, transcendent phenomena,” et
cetera. They are defined because of being defined by their particularity (sabhāva) by this
method: “there is contact, there is feeling,” et cetera. And the phenomena are said to
exceed by this method: “dhammas that are extensive, dhammas that are immeasurable,
dhammas that are incomparable,” et cetera. Because of this the following is said for the
sake of proficiency in the meaning of the utterance:

[The Abhidhamma] is called abhidhamma because by it phenomena (dhammas)

[can be] grown, have characteristics, are honored, are defined, and exceed what
has been said.

Now here there is a general sense—​

People learned about the meaning of piṭaka refer to piṭaka in the sense of an
area of study (pariyatti) and basket. Because of this, having collected them to-
gether, they are known as three, starting with the Vinaya.

Now, that a piṭaka is an area of study is said in such quotations as “do not go by mastery
of an area of study,” and that it is also a basket is evident in such statements as “if a man
were to pick up a hoe and basket and go.” Therefore, those learned in the meaning of
“piṭaka” refer to piṭaka in the sense of an area of study and basket.
Now, because of this, having collected them together, they are known as three, starting
with the Vinaya—​a compound is made with the word piṭaka in its two meanings in this
way, and so Vinaya is a piṭaka because of being an area of study, and, from that meaning,
it is also the Vinaya Piṭaka by being a basket. And by the same method, the Suttanta is
 Appendix C 249

also a piṭaka [by being an area of study] and it is the Suttanta Piṭaka. Likewise for the
Abhidhamma as a piṭaka and as the Abhidhamma Piṭaka, and so they are known as
three, starting with the Vinaya.
Having understood this, there more [to be said] for the sake of proficiency in the
various kinds of these piṭakas:

One may illuminate the types of teaching, instruction, and discourse as appro-
priate according to training (sikkhā), avoidance (pahāna), and depth (gambhīra).
This should be made entirely clear: the monk who attains success and failure in
the different kinds of study.

And here is the illuminating and the making clear:  these three piṭakas are said
to be, in order, teachings as commands (āṇā), as colloquial speech (vohāra), and
[teachings] in the furthest sense (paramattha); [they are] instruction according to
transgression, according to what is suitable, and according to the dhammas; and
they are discourses on the kinds of restraint, unraveling views, and distinguishing
name-​and-​form.
Now, the Vinaya Piṭaka is a teaching by commands due to a preponderance
of commands, because of being taught by the Bhagavan who is worthy of giving
commands. The Suttanta Piṭaka is a colloquial teaching due to a preponderance of the
colloquial, because of being taught by the Bhagavan who is skillful in the colloquial.
The Abhidhamma Piṭaka is said to be teachings in the furthest sense due to its preponder-
ance of further meaning (paramattha), because of being taught by the Bhagavan who is
skillful in further meaning.
And now, instruction according to transgression is when he instructs concerning
beings with repeated transgressions according to transgression. Secondly, instruc-
tion according to what is suitable is when he instructs beings according to their various
inclinations, biases, practices, and dispositions. And, thirdly, instruction according to the
dhammas is when he teaches to beings who attribute “I” and “mine” to a mere heap of
phenomena (dhammas).
And now, discourse on the kinds of restraint refers to the kinds of restraint described
here as opposing infractions; and the kinds of restraint are the lesser and greater kinds
of restraint [in a compound interpreted as] “kinds of restraint” [saṃvarāsaṃvaro]
which is like [the compounds] “kinds of kamma” (kammākamma) and “kinds of fruit”
(phalāphala). Secondly, discourse unraveling views is described here as the unraveling of
views that opposes the sixty-​two views [of the Brahmajāla Sutta]. And, thirdly, it is said
that discourse distinguishing name-​and-​form is described here as the distinguishing of
name-​and-​form that opposes passion, et cetera.
And there should be understood that there are three trainings, three avoidances, and
four kinds of depth in each of the three piṭakas. Now, in the Vinaya Piṭaka there is said
to be specifically training in higher moral precepts (adhisīla), in the Suttanta Piṭaka
250 Appendix C

training in higher awareness (adhicitta), and in the Abhidhamma Piṭaka, training in

higher understanding (adhipaññā).
And in the Vinaya Piṭaka there is the avoidance of transgression because of moral
precepts opposing transgression of the defilements; in the Suttanta Piṭaka there
is the avoidance of being overpowered because of concentration opposing being
overpowered; and in the Abhidhamma Piṭaka there is the avoidance of latent bias be-
cause of understanding opposing latent bias. Moreover, in the first case there is the cat-
egorical avoidance of the defilements, while in the other two there is avoidance in the
sense of withdrawing support and cutting off. And in the first there is avoidance of the
defilements of bad conduct, while in the other two there is avoidance of the defilements
of craving and view.
And in each one of them, the four kinds of depth should be understood as depth
in the Dhamma [or dhammas], the meaning [or purpose, attha], the teaching, and the
comprehension. Here Dhamma means the scripture. Meaning is the meaning just of
that. Teaching is teaching with the scripture that one has already understood with one’s
own mind. Comprehension is correct understanding of the scripture and the meaning
of the scripture. And the Dhamma, the meaning, the teaching, and the comprehension,
applies to all three [piṭakas]. They are considered depths because they are hard to get a
footing in and hard to wade into by those of slow intellect, just as the great ocean is for
hares. And these four depths are to be understood in this way in each case.
An alternative method:  dhamma refers to cause (hetu). For it is said:  “analysis
(paṭisambhidā) of a dhamma is knowledge about a cause.” Purpose (attha) is the result of
a cause. For it is said: analysis of purpose is knowledge about a result of a cause. Teaching
is making known—​the intention is speaking about dhammas according to dhammas;
or else it is speaking [of them], whether in order or backwards, in brief or in detail,
et cetera. Comprehension is clear understanding, which is worldly and transcendent.
There is full knowledge with regard to range and without confusion about dhammas
according to purpose, about purpose according to dhammas, and about what has been
made known according to the ways of making it known. Of each of them and in each
case there should be the penetration of [each of ] the dhammas as they are described,
[by] its distinct particularity (sabhāva) and naming its characteristic (lakkhaṇa).
Now, in whichever way [of these two options] the meaning is to be understood for
the occurrence of dhamma and for the occurrence of attha, there will be an encounter
with knowledge for the listener, for each of them is a teaching that casts light on the
meaning of [the piṭakas]. And here too whichever comprehension described by full
knowledge of what is distinctive should be penetrated for each of the [two meanings
of ] Dhamma/​dhamma in its distinct particularity and naming its characteristic. For in
every way they are depths because they are hard to get a footing in and hard to wade into
by those of weak understanding who have not amassed the good conditions [for under-
standing], just as the great ocean is for hares. And their state of being deep in these four
ways should be understood in each case.
 Appendix C 251

And so this is the meaning of the verse that was spoken –​

One may illuminate the types of teaching, instruction, and discourse as ap-
propriate according to training (sikkhā), avoidance (pahāna), and depth
(gambhīra).
This should be made entirely clear: the monk who attains success and failure
in the different kinds of study.

[Piṭaka in the sense of ] area of study (pariyatti) is to be seen here in reference to the
three piṭakas. For there are three kinds of study: that of a person catching a snake, that
of a person seeking escape, and the study of the treasurer.
Here those catching a snake have learned it badly, doing it for the sake of reproaching
[others], et cetera. This has been said in this connection: “Monks, just as a man catching
a snake, searching for a snake, and wanting a snake roams around until he sees a huge
snake and then grabs its hood or its tail. The snake would whip around and bite his
hand or arm or any other limb or extremity; this would be the source of death or deadly
pain for him. Why is that? Because of grasping the snake badly, monks. And in this
way too, monks, certain foolish people learn the Dhamma, whether the suttas [or any
of the other treatises up to] the questionings, and having learned the Dhamma without
investigating its meaning through understanding the teachings, they cannot gain insight
into a meaning that is not investigated for they only learn the teaching for the purpose
of reproaching others and for the purpose of being free of criticism. Those learning the
Dhamma for such purposes do not acquire its meaning. These teachings badly learned
lead to their harm and suffering for a long time. What is the reason? Because, monks, of
the poor grasp of the teachings” (M i.133–​34).
But those who learn with a desire to fulfill the sections on the moral precepts and
not for reason of reproaching others are those seeking escape [from saṃsāra]. It is said in
connection to this that “monks, these teachings lead to their welfare and happiness for
a long time. What is the reason for this? Because of the good grasp of the teachings.”
The study of the treasurer is that of the arhat for whom the aggregates are well under-
stood, the defilements are eliminated, the Path is cultivated, the attainment is steadfast,
and cessation is realized, who learns only for the sake of preserving the tradition and
protecting the lineage.
Now the monk deeply learned in Vinaya, by achieving the moral precepts, obtains
the three knowledges that have been described here in the word-​analysis. One deeply
learned in the Sutta, by achieving concentration, obtains six higher knowledges that
have been described here in the word-​analysis. One deeply learned in Abhidhamma, by
achieving understanding, obtains the four kinds of analysis (paṭisambhidā) that have
been described here in the word-​analysis. And so one deeply learned in these succes-
sively in this way obtain success in the three knowledges, the six higher knowledges, and
the four analyses.
252 Appendix C

But one poorly accomplished in Vinaya perceives as blameless contact with things
causing attachment that are forbidden because of [thinking that they are] the same as
contact with blankets and cloaks that are pleasing to the touch and not proscribed.
For this is said: “I know the Dhamma taught by the Bhagavan, wherein some things
mentioned by the Bhagavan are injurious but are not harmful when practiced [by me].”
From this one obtains a condition of bad moral precepts. One poorly accomplished in
Sutta, not knowing the intended meaning in such statements as “there are four kinds
of people in this world,” et cetera, has a poor grasp of things. In connection to this, it
is said: “because of his poor grasp he accuses us, uproots himself, and generates much
demerit.” He comes to have false views. One poorly accomplished in Abhidhamma
overruns thinking on dhammas and thinks about imponderables, and from this
becomes deranged. For this is said: “monks, there are four imponderables, not to be
thought about since one thinking of them gets deranged and vexed.” So it is that one
poorly accomplished in these [piṭakas] achieves failure in each of these in succession: a
condition of bad precepts, false views, and deranged mind. And so:

This should be made entirely clear: the monk who attains success and failure in
the different kinds of study.

This is the meaning of the verse that was spoken.

Knowing the piṭakas by these various ways, the three piṭakas are the whole
[buddhavacana] arranged by them.5

5. This final line is rendered differently (and more clearly) in the commentaries on the Vinaya
and the Dīgha Nikāya: “Knowing the piṭakas by these various ways, the threefold buddhavacana
should be known by means of them” (Sp i.26; Sv i.22).
 Bibliography

Adikaram, E. W. Early History of Buddhism in Ceylon. Colombo: Gunasena, 1953.

Allon, Mark. Style and Function: A Study of the Dominant Stylistic Features of the Prose
Portions of Pāli Canonical Sutta Texts and Their Mnemonic Function. Tokyo: The
International Institute for Buddhist Studies of the International College for
Advanced Buddhist Studies, 1997.
Almond, Ian. “The Meaning of Infinity in Sufi and Deconstructive Hermeneutics: When
Is an Empty Text an Infinite One?” Journal of the American Academy of Religion 72,
no. 1 (2004): 97–​117.
An, Yang-​Gyu, trans. The Buddha’s Last Days:  Buddhaghosa’s Commentary on the
Mahāparinibbāna Sutta. Oxford: Pali Text Society, 2005.
Anālayo. “The Brahmajāla and the Early Buddhist Oral Tradition.” Annual Report of
the International Research Institute for Advanced Buddhology at Soka University 17
(2014): 41–​59.
Anālayo. “The Brahmajāla and the Early Buddhist Oral Tradition (2).” Annual Report
of the International Research Institute for Advanced Buddhology at Soka University
18 (2015): 79–​94.
Anālayo. “The Buddha and Omniscience.” Indian International Journal of Buddhist
Studies 7 (2006): 1–​20.
Anālayo. The Dawn of Abhidharma. Hamburg: Hamburg University Press, 2014.
Anālayo. Satipaṭṭḥāna: The Direct Path to Realization. Birmington, U.K.: Windhorse
Publications, 2006.
Anālayo. “Views and the Tathāgata—​A Comparative Study and Translation of the
Brahmajāla in the Chinese Dīrgha-​āgama.” In Buddhist and Pali Studies in Honour
of the Venerable Professor Kakkapalliye Anuruddha, edited by K. L. Dhammajoti and
Y. Karunadasa, 183–​234. Hong Kong:  Centre of Buddhist Studies, University of
Hong Kong, 2009.
Aung, Shwe Zan, and C. Rhys Davids, trans. Points of Controversy. London: Pali Text
Society, 1979.
254 Bibliography

Balcerowicz, Piotr. “Some Remarks on the Opening Sections in Jaina Epistemological

Treatises.” In Śāstrārambha: Inquiries into the Preamble in Sanskrit, edited by Walter
Slaje, 25–​81. Wiesbaden: Harassowitz Verlag, 2008.
Bapat, P. V. Vimuttimagga and Visuddhimagga:  A Comparative Study. Kandy, Sri
Lanka: Buddhist Publication Society, 2014. (Originally published 1937.)
Bapat, P. V., and Hirakawa, A., trans. Shan-​Chien-​P’i-​P’o-​Sha:  A Chinese Version
by Saṅghabhadra of Samantapāsādikā. Poona:  Bhandarkar Oriental Research
Institute, 1970.
Bareau, André, Sara Boin-​Webb, and Andrew Skilton. The Buddhist Schools of the Small
Vehicle. Translated from the French by Sara Boin-​Webb. Honolulu: University of
Hawaiʻi Press, 2013.
Bodhi, Bhikkhu, trans. The All-​Embracing Net of Views:  The Brahmajāla and Its
Commentaries. Kandy, Sri Lanka: Buddhist Publication Society, 2007.
Bodhi, Bhikkhu. “Anattā as Strategy and Ontology.” In Investigating the Dhamma: A
Collection of Papers by Bhikkhu Bodhi, 25–​ 39. Kandy, Sri Lanka:  Buddhist
Publication Society, 2015.
Bodhi, Bhikkhu. “The Buddha’s Survey of Views.” In Recent Researches in Buddhist
Studies:  Essays in Honour of Professor Y.  Karunadasa, edited by Bhikkhu
Dhammajoti, A. Tilakaratne, and K. Abhayawansa, 51–​69. Hong Kong: Chi Ying
Foundation, 1997.
Bodhi, Bhikkhu, and Mahāthera Nārada, trans. A Comprehensive Manual of
Abhidhamma:  The Abhidhammattha Sangaha of Ācariya Anuruddha. Kandy, Sri
Lanka: Buddhist Publication Society, 1993.
Bodhi, Bhikkhu, trans. The Connected Discourses of the Buddha:  A Translation of the
Saṃyutta Nikāya. Boston: Wisdom Publications, 2000.
Bodhi, Bhikkhu, trans. The Discourse on the Root of Existence: The Mūlapariyāyasutta
and Its Commentaries. Kandy, Sri Lanka: Buddhist Publication Society, 2006.
Bodhi, Bhikkhu, trans. The Numerical Discourses of the Buddha: A Translation of the
Aṅguttara Nikāya. Boston: Wisdom Publications, 2012.
Blackburn, Anne M. Buddhist Learning and Textual Practice in Eighteenth-​Century
Lankan Monastic Culture. Princeton, NJ: Princeton University Press, 2001.
Bond, George. “The Gradual Path as a Hermeneutical Approach to the Dhamma.”
In Buddhist Hermeneutics, edited by Donald Lopez, 29–​45. New Delhi:  Motilal
Banarsidass, 1993.
Bond, George. “The Netti Pakaraṇa:  A Theravāda Method of Intepretation.” In
Buddhist Studies in Honour of Walpola Rahula, edited by S. Balasooriya, A. Bareau,
R. Gombrich, S. Gunasingha, W. Mallawarachchi, and E. Perry. London: Gordon
Fraser, 1980.
Bond, George. “Theravada Buddhism and the Aims of Buddhist Studies.” In Studies
in the History of Buddhism, edited by A. K. Narain, 16–​28. Delhi: B.R. Publishing
Corporation, 1980.
Bond, George. The Word of the Buddha. Colombo: Gunasena, 1982.
 Bibliography 255

Braun, Erik. The Birth of Insight: Meditation, Modern Buddhism, and the Burmese Monk
Ledi Sayadaw. Chicago: University of Chicago Press, 2013.
Bronkhorst, Johannes. Buddhist Teaching in India. Somerville, MA:  Wisdom
Publications, 2009.
Brough, John. “Thus Have I Heard  . . . .” Bulletin of the School of Oriental and African
Studies 13, no. 2 (1950): 416–​26.
Buddhadatta, A. P. “Who Was Buddhaghosa?” In Corrections of Geiger’s Mahāvaṃsa,
etc.: A Collection of Monographs, edited by Wilhelm Geiger, 142–​57. Ambalangoda,
Ceylon: Ananda Book Co., 1957.
Buswell, Robert E., and Donald S. Lopez, eds. The Princeton Dictionary of Buddhism.
Princeton, NJ: Princeton University Press, 2014.
Cabezón, José. Buddhism and Language:  A Study of Indo-​ Tibetan Scholasticism.
Albany: State University of New York Press, 1994.
Cantwell, Cathy, and Elisa Freschi. “Introduction:  Reuse and Intertextuality in the
Context of Buddhist Texts.” Buddhist Studies Review 33, no. 1–​2 (2016): 1–​7.
Carbine, Jason. Sons of the Buddha: Continuities and Ruptures in a Burmese Monastic
Tradition. New York: de Gruyter, 2011.
Chaṭṭhasaṅgāyana Saṅgāyana Tipiṭaka. Igatpuri: Vipassana Research Insitute, 1995.
Collins, Steven. Nirvana and Other Buddhist Felicities: Utopias of the Pali Imaginaire.
Cambridge: Cambridge University Press, 1998.
Collins, Steven. “On the Very Idea of the Pāli Canon.” Journal of the Pali Text Society
15 (1990): 89–​126.
Collins, Steven. A Pali Grammar for Students. Chiang Mai, Thailand:  Silkworm
Books, 2006.
Collins, Steven. “Remarks on the Visuddhimagga, and on its Treatment of the Memory
of Former Dwelling(s) (pubbenivāsānussatiñāṇa).” Journal of Indian Philosophy 37
(2009): 499–​532.
Cone, Margaret. A Dictionary of Pali: Part 1. Oxford: Pali Text Society, 2001.
Cone, Margaret. A Dictionary of Pali: Part 2. Oxford: Pali Text Society, 2011.
Cousins, Lance. “The Case of the Abhidhamma Commentary.” Journal of the
International Association for Buddhist Studies 36/​37 (2015): 389–​422.
Cousins, Lance. “Dhammapāla and the Ṭīkā Literature.” Religion 2,  no. 2
(1972): 159–​65.
Cousins, Lance. “Good or Skilful? Kusala in Canon and Commentary.” Journal of
Buddhist Ethics 3 (1996): 136–​64.
Cousins, Lance. “On the Vibhajjavādins:  The Mahiṃsāsaka, Dhammaguttaka,
Kassapiya and Tambapaṇṇiya Branches of the Ancient Theriyas.” Buddhist Studies
Review 18 (2001): 131–​82.
Cousins, Lance. “Pali Oral Literature.” In Buddhist Studies: Ancient and Modern, edited
by Philip Denwood and Alexander Piatigorsky, 1–​11. London: Curzon Press, 1982.
Cousins, Lance. “The Paṭṭhāna and the Development of the Theravādin Abhidhamma.”
Journal of the Pali Text Society 9 (1981): 22–​46.
256 Bibliography

Cox, Collett. “From Category to Ontology:  The Changing Role of Dharma in

Sarvāstivāda Abhidharma.” Journal of Indian Philosophy 32 (2004): 543–​97.
Cox, Collett. “The Unbroken Treatise:  Scripture and Argument in Early Buddhist
Scholasticism.” In Innovation in Religious Traditions: Essays in the Interpretation of
Religious Change, edited by Michael Williams, Collett Cox, and Martin Jaffee, 143–​
89. Berlin: Mouton de Gruyter, 1992.
Crosby, Kate. “Sāriputta’s Three Works on the Samantapāsādikā.” Journal of the Pali
Text Society 28 (2006): 49–​59.
Crosby, Kate. Theravada Buddhism:  Continuity, Diversity, and Identity. West
Sussex: Wiley, 2004.
Crosby, Kate. Traditional Theravada Meditation and Its Modern-​Era Suppression. Hong
Kong: Buddha Dharma Centre of Hong Kong, 2013.
Davidson, Ronald M. “Appendix:  An Introduction to the Standards of Scriptural
Authenticity in Indian Buddhism.” In Chinese Buddhist Apocrypha, edited by
Robert Buswell, 291–​325. Honolulu: University of Hawai’i Press, 1990.
De Silva, Lily. “Introduction” to her edition of Dīghanikāya-​aṭṭhakathā-​tīkā
Līnatthavaṇṇanā. London: Pali Text Society, 1970.
Detienne, Marcel, and Jean-​Pierre Vernant. Cunning Intelligence in Greek Culture
and Society, translated by Janet Lloyd. Atlantic Highlands, NJ: Humanities Press,
1978. Originally published in French as Les ruses d’intelligence: La mētis des grecs.
Paris: Flammarion, 1974.
Dhirasekera, Jotiya. Buddhist Monastic Discipline: A Study of Its Origin and Development
in Relation to the Sutta and Vinaya Pitakas. Colombo: Gunasena, 1982.
Ehara, N. R.  M., Soma Thera, and Kheminda Thera, trans. The Path of Freedom
(Vimuttimagga) by the Arahant Upatissa. Colombo: The Saman Press, 1961.
Endo, Toshiichi. Buddha in Theravada Buddhism. Dehiwela: Systematic Print, 1997.
Endo, Toshiichi. “The Buddha’s Fathom Long Halo (Byamappabhā) and Rays
(Ramsi):  A Critical Survey in the Pali Commentarial Literature.” Journal of the
Centre for Buddhist Studies, Sri Lanka 12 (2016): 91–​106.
Endo, Toshiichi. Studies in Pāli Commentarial Literature: Sources, Controversies, and
Insights. Hong Kong: Centre of Buddhist Studies, 2013.
Evans, Stephen. “Epistemology of the Brahmajāla Sutta.” Buddhist Studies Review 26,
no. 1 (2009): 67–​84.
Finot, M. Louis. “The Legend of Buddhagosa.” Translated from the French by P. C.
Bagchi. Journal of the Department of Letters, University of Calcutta 2 (1924): 65–​86.
Frauwallner, E. The Earliest Vinaya and the Beginnings of Buddhist Literature. Rome: Is.
M.E.O., 1956.
Frauwallner, E. Studies in Abhidharma Literature and the Origins of Buddhist
Philosophical Systems. Translated from the German by Sophie Frances Kidd.
Albany: State University of New York Press, 1995.
Geiger, Wilhelm, trans. Cūlavaṃsa. Translated from the German by C. Mabel Rickmers.
London: Pali Text Society, 1992.
 Bibliography 257

Geiger, Wilhelm. Pāli Literature and Language. Translated by Batakrishna Ghosh. New
Delhi: Oriental Books Reprint Corporation, 1978. Originally published 1943.
Gethin, Rupert. “The Five Khandhas:  Their Treatment in the Nikāyas and Early
Abhidhamma.” Journal of Indian Philosophy 14 (1986): 35–​53.
Gethin, Rupert. “The Mātikās:  Memorization, Mindfulness, and the List.” In In the
Mirror of Memory:  Reflections on Mindfulness and Remembrance in Indian and
Tibetan Buddhism, edited by Janet Gyatso, 149–​72. Albany:  State University of
New York Press, 1992.
Gethin, Rupert. “On the Nature of Dhammas:  A Review Article.” Buddhist Studies
Review 22 (2005): 175–​94.
Gethin, Rupert. “Was Buddhaghosa a Theravādin? Buddhist Identity in the Pali
Commentaries and Chronicles.” In How Theravāda Is the Theravāda? Exploring
Buddhist Identities, edited by P. Skilling, J. Carbine, C. Cicuzza, and S. Pakdeekham,
1–​63. Chiang Mai: Fragile Palm Leaves Foundation; Silkworm Books, 2012.
Gethin, Rupert. “Wrong View (micchā-​diṭṭhi) and Right View (sammā-​diṭṭhi) in the
Theravāda Abhidhamma.” Contemporary Buddhism 5, no. 1 (2004): 15–​28.
Gold, Jonathan. Paving the Great Way:  Vasubandhu’s Unifying Buddhist Philosophy.
New York: Columbia University Press, 2016.
Gomez, Luis O. “Buddhist Books and Texts:  Exegesis and Hermeneutics.” In
Encyclopedia of Religion, 2nd ed., vol. 2, edited by Lindsay Jones, 1268–​ 78.
Detroit: Thomson Gale, 2005.
Goonesekere, L. R. Buddhist Commentarial Literature. Kandy, Sri Lanka:  Buddhist
Publication Society, 2008. http://​www.bps.lk/​olib/​wh/​wh113.pdf
Gornall, A. M. “Some Remarks on Buddhaghosa’s use of Sanskrit Grammar: Possible
Hints of an Unknown Pāṇinian Commentary in Buddhaghosa’s Grammatical
Arguments.” Journal of the Oxford Centre for Buddhist Studies 1 (2011): 89–​107.
Granoff, Phyllis. “Buddhaghosa’s Penance and Siddhasena’s Crime:  Remarks on
Some Buddhist and Jain Attitudes Towards the Language of Religious Texts.” In
From Benares to Beijing:  Essays on Buddhism and Chinese Religion in Honour of
Prof. Jan Yün-​hua, edited by Koichi Shinohara and Gregory Schopen, 17–​33.
New York: Mosaic Press (1991).
Gray, James, trans. Buddhaghosuppatti; or, The Historical Romance of the Rise and Career
of Buddhaghosa. Oxford: Pali Text Society, 2001. First published 1892.
Grero, C. Ananda. “The Philosophy of Theravada Vinaya.” In Recent Researches in
Buddhist Studies:  Essays in Honour of Professor Y.  Karunadasa, edited by K. L.
Dhammajoti, A. Tilakaratne, and K. Abhayawansa, 242–​57. Hong Kong:  Y.
Karunadasa Felicitation Committee, 1997.
Griffiths, Paul J. “Denaturalizing Discourse:  Ᾱbhidhārmikas, Propositionalists, and
the Comparative Philosophy of Religion.” In Myth and Philosophy, edited by Frank
Reynolds and David Tracy, 57–​91. Albany: State University of New York Press, 1990.
Griffiths, Paul J. Religious Reading:  The Place of Reading in the Practice of Religion.
New York: Oxford University Press, 1999.
258 Bibliography

Grimes, John. A Concise Dictionary of Indian Philosophy:  Sanskrit Terms Defined in

English. Albany: State University of New York Press, 1996.
Hadot, Pierre. Philosophy as a Way of Life: Spiritual Exercises from Socrates to Foucault.
Edited by Arnold I. Davison, translated by Michael Chase. Malden, MA: Blackwell
Publishing, 1995.
Hallisey, Charles. “Buddhaghosa on the Futures of Scriptures.” AAR paper, November
22, 2014.
Hallisey, Charles. “Councils as Ideas and Events in the Theravāda.” Buddhist Forum 2
(1991): 133–​48.
Hallisey, Charles. Devotion in the Medieval Literature of Sri Lanka. PhD diss., University
of Chicago, 1988.
Hallisey, Charles. “In Defense of Rather Fragile and Local Achievement: Reflections
on the Work of Gurulugomi.” In Religion and Practical Reason: New Essays in the
Comparative Philosophy of Religions, edited by Frank Reynolds and David Tracy,
121–​60. Albany: State University of New York Press, 1994.
Harvey, Van A. “Hermeneutics.” In Encyclopedia of Religion, 2nd ed., edited by Lindsey
Jones, 3930–​36. Farmington Hills, MI: Thomson Gale.
Heim, Maria. “Buddhaghosa on the Phenomenology of Love and Compassion.” In
The Oxford Handbook of Indian Philosophy, edited by Jonardon Ganeri, 171–​89.
New York: Oxford University Press.
Heim, Maria. “The ‘Fecundity of Dialogue’ and the ‘Philosophy of Incompletion.’” In
Buddhist Spiritual Practices: Thinking with Pierre Hadot on Buddhism, Philosophy,
and the Path, edited by David V. Fiordalis, 181–​214. Berkeley, CA:  Mangalam
Press, 2018.
Heim, Maria. The Forerunner of All Things:  Buddhaghosa on Mind, Intention, and
Agency. New York: Oxford University Press, 2014.
Heim, Maria. “She Who Heard Much.” International Journal of Hindu Studies 19, no.
1–​2 (2015): 139–​56.
Heim, Maria. “Visuddhimagga/​Buddhaghosa.” In Oxford Bibliographies for
Buddhism, edited by Richard Payne. New  York:  Oxford, 2017 http://​www.
oxfordbibliographies.com/​obo/​page/​buddhism.
Heim, Maria, and Chakravarthi Ram-​ Prasad. “In a Double Way:  Nāma-​Rūpa
in Buddhaghosa’s Phenomenology.” Philosophy East and West 69, no. 1
(forthcoming, 2019).
Holt, John. Discipline:  The Canonical Buddhism of the Vinayapiṭaka. Delhi:  Motilal
Banarsidass, 1983.
Horner, I. B., trans. The Book of the Discipline (Vinaya-​Piṭaka). Vol. IV. Lancaster, UK:
Pali Text Society, 2007 (1951).
Hultzsch, E. Inscriptions of Asoka. Oxford: Clarendon Press, 1925.
Huxley, Andrew. “Buddhist Case Law on Theft:  The Vinītivatthu on the Second
Pārājika.” Journal of Buddhist Ethics 6 (1999): 313–​30.
 Bibliography 259

Jaini, Padmanabh. “On the Sarvajñatva (Omniscience) of Mahāvīra and the Buddha.”
In Collected Papers on Buddhist Studies, 97–​122. Delhi: Motilal Banarsidass, 2001.
Jayatilleke, K. N. Early Buddhist Theory of Knowledge. Delhi:  Motilal Banarsidass
Publishers, 1980.
Jayawickrama, N. A., trans. The Inception of the Discipline and the Vinaya Nidāna.
London: Pali Text Society, 1986.
Jayawickrama, N. A., trans. The Story of Gotama Buddha (Jātaka-​nidāna). Oxford: Pali
Text Society, 2002.
Kahrs, Eivind. Indian Semantic Analysis:  The Nirvacana Tradition.
Cambridge: Cambridge University Press, 1998.
Kalupahana, David. A History of Buddhist Philosophy, Continuities and Discontinuities.
Delhi: Motilal Banarsidass, 2006.
Kalupahana, David. Nāgārjuna:  The Philosophy of the Middle Way. Albany:  State
University of New York Press, 1986.
Karunadasa, Y. “The Buddhist Theory of Double Truth.” Kalyāni: Journal of Humanities
and Social Sciences of the University of Kelaniya, Colombo 3–​4 (1985): 23–​55.
Karunadasa, Y. The Dhamma Theory:  Philosophical Cornerstone of the Abhidhamma.
Kandy, Sri Lanka: Buddhist Publication Society, 1996.
Keown, Damien. “Morality in the Visuddhimagga.” Journal of the International
Association for Buddhist Studies 6, no. 1 (1983): 61–​75.
Keyes, Charles F. “Merit-​Transference in the Kammic Theory of Popular Theravāda
Buddhism.” In Karma: An Anthropological Inquiry, edited by C. F. Keyes and E. V.
Daniels, 261–​86. Berkeley: University of California Press.
Kieffer-​Pülz, Petra. “Reuse of Text in Pāli Legal Commentaries.” Buddhist Studies
Review 33, no. 1–​2 (2016): 9–​45.
Lamotte, Étienne. “Assessment of Textual Interpretation in Buddhism.” In Buddhist
Hermeneutics, edited by Donald Lopez, 11–​27. New Delhi: Motilal Banarsidass, 1993.
Lamotte, Étienne. History of Indian Buddhism:  From the Origins to the Saka Era.
Translated from the French by Sara Webb-​Boin. Louvain-​La.-​Neuve:  Université
catholique de Louvain, Institut orientaliste, 1988.
Law, B. C. Buddhaghosa. Bombay: Royal Asiatic Society, 1946.
Law, B. C., trans. The Debates Commentary (Kathāvatthuppakaraṇa-​Aṭṭhakathā).
London: The Pali Text Society, 1969.
Le Roux, Ina. “The Brahmajāla Sutta:  A Rediscovery of Its Oral Features.” In Oral
Tradition and Innovation: New Wine in Old Bottles?, edited by E. R. Sienaert, A.
N. Bell, and N. Lewis, 47–​67. Natal: University of Natal Oral Documentation and
Research Centre, 1991.
Lopez, Donald, ed. Buddhist Hermeneutics. New Delhi: Motilal Banarsidass, 1993.
Malalasekera, G. P. Dictionary of Pāli Proper Names. Vols. 1–​2. New Delhi: Munshiram
Manoharlal Publishers, 1995.
Malalasekera, G. P. The Pāli Literature of Ceylon. Kandy, Sri Lanka:  Buddhist
Publication Society, 1994.
260 Bibliography

Masefield, Peter, trans. The Commentary on the Itivuttaka (The Itivuttaka-​aṭṭhakathā of

Dhammapāla). Vols. 1–​2. Oxford: Pali Text Society, 2008–​2009.
Maskell, Dorothy, ed. Kaṅkhāvitaraṇī nāma Mātikaṭṭhakathā:  Buddhaghosa’s
Commentary on the Pātimokkha. London: Pali Text Society, 1956.
McClintock, Sara. Omniscience and the Rhetoric of Reason: Śāntarakṣita and Kamalaśīla
on Rationality, Argumentation, and Religious Authority. Boston:  Wisdom
Publications, 2010.
McQueen, Graeme. “Inspired Speech in Early Mahāyāna Buddhism I. Religion 11
(1981): 303–​19.
Minkowski, Christopher. “Why Should We Read the Maṅgala Verses?” In
Śāstrārambha: Inquiries into the Preamble in Sanskrit, edited by Walter Slaje, 1–​24.
Wiesbaden: Harassowitz Verlag, 2008.
Monier-​Williams, Monier. A Sanskrit-​ English Dictionary. New Delhi:  Asian
Educational Services, 1999.
Mori, Sodo. “The Milindapañha and the Pāli Aṭṭhakathā Literature.” Indologica
Taurinesis 23–​24 (1998–​1999): 291–​312.
Mori, Sodo. “The Vijñānavādin View as Depicted in the Pāli Commentaries with Special
Reference to the Nirayapālakathā.” In Bauddhavidyāsudhākaraḥ: Studies in Honour
of Heinz Bechert on Occasion of His 65th Birthday, edited by Petra Kieffer-​Pülz and
Jens-​Uwe Hartmann, 453–​64. Swisttal-​Odendorf: Indica & Tibetica, 1997.
Morris, Richard, ed. The Puggala-​Paññatti. London: Pali Text Society, 1883.
Ñāṇamoli, Bhikkhu, trans. The Dispeller of Delusion (Sammohavinodanī). Revised
for Publication by L. S. Cousins, Nyanaponika Mahāthera, and C. M.  M. Shaw.
Oxford: Pali Text Society, 1996
Ñāṇamoli, Bhikkhu. The Guide (Nettippakaraṇa). London: The Pali Text Society, 1977.
Ñāṇamoli, Bhikkhu, trans. The Minor Readings and Illustrator (Khuddaka Pāṭha).
London: Pali Text Society, 1978.
Ñāṇamoli, Bhikkhu, trans. The Path of Discrimination (Paṭisambhidāmagga).
London: Pali Text Society, 1982.
Ñāṇamoli, Bhikkhu, trans. The Path of Purification (Visuddhimagga) by Bhadantācariya
Buddhaghosa. Kandy, Sri Lanka: Buddhist Publication Society, 1975.
Ñāṇamoli, Bhikkhu, and Bhikkhu Bodhi, trans. The Middle Length Discourses of
the Buddha:  A Translation of the Majjhima Nikāya. Somerville, MA:  Wisdom
Publications, 1995.
Nance, Richard. “Buddhist Hermeneutics.” In Oxford Bibliographies, edited by Richard
Payne. New York: Oxford University Press, 2013. (http://​www.oxfordbibliographies.
com/​view/​document/​obo-​9780195393521/​obo-​9780195393521-​0155.xml)
Nance, Richard. “Mindsets and Commentarial Conventions among Indian Buddhists.”
Journal of the American Academy of Religion 83, no. 1 (2015): 210–​35.
Nance, Richard. Speaking for Buddhas:  Scriptural Commentary in Indian Buddhism.
New York: Columbia University Press, 2012.
 Bibliography 261

Nance, Richard. “The Voice of Another:  Speech, Responsiveness, and Buddhist

Philosophy.” In A Companion to Buddhist Philosophy, edited by Steven M.
Emmanuel, 366–​76. Oxford: Wiley, 2013.
Naughton, Alex. “Buddhist Omniscience.” Eastern Buddhist 24, no. 1 (1991): 28–​51.
Norman, K. R. Pāli Literature. Wiesbaden: Otto Harrossowitz, 1983.
Nyanaponika Thera. Abhidhamma Studies:  Buddhist Explorations of Consciousness
and Time. Edited with an Introduction by Bhikkhu Bodhi. Boston:  Wisdom
Publications, 1998.
Nyanatiloka Mahāthera. Guide through the Abhidhamma Piṭaka. Kandy, Sri
Lanka: Buddhist Publication Society, 2008. Originally published 1938.
Palihawadana, Mahinda. “Dhammapada 1 and 2 and their Commentaries.” Buddhist
Studies in Honour of Hammalava Saddhātissa, ed. H. Saddhātissa, G. Dhammapala,
R. Gombrich, and K.R. Norman, 189–​202. Nugegoda, Sri Lanka: University of Sri
Jayewardenepura, 1984.
Pe Maung Tin, trans., with editing by Caroline Rhys Davids. The Expositor. Oxford: Pali
Text Society, 1999.
Perrett, Roy W. “Omniscience in Indian Philosophy of Religion.” In Indian Philosophy
of Religion, edited by Roy W. Perrett, 125–​42. Springer: Studies in Philosophy and
Religion, 1989
Pind, Ole Holten. “Buddhaghosa:  His Works and Scholarly Background.” Bukkyō
kenkyū 21 (1992): 135–​56.
Pind, Ole Holten. “Pāli Miscellany.” In Bauddhavidyāsudhākaraḥ: Studies in Honour of
Heinz Bechert on the Occasion of His 65th Birthday, edited by Petra Kieffer-​Pülz and
Jens-​Uwe Hartmann, 515–​36. Swisttal-​Odendorf: Indica et Tibetica, 1997.
Pind, Ole Holten. “Studies in the Pāli Grammarians I:  Buddhaghosa’s Reference to
Grammar and the Grammarians.” Journal of the Pali Text Society 13 (1989): 33–​82.
Pind, Ole Holten. “Studies in the Pāli Grammarians II.” Journal of the Pali Text Society
14 (1990): 175–​218.
Pollock, Sheldon. “Future Philology? The Fate of a Soft Science in a Hard World.”
Critical Inquiry 35 (2009): 931–​61.
Pollock, Sheldon. “Introduction.” In World Philology, edited by Sheldon Pollock,
Benjamin A. Elman, and Ku-​ming Kevin Chang. Cambridge, MA:  Harvard
University Press, 2015.
Pollock, Sheldon. “What Was Philology in Sanskrit?” In World Philology, edited
by Sheldon Pollock, Benjamin A. Elman, and Ku-​ming Kevin Chang, 114–​36.
Cambridge, MA: Harvard University Press, 2015.
Prasad, C. S. “Theravāda and Vibhajjavāda: A Critical Study of Two Appellations.” East
and West 22 (1972): 101–​13.
Ramanujan, A. K. “Is There an Indian Way of Thinking? An Informal Essay.”
Contributions to Indian Sociology, n.s., 23, no. 1 (1989): 41–​58.
Ram-​Prasad, Chakravarthi. Human Being, Bodily Being. London: Oxford University
Press, 2018.
262 Bibliography

Ram-​Prasad, Chakravarthi. Indian Philosophy and the Consequences of Knowledge: Themes

in Ethics, Metaphysics and Soteriology. Abingdon: Routledge, 2007.
Rhys Davids, Caroline A.  F., trans. A Buddhist Manual of Psychological Ethics
(Dhammasaṅgaṇī). London: Royal Asiatic Society, 1900 (1975).
Rhys Davids, T. W., and William Stede. The Pali Text Society’s Pali-​English Dictionary.
Oxford: Pali Text Society, 1925 (1992).
Ricoeur, Paul. Interpretation Theory:  Discourse and the Surplus of Meaning. Fort
Worth: Texas Christian University Press, 1976.
Ronkin, Noa. Early Buddhist Metaphysics:  The Making of a Philosophical Tradition.
New York: RoutledgeCurzon, 2005.
Samtani, N. H. “The Opening of the Buddhist Sūtras.” Bharati: Bulletin of the College
of Indology 8, no. 2 (1964): 47–​63.
Sasaki, Genjun H. “Pariyāya und Nippariyāya.” Wiener Zeitschrift für die Kunde Süd-​
und Ostasiens 6 (1962): 47–​59.
Schlingloff, Dieter von. “Zur Interpretation des Prātimokṣasūtra.” Zeitschrift der
Deutschen Morgenländischen Gesellschaft 113 (1964): 536–​51.
Sayadaw, Ledi. “Some Points in Buddhist Doctrine.” Journal of the Pali Text Society 7
(1913–​1914): 115–​69.
Scholem, Gershom. Kabbalah. Jerusalem: Keter Publishing House, 1974.
Scott, James C. Seeing Like a State. New Haven, CT: Yale University Press, 2008.
Sernesi, Marta. “The Collected Sayings of the Master: On Authorship, Author-​Function,
and Authority.” Journal of the International Association of Buddhist Studies 36/​37
(2013–​2014): 459–​98.
Shaw, Sarah. “In What Way Is There a Saṅghavacana? Finding the Narrator, Author and
Editor in Pāli Texts.” Journal of the International Association of Buddhist Studies 36/​
37 (2013–​2014): 423–​58.
Shusterman, Richard. “Philosophy as a Way of Life: As Textual and More Than Textual
Practice.” In Philosophy as a Way of Life: Ancients and Moderns: Essays in Honor of
Pierre Hadot, edited by Michael Chase, Stephen R. L. Clark, and Michael McGhee,
40–​56. West Sussex: Wiley, 2013.
Siderits, Mark. Buddhism as Philosophy: An Introduction. Hants, U.K.: Ashgate, 2007.
Skilling, Peter. “Dharma, Dhāraṇī, Abhidharma, Avadāna:  What was taught in
Trayastriṃśa?” Annual Report of The International Research Institute for Advanced
Buddhology 11 (2008): 37–​60.
Skilling, Peter. “Mahāyāna and Bodhisattva:  An Essay towards Historical
Understanding.” In Bodhisattvaparami and Thai Society in the New Millennium,
edited by Pakorn Limpanusor and Chalermpon Iampakdee, 139–​ 56.
Bangkok: Thammasat University Press, 2004.
Skilling, Peter. “Scriptural Authenticity and the Śrāvaka Schools: An Essay Towards an
Indian Perspective.” Eastern Buddhist 41, no. 2 (2010): 1–​47.
Skilling, Peter. “Vasubandhu and the Vyākhyāyukti Literature.” Journal of the
International Association of Buddhist Studies 23, no. 2 (2000): 297–​350.
 Bibliography 263

Skilling, Peter, Jason Carbine, Claudio Cicuzza, and Santi Pakdeekham, eds. How
Theravada Was the Theravada? Seattle: University of Washington Press, 2012.
Skilton, Andrew. “Theravāda.” In A Companion to Buddhist Philosophy, edited by
Stephen M. Emmanuel, 71–​85. West Sussex: Blackwell, 2013.
Sodo Mori, Aichi. “The Vijñānavādin View as Depicted in the Pāli Commentaries with
Special Reference to the Nirayapāla-​kathā.” In Bauddhavidyāsudhākaraḥ: Studies
in Honour of Heinz Bechert on the Occasion of His 65th Birthday, edited by Petra
Kieffer-​Pülz and Jens-​Uwe Hartmann, 453–​64. Odendorf, 1997.
Strong, John S. “Buddhaghosa.” In Encyclopedia of Buddhism, vol. 1, edited by Robert E.
Buswell, 75. New York: Thomson Gale 2004.
Swearer, Donald K., ed. Me and Mine:  Selected Essays of Bhikkhu Buddhadasa.
Albany: State University of New York Press, 1989.
Syrkin, A. “On the First Work of the Sutta Piṭaka: The Brahmajāla Sutta.” In Buddhist
Studies: Ancient and Modern, edited by Philip Denwood and Alexander Piatigorsky,
153–​166. London: Curzon Press, 1982
Thanissaro Bhikkhu, trans. “Mulapariyaya Sutta: The Root Sequence” (MN 1). Access
to Insight (Legacy Edition). November 30, 2013. http://​www.accesstoinsight.org/​
tipitaka/​mn/​mn.001.than.html.
Thanissaro Bhikkhu. “The Non-​ Self Strategy.” Access to Insight. http://​www.
accesstoinsight.org/​lib/​authors/​thanissaro/​notselfstrategy.pdf.
Thanissaro Bhikkhu, trans. “Ratha-​vinita Sutta:  Relay Chariots” (MN 24). Access to
Insight (Legacy Edition). November 30, 2013. http://​www.accesstoinsight.org/​
tipitaka/​mn/​mn.024.than.html.
Thittila, Ashin, trans. The Book of Analysis (Vibhaṅga). Oxford: Pali Text Society, 2002.
U Nārada, trans. Conditional Relations (Paṭṭhāna). London: Pali Text Society, 1969.
Upasak, C. S. Dictionary of Early Buddhist Monastic Terms. Varanasi:  Bharati
Prakashan, 1975.
von Hinüber, Oskar. “Building the Theravāda Commentaries:  Buddhaghosa and
Dhammapāla as Authors, Compilers, Redactors, Editors and Critics.” Journal of the
International Association for Buddhist Studies, 36/​37 (2015): 353–​88.
von Hinüber, Oskar. A Handbook of Pali Literature. New Delhi:  Munshiram
Manoharlal.
von Hinüber, Oskar. “Zu einer Göttingere Dissertation über das Buddhistische Recht.”
Wiener Zeitschrift für die Kunde Südasiens /​Vienna Journal of South Asian Studies
40 (1996): 101–​13.
Walshe, Maurice, trans. The Long Discourses of the Buddha: A Translation of the Dīgha
Nikāya. Boston: Wisdom Publications, 1995.
Walshe, Maurice O’C. Meister Eckhart:  German Sermons and Treatises.
London: Watkins, 1979.
Walters, Jonathan. “Buddhist History:  The Sri Lankan Pāli Vaṃsas and Their
Commentary.” In Querying the Medieval: Texts and the History of Practices in South
Asia, 99–​164. New York: : Oxford University Press, 2000.
264 Bibliography

Walters, Jonathan. “Mahāyāna Theravāda and the Origins of the Mahāvihāra.” Sri
Lanka Journal of the Humanities 23, no. 1–​2 (1997): 100–​19.
Walters, Jonathan. “Suttas as History: Four Approaches to the Sermon on the Noble
Quest (Ariyapariyesanasutta).” History of Religions 38, no. 3 (1999): 248–​84.
Warder, A. K. Indian Buddhism. Delhi: Motilal Banarsidass, 1980.
Warder, A. K. “Some Problems of the Later Pali Literature.” Journal of the Pali Text
Society 9 (1981): 198–​202.
Westerhoff, Jan. “Metaphysical Issues in Indian Buddhist Thought.” In A
Companion to Buddhist Philosophy, edited by Steven Emmanuel, 129–​ 50.
Oxford: Wiley-​Blackwell, 2013
Williams, Paul. “On the Abhidharma Ontology.” Journal of Indian Philosophy 9, no. 3
(1981): 227–​57.
 Index

Page numbers followed by n indicate notes.

Abhidhamma, 144–​83 Abhinavagupta, 83n74

and Buddha’s omniscience, 17, 25, Adikaram, E. W., 21n52, 27n69
45–​46, 53, 56 Ajātasattu, 133n61
Buddha’s teaching of, 40–​43, 157–​62, ākāra (mode, aspect, register), 74, 76n56,
183, 187 89n93, 169n46, 171n49, 176n64,
commentaries on, 19, 19n48, 20, 23, 24, 177n67, 237
145n5, 200, 239–​52 Almond, Ian, 16
introduction (nidāna) to, 5, 53–​54, Analysis, see Asking Questions;
117–​18, 156–​61, 181, 198, 216, 221 paṭisambhidā; Vibhajjavāda
methods of, 6, 15, 42–​43, 46, 49–​57, in the Abhidhamma, 29, 37, 46, 49,
63, 73, 93–​94, 94n104, 101–​103, 64, 92, 144–​83, 221, 239
106, 109, 130, 146–​68, 174–​82, 196, from all sides, 174–​80
230, 232, 239–​41, 244 nirutti/​nerutta analysis, 79–​85, 79n68,
as nippariyāya or abstract teaching, 29, 83n74, 118, 190n13, 193, 203, 225,
97–​103, 105, 110, 162–​63, 175, 181, 221 229–​30, 235–​38, 240 (see also
as oceanic and immeasurable, 49, 52–​ paṭisambhidā)
57, 154–​63, 178–​79, 221, 239–​52 paramattha analysis, 89–​94, 173
as ontology or metaphysics, 146–​49, phenomenological, 23, 94n104,
147n7, 168–​74, 180, 182 144–​83,  221
as paramattha (furthest sense) analysis, vibhajja, 68, 70, 174, 186
7, 7n13, 89–​94, 96n109, 168–​74, 177 vibhaṅga, 61n3, 64, 174, 186, 189–​90, 193,
Abhidhamma Analysis 213–​14n87, 246 (see also Vibhaṅga;
(abhidhammabhājanīya), 45–​46, Suttavibhaṅga)
64, 93, 151, 155, 167, 174–​77 Analysts (Vibhajjavādins), 67–​70, 93, 176
Abhidhammatthasaṅgaha, 91n98, Ānanda, 34, 37, 41, 44, 112, 116–​17, 186,
149,  167–​68 209, 214–​15, 244
266 Index

āṇāpātimokkha (Pātimokkha of the bhāvanānaya (method of

Rules), 212 development),  68–​69
ancient commentators, 21, 66, 66n25 Bhikkhu, Thanissaro, 140, 182n84
Andhra language, 88 Bodhi, Bhikkhu, 91n98, 95n106,
Aṅguttara Nikāya, 8, 36, 68n32, 73–​74, 96n109, 117n17, 125n35, 138, 140,
94, 98, 116n14, 156n23, 202, 246 148n11, 182n84
Anuruddha,  167–​68 Bodhi Tree, 1, 35, 45, 54, 81, 81n71,
Apadāna, 36n8, 246 156–​60, 234–​35, 239, 241, 245
Araha (Worthy), 80–​82, 203, 230–​31, Bond, George, 17, 17n40, 26, 65–​66,
235–​38. See also itipiso contemplation 68–​69, 70n38, 84, 105, 118–​19
Ariyapariyesanasutta (Sermon on the brahmacariya (religious life), 10n19, 224
Noble Quest), 26, 114 Brahmadatta,  131–​33
Asking Questions (pañhāpucchaka), Brahmajāla Sutta, 49, 117, 126–​36, 142,
174–​75,  177–​78 161, 192, 192n18, 197, 216, 249
Assaji, 84, 99, 102–​3 Buddhaghosa’s reading of, 128–​32, 135
Aśvaghoṣa,  114–​15 nidāna (introduction), 124, 126
attha, 7, 10, 21, 29, 39–​40, 48, 50–​52, Brahmins, 133, 206, 231, 234–​35
60, 62, 65, 129, 193, 250. See also in the Mūlapariyāya story, 137, 140, 142
meaning Brahmin Verañja 20, 185, 197, 198,
and phrasing (byañjana), 10, 48, 60, 201–​2, 203n51, 206–​9, 213, 216–​17,
70, 73–​85, 119, 120, 190, 206, 213, 223,  229–​38
214, 222–​26, 231 brief teachings (saṃkhitta, saṃkhepa), 7,
aṭṭhakathā, 5, 29, 57, 60, 65, 68–​70, 29, 42–​43, 51, 60, 64, 68–​73, 84, 143,
94n104, 117, 171–​72, 199–​200n43 189n8, 193, 211–​12, 221, 222, 237, 250
authorship, 5n8, 5n9, 18–​22, 65, 189n8 Buddha
and Visuddhimagga, 20, 22, 26, 70, 72, biography, 114, 117, 181, 186–​88n3, 191,
80n70, 220 203, 216
Atthasālinī, 5, 16n33, 24, 35n6, 42n24, devotion to, 15, 204, 216–​17
40, 45, 49, 54, 69n38, 96–​97, 101, epithets, 80–​83, 85, 190n13, 202–​7,
117–​18, 145n5, 154n17, 156–​61, 216–​17, 229–​30,  248
171–​73, 196n30, 200n44, 239–​52 omniscience, 2, 14–​18, 25, 33–​59,
atthavaṇṇanā (meaning commentary), 103–​10, 113, 121, 122–​24, 126–​31,
65–​66,  84 133–​36, 138, 139–​41, 143, 149, 154–​55,
157, 162–​63, 179–​81, 184–​85, 195,
bahirnidāna (External Nidāna), 205–​206, 208, 217–​19, 221, 239–​52
196–​97,  202 recollections of, 69, 80, 83,
Bahuvedanīya Sutta (Many Types of 203–​5,  222–​23
Feeling Sutta), 98n114, 150–​51, teachings as immeasurable, 2, 8–​9,
153, 170 14–​18, 21, 27, 33–​59, 68, 73–​85,
Balcerowicz, Piotr, 22–​23, 45n35 103–​7, 110–​11 (see also omniscience)
Bhairava, 83n74 Buddha (term), 47–​49
 Index 267

Buddhacarita (Aśvaghoṣa),  114–​15 Indian, 17, 27n69, 147n7

Buddhadāsa, Bhikkhu, 24 Mahāyāna, 15, 78n63, 83–​84, 91–​92,
Buddhaghosa (“Voice of the Buddha”) 104–​5,  118n22
as editor, translator and author, 5, Theravada, 17, 23–​24, 27–​28
5n8, 18–​24, 19n48, 19n49, 21n52, buddhology, 14–​15, 26, 33, 44, 85, 114,
33n1, 46–​7, 66–​9, 72, 67n27, 81n71, 123, 207
83, 145n5, 146, 189n8, 199n43, Burmese tradition, 24, 36n8, 178
204, 211n79 byañjana (phrasing), 7, 10, 48, 60, 70,
legend of, 1–​2, 22, 42 73–​85, 120, 190, 203, 213–​14, 222,
as Vibhajjavādin (Analyst), 67–​68, 70, 223–​26, 234. See also attha, nirutti
104–​5, 149, 153–​54, 176
Buddha Metteyya, 2 Cabezón, José, 58–​59, 58n76
buddhānussati (Recollection of the Cariyāpiṭaka, 36n8, 246
Buddha) contemplation, 69, 80, 83, categorical (nippariyāya) teachings, 7, 29,
203–​5,  222–​23 97–​103, 112–​13, 162–​63, 175,  221–​22
Buddha Padumuttara, 125–​26 Catuḥpratisaraṇasūtra, 83–​84,  104–​5
buddhas, 12, 41, 47, 57–​58, 66, 120, 130, Chaṭṭha Saṅgāyana, 6n11
157, 165, 184, 192, 196, 210–​212, 214, Collins, Steven, 26, 113–​14
215, 219, 225, 240–​41, 244–​45 colloquial talk (vohāra), 6n12, 63, 89,
buddhavacana (Buddha’s word), see also 91n97, 94, 105–​6, 249. See also
specific texts sammuti
Buddhaghosa’s conception of, 29, commentary, 65-​66, see also aṭṭhakathā
33–​59, 60–​61, 73–​85, 105–​10, 138, contemplative exercises, see meditation
180, 184, 187, 195 contextual teachings, see pariyāya
distinctions about, 7, 60–​106, conventional language and teachings, see
221, 252n5 sammuti, vohāra
divisions of, 34–​38, 187 Cousins, Lance, 19n48, 145n5
as endless and immeasurable, 2, 8–​9, Crosby, Kate, 17, 78
14–​18, 29, 33–​59, 155, 222, 243–​44 Cūḷavagga, 71n43, 72, 186–​90, 201–​2, 215
expansive nature of, 12, 43, 49,
58–​59, 73, 109, 155, 179, 184, 200, Dānasāgara, 53n61
217–​18,  221–​22 Davidson, Ronald, 17, 18
at First Council, 18, 38, 43, 58–​59, 65, Deer Park of Isipatana, 158
112, 116–​17, 121, 186 definitive (nītattha) language, 7, 29,
language of, 77–​79, 83, 89 60–​61, 94–​97, 99n120,  221–​22
as occasioned, 137, 184–​85, 198 deities (devatā), 35, 78, 82, 99, 132, 157–​61,
praises of, 7, 8, 9, 10–​14 223, 231, 233, 237, 242–​45
well–​spoken words, 10, 33, 59, 76, dependent origination, 67, 81, 125–​26,
141,  150–​52 130–​31, 146, 148, 149, 153, 163,
Buddhavaṃsa, 36n8, 246 174–​76, 178, 181, 230–​37
Buddhism, 25, 120, 140, 182, 194 Derrida, Jacques, 16
268 Index

de Silva, Lily, 5n9, 70n38, 118–​19 Endo, Toshiichi, 26, 46, 154n17

detailed teachings (vitthārena), 7, 29, epistemology, 22, 46–​47, 57n73, 59, 85,
42–​43, 51, 60, 64, 68–​73, 71, 73, 84, 91n98, 171
143, 175, 221, 250 expertise in piṭakas (pariyatti), 6, 49–​58, 61,
Devī, 15 156, 190–​91, 196, 203–​4, 215, 230, 235
Dhamma, the, 34–​35 exposition (vinicchaya), 21, 66, 67n28
immediate impact, 13–​14 External Nidāna (bahirnidāna),
ninefold division, 34–​37, 37n10 196–​97,  202
Recollection of the Dhamma, 9–​14,
69, 206, 222, 223–​28 false speech, 91, 91n98
Turning of the Wheel of the feeling (vedanā), 127–​29, 131, 142, 150–​52,
Dhamma, 158 156, 156n26, 161–​64, 166, 169–​71,
Dhammapada, 35n6, 36n8, 235, 246 198, 236–​37, 248
Dhammapāla, 5, 5n9, 46, 46n40, 47, figurative teaching, see pariyāya
70n38, 85n82, 91n98, 133 First Council, 18, 34–​38, 43, 58–​59, 65,
commentaries, 91n98, 200n44, 223 69, 112, 116–​17, 121, 184–​86, 201–​2,
on nidānas,  118–​23 214–​15,  220
dhammas (phenomena), 46–​47, 51n55, Frauwallner, Erich, 105n135, 114n9, 187n3
62, 63, 59n77, 90–​94, 94n104, 129, furthest sense, see paramattha
146, 148n10, 149, 152, 156, 160,
164–​74, 176, 198, 226, 227, 237, 241, genres, 4–​7, 19, 23–​24, 26, 28, 29, 33, 36,
243, 245, 247–​50, 252 38, 49, 50, 54, 58, 60. See also piṭaka
Dhammasaṅgaṇī (Enumeration of Buddhaghosa on, 89, 192, 220–​21
Dhammas), 5, 62–​65, 101–​2, 129, for Ricouer, 50
146–​48, 152–​54, 160–​61, 164–​65, Gethin, Rupert, 43n28, 68, 148n11,
167, 169–​72, 174–​75, 177–​79, 171–​72, 172n52, 178n72
197–​98, 242, 246 Gotamaka Sutta, 139, 141
Dhātukathā (Discourse on Elements), Great Exposition on Kamma, 64n16
146, 242
Dhirasekera, Jotiya, 188–​89 Hadot, Pierre, x, 141–​42
Dīgha Nikāya, 5, 36, 116n14, 117, 126–​27, Hallisey, Charles, 26, 76n56, 81n71,
215, 246, 252n5 87–​88, 98, 100n123, 102n131, 204,
Diṅnāga,  22–​23 205n57,  216–​17
Dīpa, 187 hermeneutics, 2, 4n7, 16–​17, 25–​27,
Dīpavaṃsa, 37n10 26n64, 34, 96, 140, 220
Dispensation (sāsana), 11, 21, 23, 63, 66, Hinduism, 15
123, 177, 195, 214, 215 Holt, John, 194
docetism, 115, 136 Horner, I. B., 84n80, 188

Eckhart, Meister, 17, 53n61 Ibn ‘Arabi, 16, 17

eel-​wrigglers,  133 immeasurability of Buddha’s words
 Index 269

in the Abhidhamma, 49, 52–​57, 63, Lamotte, Étienne, 34n2, 39n17, 83–​84,
154–​63, 178–​79, 221,  239–​52 95n106, 96, 104, 187n3
Buddhaghosa on, 1–​30, language, see Māgadha; nirutti
38–​44,  218–​22 Law, B. C., 26, 64n16, 77n59,
in his meanings and phrasings, 91n97,  118–​19
33–​59,  73–​85 linguistic analysis, 39–​40, 61, 76n56,
in his teachings, 2, 8–​9, 14–​18, 21, 27, 77–​79, 230. See also byañjana;
33–​59, 68, 73–​85, 103–​7,  110–​11 nirutti; paṭisaṃbhidā
interpretable (neyyatha) statements, linguistic capacity, 78n63
7, 29, 60–​61, 64n14, 94–​97, lokiya, 51n57
102,  221–​22 lokuttara, 51n57
Isipatana, 158, 186–​87
itipiso contemplation, 83, 202–​5, 207, Madhupiṇḍika Sutta, 43
216–​17, 222,  229–​38 Madhyamaka, 147, 168–​69, 174
Itivuttaka, 36n8, 258 Māgadha, 1–​2, 77–​79, 78n65,
79n68, 83, 85
Jainism, 22, 56–​57, 57n73 Mahāgatigamiyatissa, 54–​55, 155, 241
Jains, 15, 44, 45n35, 57n73, 97n113, 99 Mahākaccāna, 43, 43n26, 64, 71
Jātakanidāna, 5n9, 45n33, 117–​18, Mahākammavibhaṅga Sutta, 64
157–​58,  216 Mahākassapa, 34–​36, 116–​17, 186, 214–​16
jātaka stories, 36n8, 50n49, 114, 117, Mahākoṭṭḥita, 64
159–​60,  246 Mahānidāna Sutta, 125n35, 125–​26, 130
jhāna, 101–​2, 156, 169, 204, 206, 227 Mahāniddesa, 47n47, 64–​65
Judaism,  16–​17 Mahāpaduma Thera, 211, 211n79
Mahāparinibbāna Sutta, 114n9,
Kabbalists,  16–​17 187–​88n3,  215n92
Kakusandha, 210 Mahāṭṭhakathā, 65
kammavācā requirements, 186–​88 Mahāvagga, 72, 186–​87
Kaṅkhāvitaraṇī, 189n8 Mahāvihāra (Great Monastery), 1–​2, 22,
Kassapa, 210, 245 27–​28, 34n2, 40n21, 67–​68, 196
Kathāsaritsāgara, 53n61 Mahāvīra, 15, 44, 44n29
Kathāvatthu (Issues for Discussion), 43, Mahāyāna Buddhism, 15, 78n63, 83–​84,
45n33, 88–​89, 92–​93, 146, 242 91–​92, 104–​5,  118n22
Khandhaka, 36, 72, 186–​88, 246 Majjhima Nikāya, 36, 64n16, 101,
Khuddaka Nikāya, 5n9, 36, 36n8, 37–​38, 136, 246
223, 246 Malliṣeṇa, 57n73
Khuddakapāṭha, 21, 36, 36n8, Many Types of Feeling Sutta
120n27, 246 (Bahuvedanīya Sutta), 98n114,
commentary on (Paramatthajotikā), 150–​51, 153, 170
5n10, 19, 21–​22, 65–​66 mātikā (matrix), 7, 43, 66, 71, 71n43, 73,
Konāgamana, 210 187, 193, 246
270 Index

meaning, 3, 4, 7, 26, 28, 61, 86, 88, 118, 153, Milindapañha, 36n8, 147n7
190, 198, 207, 247, 251. See also attha modal, 97n113, 98, 152, 163, 165–​66, 168,
expanding or developing, 12, 15, 16–​18, 174, 207–​8, 230
20, 43, 59, 62, 64, 66–​68, 70, 85, mode, see ākāra
200, 203, 220, 230, 247–​48 modernism, Buddhist, 24, 115, 221
surplus of, 16, 50n51 modularity, 15, 19–​20, 28, 38, 81, 81n71, 146,
meaning commentary (atthavaṇṇanā), 152, 154, 162–​65, 174, 203, 220, 222
65–​67, 119, 231 Moggaliputta Tissa, 43, 146, 217
meditation, 11–​13, 20, 23, 69, 172, 180n81, Moggallāna, 132–​33, 202, 208–​10, 217
182n83, 227, 237, 243 moral precepts, see sīla
the Buddha’s contemplations, 35, 40, Mūlapariyāya Sutta, 126, 136–​43, 162, 246
55–​56, 154–​55, 156n26, 158, 186,
239, 241 Nāgārjuna, 85, 85n82, 91n97
calming, 11, 12, 48n48, 224–​25, 227 Ñāṇamoli, Bhikkhu, 5n10, 10n18, 26n64,
(see also jhāna) 57n75, 66, 70n38, 76n56, 169n46,
contemplative exercises, 9, 20, 23, 55, 173n54, 230
68–​69, 83, 90, 92, 98, 102, 110, 113, Nance, Richard, 39n18, 70–​71, 77n61,
155, 160, 173n53, 180n81, 181–​83, 185, 110–​11,  118n22
196, 203–​207, 214, 216, 218, 220, 230 Nāṭyaśāstra, 53n61
insight, 11, 12, 224–​225 naya (methods), 6, 15n32, 25n62, 42, 52,
Recollections, 9, 20, 69, 80n70, 83, 53n62, 55–​58, 60n1, 61n3, 65, 69, 79,
113, 173, 185, 203–​204, 206, 216, 97n113, 146, 155, 157n28, 163, 175,
222–​23,  227 193, 240, 244, 246–​47
Methods (see also naya; pariyāya) Nettippakaraṇa (The Guide), 4–​5, 10n20,
Abhidhamma methods, 15, 42–​43, 46, 25–​26, 36n8, 37n10, 57, 69–​71,
49–​57, 63, 73, 93–​94, 94n104, 74–​75, 118, 223
101–​103, 106, 109, 146–​168, neyyattha (to be interpreted), 7, 29,
174–​182, 196, 239–​41, 244 60–​61, 64n14, 94–​97, 102, 221–​22
of the Buddha’s pedagogy, 7, 42, 57, nibbana, 11, 35, 85–​86, 99, 145, 201,
60–​61, 66, 68, 105–​106, 138, 155, 224–​27, 238, 246
164, 175–​77, 207, 225 nidānas (introduction, origin, context)
interpretative, 3, 6, 10n20, 15, 23, canonical, 11, 29, 105, 110, 112–​23,
25–​28, 56–​58, 61, 64–​79, 83, 96, 131–​32, 140, 185, 220, 224
103–​106, 113–​14, 141, 164, 173, 182, commentarial, 5–​6, 19, 115, 117, 123, 126,
203, 213, 222, 234 131–​35, 156–​61, 185–​86, 197–​202,
“ocean of,” 6n11, 15, 25n62, 48–​57, 205,  213–​20
154–​63, 196, 239, 241 definition of, 110, 116–​23, 135, 141,
piṭakas as, 6, 29, 38, 49–​58, 61, 63, 143, 193n22
101–​103, 113–​16, 155, 177, 193, 203, introduction to each piṭaka, 5–​6, 19,
220–​22, 235–​37,  239–​52 113, 117, 119, 185–​86, 197, 200, 216,
mētis (practical knowledge), 104, 105, 221 220, 239
 Index 271

to the Abhidhamma, 156–​61, 216 omniscience, Buddha’s, 14–​18, 25, 103–​10,

to the Brahmajāla Sutta, 124, 121, 122–​24, 126–​31, 138, 139–​41,
126,  130–​36 143, 149, 154–​55, 157, 162–​63,
to the Mahānidāna Sutta, 125, 130 179–​81, 195, 205–​206, 208, 239–​52
to the Mūlapariyāya Sutta, 136–​43 features of, 48, 133–​36
to the Vinaya, 193n22, 196–​97, knowledge of, 33–​59
202, 205, 213, 216, 219 (see also teachings on, 217–​19, 221
bahirnidāna (External Nidāna) unfolding quality, 2, 49, 109, 113,
niddesa (expository commentary or 183–​85, 217–​19,  221
description), 4, 7, 36n8, 47n47, ontology, 94, 147–​49, 168–​74, 173n54
64–​65, 71, 74n51, 76n56 ordinary persons, 79, 137–​41
Niddesa, 36n8, 47n47, 246 ovādapātimokkha (Pātimokkha of
nikāyas, 18n46, 19, 34–​36, 37n9, 41, Exhortation), 212
86n84, 119, 124, 246
nippariyāya (categorical, abstract) padavaṇṇanā (word commentary),
teachings, 7, 29, 61, 97–​106, 110, 65–​66,  84
112–​13, 156, 162, 175, 181, 207–​8, Padesavihārasutta, 156, 156n26
216,  221–​22 Pali language, 1–​2, 22, 27–​28, 77–​79,
nirukta analysis, 79n68 223–​24. See also Māgadha
nirukti—​pratisaṃvid, 78n63 Pañcakaṅga,  150–​52
nirutti (language), 39–​40, 61, 62n4, Pañcapakaraṇa-​aṭṭhakatha, 145n5
75,  77–​82 pañhākamma (set of questions), 66, 72
nirutti analysis, 61, 62n4, 75, 78–​85, pañhāpucchaka (Asking Questions),
83n74, 190n13, 193, 203, 214, 225, 174–​75,  177–​78
230, 234, 234n1, 240 paññā, 11–​12, 69, 72, 144, 168, 181–​82,
nītattha (definitive) language, 7, 29, 222, 237
60–​61, 94–​97, 99n120,  221–​22 paramārthasatya, 92
Noble Truths, four, 38, 74, 75, 86, 100, paramattha (furthest sense) language
149, 153, 175, 179n76 or teaching, 7, 29, 60, 63n11,
Nyanaponika Thera, 94n104, 147–​49, 85–​96, 102, 173–​74, 177, 180–​81,
167, 182 221–​22,  249
Paramatthajotikā, 5n10, 19, 21–​22, 65–​66
occasion, see samaya Parivāra, 36, 72, 186–​87, 201, 246
oceans, 15–​17, 15n32, 28, 42, 146, pariyatti (learning), 6, 49–​52, 58, 61,
177, 216 77n61, 129n46, 137n72, 155–​56, 192,
in Abhidhamma, 73, 154–​63, 196, 224, 248, 251
221,  239–​52 pariyāya (contextual, qualified, figurative)
of knowledge, 15n32, 49, 53n61, teachings, 7, 29, 61, 97–​106, 110,
56, 56n71 112–​13, 175, 181, 207–​8, 216, 221–​22
of methods, 6n11, 15, 15n32, 25n62, as modes of instruction, 50n49, 67,
48–​57, 154–​63, 196, 239, 241 97n113, 124, 151–​52, 206–​7
272 Index

Patañjali, 10n20 Asking Questions (pañhāpucchaka),

Path of Purification, see Visuddhimagga 174–​75,  177–​78
Pātimokkha, 36, 61, 71, 186, 188n4, 189, pañhākamma (set of questions),
193, 210–​12, 246 66, 72
paṭisambhidā (analysis), 39–​40, 42, ways of answering questions,
51–​52, 58, 64–​65, 76–​79, 225, 230, 68, 68n32
244, 250, 251 Quran, 16
Paṭisambhidāmagga (The Path of
Analysis), 36n8, 39n18, 45–​49, 246 Ramanujan, A. K., 112n5, 141
Paṭṭhāna (Starting Points), 54–​56, Ram-​Prasad, C., 57n73, 153n15, 180n81
73, 146, 148, 154, 155, 174–​80, Ratnavinīta Sutta, 145n4
182n85,  239–​52 Recollection of the Buddha, 69, 80, 83,
Pe Maung Tin, 102n130, 157n29, 169n46 203–​5,  222–​23
Peṭakopadesa (Instruction in the Piṭaka), Recollection of the Dhamma, 9–​14, 69,
4–​5, 25–​26, 36n8, 37n10, 57, 69–​70 206, 222, 223–​28
Petavatthu, 36n8, 246 Recollection of the Sangha, 223
phenomenology, 23, 94n104, 144–​83, 221 Ricoeur, Paul, 6, 16, 50
philology, 3–​7, 25, 27, 33, 218, 220 Ronkin, Noa, 39n18, 94n104, 147n9,
philosophy, 22–​23, 111, 114, 124, 141–​42, 171–​74, 173nn53–​54,  174n58
146, 160, 220
phrasing, see byañjana sabbaññutañāṇa, see omniscience
piṭakas. See also Abhidhamma; genre; sabhāva, 91n98, 168–​74, 248, 250
pariyatti; Suttanta; Vinaya Saccaka, 99, 103
definitions of, 6, 15n32, 25n62, 35–​38, Sāmaññaphala Sutta, 117, 133n61
49–​58, 50n49, 60, 61, 63–​64, 89, Samantapāsādikā (Completely Pleasing),
103, 126, 149, 153n15, 155–​56, 162, 5, 20, 29, 50, 61–​67, 83, 116–​18,
185, 188, 192–​93, 220, 239–​52 184–​217, 221–​23,  252n5
introductions (nidānas) to, 5–​6, authorship, 19, 189n8, 213
19, 24, 28–​29, 49, 53–​54, 61, 103, External Nidāna
113–​14, 117–​18, 156–​57, 200, 216, (bahirnidāna),  196–​97
220–​21,  239–​52 translations, 5n10, 34n2, 196n33
Pollock, Sheldon, 3–​4, 218 samaya (occasion), 148, 184, 192,
porāṇas, see ancient commentators 197–​201, 214, 219, 229–​32, 234, 241
Pramāṇa-​samuccaya (Diṅnāga),  22–​23 Sammohavinodanī (Dispeller of
Prasad, C. S., 68 Delusion), 39, 65, 67, 77–​79, 86n83,
Pubbācariyas, see ancient commentators 100, 154, 175–​77
Puggalapaññatti (Describing Persons), 93, sammuti (conventional language or
146, 173n55, 242 teaching), 6, 7, 29, 60, 70, 85–​96,
Puggalavādins,  92–​93 102, 177, 221–​22
saṃsāra, 13, 15n32, 51n57, 52–​53, 81–​2, 128,
questioning, 65, 146, 182, 187, 251 130n48, 142, 169n46, 182, 192, 226,
analytical, 66, 72 230, 235–​38, 240–​41, 246, 251
 Index 273

saṃvṛti, 85n82, 91–​92n99 Sumedha,  157–​58

Saṃyutta Nikāya, 36, 246 Suppiya, 127, 131–​33, 136, 142, 161, 216
Sandaka Sutta, 44n32 Suttanipāta, 36n8, 65, 246
Saṅgīti Sutta,  64–​65 Suttanta Analysis (suttantabhājanīya),
Sañjaya, 132–​33, 133n61 174–​77,  204
Sanskrit, 75, 78, 78n65, 91–​92n99, Suttanta Piṭaka. See also pariyatti;
120n27, 168, 247n3 piṭakas; nidāna
Sāriputta, 23–​24, 42–​43, 56, 64, 84, 99, as conventional, contextual, dialogical
132–​33, 155, 161, 197–​99, 202, teachings, 23, 25, 29, 63, 89–​94, 101,
208–​17, 231, 232, 234, 239, 244, 245 103, 105–​6, 109–​43, 162, 189–​201,
Sarvāstivāda Abhidharma, 105n135, 208, 216, 220–​21
168–​69,  168n44 definitions of, 5–​7, 35–​38, 49, 50–​54,
Sarvāstivādins, 68, 91n98, 105n135 62–​64, 192n18, 203–​4, 229–​30,
sāsana (Dispensation), 11, 66 232–​35,  246–​50
śāstras, 10n20, 23 Suttavibhaṅga (commentary on the
Sayadaw, Ledi, 24, 91n98 rule), 64, 71–​72, 116n14, 186–​90,
Schlingloff, Dieter, 188, 188n5 197, 199
scholasticism, 26, 27n69, 58–​59, 169
Scholem, Gershom, 16–​17 Tamil language, 77, 88
scripture, see buddhavacana Tathāgata, 8, 9, 14, 38, 82, 94, 122, 127–​30,
(Buddha’s word) 132n55, 136, 137, 160, 161, 175, 192,
Second Council at Vesālī, 186–​88 207, 211, 234, 237, 243, 244
Sermon on the Noble Quest Tāvatiṃsa heaven, 40, 156–​57, 159, 160,
(Ariyapariyesanasutta), 114 162, 239, 243, 245
“Short Exposition on Kamma,” 64n16 techne (theoretical knowledge), 104, 221
Shusterman, Richard, 142 Theragāthā, 36n8, 37, 114, 246
Sīhala-​aṭṭhakathā, 1, 22. See also ancient Theravada Buddhism, 17, 23–​24, 27–​28
commentators Therīgāthā, 36n8, 114
sīla (morality, moral precepts), 11, 18n45, Tissa, Moggaliputta, 43, 146, 155
50, 69, 72, 81n71, 100, 132n55, 144, Tissabhūti,  156–​57
145n4, 177, 178n69, 222, 223, 249 Torah,  16–​17
Śiva, 15 trees, 1, 35, 40, 54, 90, 136, 157–​60, 197,
Stream-​Enterers, 125, 212 206, 217, 229, 230–​35, 239, 241–​43,
Strong, John, 24 245–​46. See also Bodhi Tree
Sudinna, 185, 191, 201–​2, 213, 215, 231
Sumanadeva,  156–​57 Udāna, 35n6, 36n8, 200n44, 246
Sumaṅgalavilāsiṇī, 5–​7, 10, 18n46, 20, 22, Udāyī,  150–​52
34n2, 35–​37, 37n10, 50n49, 51–​52, uddesa (outline), 7, 43, 65, 71, 193, 246
59n77, 60–​65, 86, 89, 91n97, 97, 117, U Nārada, 178
119, 125–​35, 153, 166, 192–​93, 195, Upāli, 34, 110, 116–​17, 159, 184, 186, 188,
200–​1, 204, 214, 239, 252 201–​2, 215, 233, 234
274 Index

Upaniṣads, 23 vinicchaya (exposition), 66

Uposatha ritual, 10n19, 96, 186, 189, 212 Vipassī, 210, 212, 212n82
Uttarakuru, 209, 244 Viṣṇu, 15
Visuddhimagga (Path of Purification),
Vakkali, 18n44 6, 9, 10n18, 18n45, 23–​24, 26,
vaṇṇanā,  65–​66 69, 76, 89, 149, 152, 163n35, 164,
Vasubandhu, 70–​71, 85n82, 96n108, 103, 168, 173, 176n66, 182, 206, 222,
147n7, 168n44 223–​28,  230
vedanā (feeling), 127–​29, 131, 142, 150–​52, commentaries on, 5n9, 46n40, 173n54
156, 156n26, 161–​64, 166, 169–​71, production of, 1–​2, 18, 18n47,
198, 236–​37, 248 22,  66–​67
Verañjā story, 20, 185, 197, 198, 201–​202, as a reference tool, 28, 28n70
203n51, 206–​209, 213, 216–​17, relation to aṭṭḥakathā, 19–​20, 19n48,
223,  229–​38 70, 72, 80, 80n70, 81n71, 83, 203–​4,
Vesālī, 188, 201–​2, 213 222, 223, 230
Vessabhū, 210, 212 structure of, 11, 66, 72, 81n71, 144–​45,
vibhajjavāda analysis, 64–​71, 174 145n3, 145n4, 154, 177, 181
Vibhajjavādins (Analysts), 67–​70, 93, 176 vitthārena (detailed words), 7, 29, 42–​43,
Vibhaṅga (Analysis), 4, 7, 36, 39n18, 61, 51, 60, 64, 68–​73, 71, 73, 84, 143,
64–​65, 71, 100n125, 146, 154, 175, 221, 250
174–​78, 186, 189, 190, 193, 242, 246 vohāra (colloquial talk), 6n12, 7n13, 63,
Vimānavatthu, 36n8, 246 89, 91n97, 94, 105–​6, 249
Vimuttimagga, 18–​19n47 Vyākhyāyukti,  70–​71
Vinaya Piṭaka, 182–​217. See also nidāna; vyañjanā, 75
piṭakas
Buddhaghosa’s commentary, 229–​38 Walters, Jonathan, 25–​26, 113–​16,
(see also Samantapāsādikā) 123, 136
definitions of, 6–​7, 7n14, 35–​37, 49–​ Wheel of the Dhamma, 158
52, 54, 61, 63–​64, 66n25, 71–​72, 89, word commentary (padavaṇṇanā),
97, 110, 221, 246–​52 65–​66,  84
experts in, 190–​91, 196, 203–​204,
211n79, 215, 230, 235, 240 (see also Yamaka (Pairs), 146, 242
expertise; pariyatti) Yasa,  186–​87