Journal of the International AssQciation of

Buddhist Studies
Volume 20 Number 1 Summer 1997
PAUL L. SWANSON
What's Going on Here?
Chih-i's Use (and Abuse) of Scripture 1
GEORGES DREYFUS
Tibetan Scholastic Education and The Role of Soteriology 31
ANDREW HUXLEY
Studying Theravada Legal Literature
PETER SKILLING
The Advent of Theravada Buddhism to Mainland
South-east Asia
ELI FRANCO
Distortion as a Price for Comprehensibility?
The rGyal tshab-J ackson Interpretation of DharmakIrti
ROGER R. JACKSON
"The Whole Secret Lies in Arbitrariness":
A Reply to Eli Franco
63
93
109
l33
e watermark
The Journal of the International
Association of Buddhist Studies
(ISSN 01-93-600XX) is the organ of
the International Association of
Buddhist Studies, Inc. It welcomes
scholarly contributions pertaining to
all facets of Buddhist Studies.
JIABS is published twice yearly, in
the summer and winter.
Address manuscripts (two copies)
and books for review to The
Editors, JIABS, Section de
langues et civilisations orientales,
Universite de Lausa,nne, BFSH 2,
CH-1015 Lausanne, Switzerland.
Address subscription orders and
dues, changes of address, and
business correspondence
(including advertising orders) to
Professor Joe B. Wilson, Treasurer
IABS, Department of Philosophy
and Religion, University of North
Carolina at Wilmington,
Wilmington NC 28403 USA
email: wilsonj @uncwil.edu
Fax: 910-962-7107
Subscriptions to JIABS are $40
per year for individuals and $70
per year libraries and
institutions. For informatIOn on
membership in lABS, see back
cover.
1997 by the International
Association of Buddhist Studies,
Inc. The editor gratefully
acknowledges the support of the
Center for Chinese Studies and the
Department of Asian Languages
and Cultures of the University of
Michigan.
EDITORIAL BOARD
Donald S. Lopez, Jr.
Editor-in-Chief
Robert Buswell
Steven Collins
Collett Cox
Luis O. Gomez
Oskar von Hinfiber
Roger Jackson
Padmanabh S. Jaini
Shoryu Katsura

D. Seyfort Ruegg
Ernst Steinkellner
Erik ZUrcher
Editorial Assistant
Alexander Vesey
ELI FRANCO
A Short Response to Roger Jackson's Reply
TOM J. F. TILLEMANS
On a Recent Translation of the San:zdhinirmocanasutra
JOHN R. MCRAE
Review: Zenbase CD1
JOE BRANSFORD WILSON
The International Association of Buddhist Studies
and the World Wide Web
149
153
165
175
Contributors to this issue:
GEORGES,B. DREYFUS is an Associate Professor of Religion at
Williams College. His research on Tibetan monastic education exam-
ines the ways in which intellectual practices such as commentary,
debate and memorization inform communities and contribute to the
formation of their members. His publications include Recognizing
Reality: Dharmakfrti's Philosophy and Its Tibetan Interpretations
(SUNY, 1997).
ELI FRANCO is a member of the Institute for the Culture and History
of India and Tibet, University of Hamburg. Publications include Per-
ception, Knowledge, and Disbelief" A Study of JayaraHs Skepticism
(Wiesbaden: 1987; 2nd. ed. Delhi: 1994), and Dharmakrrti on Com-
passion and Rebirth (Vienna: 1997).
ANDREW HUXLEY, MA, BCL, Barrister-at-Law, has taught law in
the University of Oxford, the University of the West Indies and the
University of Hong Kong. He spent some years practicing in the
criminal courts of London. His interest in Buddhism and Southeast
Asia began in 1984 when he took up his present post at the School of
Oriental and African Studies, London.
ROGER R. JACKSON is Professor of Religion at Carleton College,
Northfield, MN. He is presently completing two books on Mahamudra
in the dGelugs and other traditions, and co-editing a volume of essays
on Buddhist theology.
JOHN R. MCRAE teaches in the Department of Religion at Indiana
University. He is author of The Northern School and the Formation of
Early Ch' an.
PETER SKILLING is Curator of Pali manuscripts for the Pali Text
Society in Bangkok, and is responsible for the Fragile Palm leaves
manuscript preservation project. He is at present working on a three-
volume comparative study of the Mulasarvastivadin Mahasutras, vol.
44 (PTS, Oxford).
PAUL L. SWANSON is a permanent fellow of the Nanzan Institute
for Religion and Culture and Professor at Nanzan University in
Nagoya, Japan. He is the author of Foundations of Tien-fai Philos-
ophy: The Flowering of the Two Truths Theory in Chinese Buddhism
(Asian Humanities Press, 1989), and The Collected Teachings of the
Tendai Lotus School (BDK English Tripitaka 97-ll, 1995).
TOM J. F. TILLEMANS is Professor of Buddhist Studies at the Uni-
versity of Lausanne in Switzerland. His research centers on Indian and
Tibetan Buddhism, in particular the Madhyamaka and logic and
epistemology.
JOE BRANSFORD WILSON is Associate Professor of Philosophy
and Religion at the University of North Carolina at Wilmington. A
Tibetologist, he maintains a number of World Wide Web sites relevant
to Buddhist and Asian Studies and is a co-editor of the World Wide
Web Virtual Library materials on Buddhist Studies.
PAUL L. SWANSON
What's Going On Here?
Chih-i's Use (and Abuse) of Scripture
In preparing heavily annotated English translations of major portions
of the Fa-hua hsuan-i and Mo-ho chih-kuan T'ien-
t'ai Chih-i's (538-597) two most doctrinally influential works, one of
my main concerns has been to identify the original sources of scrip-
tural quotations and put them into their proper context. It seems to
me an important part of translation and interpretation that quota-
tions not be tendered autonomously and out of context-the trans-
lation should not only reflect, but also in some way incorporate, the
context from which a quoted passage comes. This involves the ques-
tion of what the "original" source says in the first place (as far as it
can be determined), and whether or not the source has been quoted
accurately or has been restated, paraphrased, or reinterpreted.
In the course of tracking down and comparing the sources that
Chih-i quotes as scriptural authority, I have been struck by passages
that seem to misrepresent their source. This raises a number of
questions. Which texts does Chih-i rely on most frequently, and
why? When these texts seem to be "misquoted," can this be attrib-
uted to deliberate misrepresentation? If so, what is the significance
of the passage, and what does this have to say about Chih-i's ideas
and his creative genius? How far ought we apply the scholarly
assumptions and requirements of our age to the scholarship of days
gone by?
In this paper I will take a close look at selected passages in the Fa-
hua hsuan-i and the Mo-ho chih-kuan, comparing closely the quotes in
Chih-i's texts with the "original" texts as we have them in the Taisho
canon. 1 I will then make some general comments concerning Chih-i' s
1. Various problems connected with considering the texts as found in the
Taish6 edition of the Buddhist canon as "original" are discussed below.
1
2 JIABS 20.1
use of scripture.and discuss the significance of these conclusions.
In addition to the comments above-which deal with Chih-i's
method of "translating" or interpreting the texts he uses; and the
attempt to translate Chih-i's texts faithfully into a modern context-
I would like to introduce (as an Appendix) yet another perspective
to the issue, namely, the problems encountered in preparing essays
about Chih-i's work in both Japanese and English. I have given a
number of papers in Japanese on the topic of Chih-i's use of scrip-
ture, and am now preparing the "same" paper in English. I have dis-
covered, however, that it is impossible to write the same paper, in
both languages-that I cannot simply "translate" my Japanese paper.
Even I myself cannot say the "same" thing I want to say, in two dif-
ferent languages; I suspect that this is true for anyone working in
more than one language (or even classical and modern versions of
the same language), especially if the languages are not of the same
linguistic family. Reflection on the reasons for this intriguing
phenomenon has led me to delineate a number of factors that con-
tribute to this situation, and which seem to work on overlapping lev-
els: differences in individual words and terms and their nuances that
steer one's train of thought in different directions; differences in the
ideas, cultural background, and other ''baggage'' that is tied up with
the language; and the differences in intended audience that
This question is addressed by Paul Harrison with regard to the Pratyut-
panna-samiidhi-sutra, but his comments are applicable to Buddhist texts in
general: "When we refer to an 'original Sanskrit text,' we must realise
from the outset that we are adopting a convention, and a potentially mis-
leading o n ~ , a t that. For there is, or was, no such thing as a single.origi-
nal Sanskrit text of the Pratyutpanna-samiidhi-si1tra, compiled around the
beginning of the Common Era andremaining unchanged while various
translations, Chinese and Tibetan, were made from it. We know that in
general Mahayana sutras underwent some degree of change in the course
of the many centuries during which they were in use, being amplified (pos-
sibly the most common pattern), shortened, re-arranged, or subject to
the introduction or modification of various doctrinal terms. The surviving
translations of the Pratyutpanna-samiidhi-sutra exhibit this 'textual fluidity'
to a marked degree .. , . We must therefore realise that when we speak of
'the original' of the Pratyutpanna-samiidhi-si1tra, we are in effect referring to
its ever-changing Sanskrit textual tradition, and not to any single entity-
a river, rather than a lake" (Harrison 1990, xxxiv).
SWANSON 3
influence the direction, flow, and content of one's prose and expres-
sion. I have also come to the conclusion that these factors are equally
important for the understanding and translation of classical texts,
and provide aD. important argument against the idea that a strict, lit-
eral translation best represents the original text. It also provides
important implications for our own understanding of Chil;1-i's work,
as well as his use of scriptural texts and authority.
Detailed Analysis of Chih-i's Use of Scripture
My initial methodology in comparing Chih-i's text with the sources
of his quotes was to physically "disassemble" the text by indulging in
a bit of old-fashioned cutting-and-pasting. This involved snipping
out sections of Chih-i's text and pasting it in horizontal columns,
then cutting out the quoted original sources and pasting them in
columns under the corresponding sections of Chih-i's quotes. This
process not only forced a detailed and careful comparison of each
phrase, but also provided an overview of the flow of the text, gave a
visual sense of the proportion of quotation in the text, and supplied
the wider context of the quoted sources for further analysis.
The following is an analysis of some representative passages,2
mainly from the Mo-ho chih-kuan and Fa-hua hsuan-i, to illustrate
Chih-i's use of scripture. .
1. On whether or not the Buddha-dharma should be taught; from
Kuan-ting's introduction
3
to the Mo-ho chih-kuan (T. 46.3all-b2).
This section deals with a question raised by a "skeptic": how can one
2. By "representative passages" I mean that almost any other section of
Chih-i's works, especially the Mo-ho chih-kuan and Fa-hua hsiian-i, could be
examined with similar results, and that the passages are not chosen arbi-
trarily to prove a predetermined point. I have deliberately avoided pas-
sages from the Fa-hua wen-chii (T. no. 1715), since it appears that Kuan-
ting imported much of Chi-tsang's work into the text during the editorial
process; see Hirai 1985). I cannot claim to have read the entire Chih-i cor-
pus, but my limited exposure to his earlier works, such as the Tz 'u-ti ch 'an-
men (T. no. 1916), Hsiao chih-kuan (T. no. 1915), and other ritual manu-
als, as well as his later commentaries on the Vimalak1.rti-siitra, leads me to
believe that my conclusions are applicable to them as well.
3. Since it is practically impossible to distinguish which parts of Chih-i's
"4 JIABS 20.1
teach or preach the Buddha-dharma when it is inconceivable and
beyond verbalization, and should we try? The skeptic quotes or uses
the words of the sutras to point out that the Dharma is "impossible
to explain in words" (Lotus Sutra, T 9.10a4), "not possible to verbalize"
(Mahiiparinirvii'Yfa-sutra, T 12. 733c9-20), and "cannot be expound-
ed or signified" (Vimalakirti-sutra, T 14.548a8-11 or 540a17-21).
Kuan-ting replies by quoting twelve times from the same sutras-
often the same passages-as well as other texts (Mulamadhyamaka-
kiirikii, * Pravara-devariija-parip!cchii-
sutra), to argue that the Buddha-dharma must be taught. The fol-
lowing points are of special interest. .
1. The same sutra passages quoted by the "skeptic" are followed up
by Kuan-ting to show that the passage in fact supports the idea
that the Buddha-dharma should be taught.
a. The skeptic opens his question with a phrase from the "Chapter
on Skillful Means" of the Lotus Sutra, though it is not identified
as a scriptural quote. Kuan-ting's third quote is from the line
that follows immediately after the line quoted by the skeptic,
and points out that the Buddha "expounds [the Dharma] by
resorting to the power of expedient means" (T 9.10a5).
b. Kuan-ting's first quote is a line from the Mahiiparinirvii'Yfa-sutra
that concludes the passage quoted by the skeptic: that "since
there .ire causes and conditions, it is also possible to expound
[them, or the Dharma?]" (T 12.733c19-20).
lectures edited by Kuan-ting (such as the Mo-ho chih-kuan and Fa-hua
hsuan-i) are attributable to whom, I treat Chih-i and Kuan-ting as a single
"person," or at least as speaking with a single voice. Although I would like
to know what Chih-i himself said (his "original discourse"), it must be
recognized that what we have as "the work of Chih-i" is a composite of
many layers-the resultant texts are records mostly (but not all) based on
lectures by Chih-i, recorded and edited by Kuan-ting, and accepted and
commented on by a variety ofT'ien-t'ai / Tendai scholars (not limited to
the monks of the T'ien-t'ai / Tendai school). To be accurate, my essay
should thus be entitled "Tentative observations on the use of 'scripture'
(as far as this term can be used in the Buddhist context) by Chih-i in
works edited and modified by Kuan-ting and handed down through the
T'ien-t'ai / Tendai tradition, as far as we can tell from textual variants as
compiled and published mainly in the Taisho Tripitaka," but my prefer-
ence for simplicity led me to retain a shorter title.
SWANSON 5
c. The skeptic's references to the Vimalakirti-sutra are reflected in
Kuan-ting's seventh quote from the same sutra, that ''liberation
is [not]4 separate from the nature of words and letters" (T.
(see the discussion of this section below).
2. Some problems:
a. Kuan-ting quotes the parable of explaining the color of milk to
a blind person in the Mahiiparinirvii!TJa-sutra (T. 12. 688c15-24)
to illustrate that the real truth (paramiirtha-satya) can be
expounded. In the sutra, however, the attempt to explain "the
whiteness of milk" to a blind person by referring to a shell
(sound), rice powder (softness), snow (cold), and a white
crane, ultimately fails, because a blind person cannot "see"
"white." The sutra concludes that "ultimately he is unable to
know the true color of milk," and that this is the same as trying
to explain that nirvana is eternal, blissful, selfhood, and pure to
a non-Buddhist:9i-Ji -such people will never understand. Thus,
though the parable does illustrate a concerted attempt to
explain with words what must be directly experienced, it is a far
cry from proving that "the real truth should be expounded."
b. The Shing t'ien-wang po-jo piJ-lo-mi ching (* Pravara-deuariija-
pariprcchii-sutra) is quoted as saying that "Although [the essence
of] a dharru;ti is [ultimately] wordless, yet words do manifest
[the meaning of] dharru;ti" (T. 8.720c5-6); and Kuan-ting claims
that this illustrates that the mundane truth (sarrtvrti-satya) can
be expounded. However, the two lines that follow in the sutra
immediately after this line say the opposite: "the great wisdom-
power of prajiiii is separate / independent from words and ver-
balization. "
c. Kuan-ting's reply to the skeptic's quotes from the Vimalakirti-
sutra could be attributed to at least two passages in the sutra.
Traditionally it has been attributed to a passage in the third
4. As Stevenson and Donner point out, "the MHCK text is apparently
corrupt here, omitting a crucial negation that stands in the Taisho text of
the Vimalak1.rti, so that the MHCK text should read here 'Not being sepa-
rate ... .' An interlinear handwritten note in the woodblock of the ehon
in L. Hurvitz's possession makes the emendation." See Donner 1976, 82,
and Donner and Stevenson 1993, 125-26.
6 JIABS 20.1
chapter, T 14.S40c18-20, near the skeptic's second quote.
Certainly the characters quoted in the Mo-ho chih-kuan are
almost identical to those found in this sutra passage, except that
it jumps over a crucial phrase of four characters 1Wi:1fJt', and
this passage in effect emphasizes the independence ofliberation
from words and letters. The passage (T 14.S48a11-1S) immedi-
ately after the skeptic's first quote from the seventh chapter of
the Vimalakirti-sutra, on the other hand, uses different charac-
ters to make the same point ("speech, words, and letters are all
marks of liberation"), and then supports the point being
by the Mo-ho chih-kuan by saying "one cannot expound libera-
tion apart from words and letters." Thus, although the charac-
ters in the phrase from the third chapter of the Vimalakirti-sutra
are closer to the phrase quoted in the Mo-ho chih-kuan, the
intent of the phrase-that one must use words to expound the
meaning of liberation-is closer to the passage in the seventh
chapter. It is difficult to conclude decisively which section Kuan-
ting had in mind-perhaps both.
3. Miscellaneous minor points:
a. In the quote froJIl the (T
lS.S0c1S-16), the character for "should" is given as "con-
stantly" -m- in the Mo-ho chih-kuan. This is certainly a copyist's
error due to visual similarity, and does not result in any
significant shift in meaning.
b. Kuan-ting's'tenth quote is a paraphrase from the Lotus Sutra (T
9.20alS-20). The three phrases ("going, coming, sitting, or
standing," "constantly proclaiming the wonderful Dharma,"
and the falling of rain") are picked up in a different order
from that in the original sutra, but without a significant change
in meaning. This is a relatively unusual example of a paraphrase
rather than direct quote of the Lotus Sutra.
c. The next quote (no. 11) starts out by saying that the Lotus Sutra
"also says," but doesn't start quoting from the Lotus Sutra until
the next phrase.
5
5. Donner and Stevenson (1993, 126) describe the early part as a para-
phrase from the same passage as the quote that follows, but, as they
admit, "the match is a dubious one."
SWANSON 7
d. Otherwise the quotations from the sutras are generally accu-
rate.
II. The Constantly-Sitting Samadhi . and the Maiijusri sutras (T
46.11a-12a)
Chih-i's exposition of the Constantly-Sitting Samadhi is based for the
most part on two suu-as, the Saptasatikii-prajiiiipiiramitii-sutra (The
"Great Prajiiaparamita-sutra Taught by Maiijusri"; T no. 233) and
the * Maiijusii-pariprcchii ("Questions of Maiijusri"; T no. 468).6 The
content is almost entirely from these sutras, even sections that are
not specifically identified as quotes from scripture. This is not par-
ticularly surprising, since Chih-i declares that his presentation is
derived from these sutras. It is important to point out here that the
Mo-ho chih-kuan text is often abbreviated and cryptic, and referring
to the original sutras provides the additional information missing or
needed to make sense of the Mo-ho chih-kuan version.
One section that does not specifically identify itself as quoting
scripture but is in fact almost verbatim from the Teachings of Maiijusii
(T 8.728b-c) is Chih-i's exposition on "the contemplation of karma"
(T 46.11c25-12al). Here all ofChih-i's phrases are from the sutra,
but he skips sentences and jumps from one section to another with
no indication that he is using the words of the sutra. One is led to
wonder whether his audience was already familiar with the content
of the sutra and was able to fill in the details themselves and thus
Chih-i took this for granted, or if Chih-i was consciously selective of
only phrases that fit his agenda. In this case I suspect the former is
more likely, and therefore in order to adequately grasp the content
of the Mo-ho chih-kuan we must familiarize ourselves with the full con-
tent of the sutras on which these passages are based.
III. The Constantly-Walking Samadhi and the Pratyutpanna-samiidhi-
sutra
Much the same (as in section II) can be said for Chih-i's exposition
of the Constantly-Walking Samadhi (T 46.12a19-13a23). Some of
the highlights are outlined below.
6. For an excellent summary of the four samidhis see Dan Stevenson's
essay on "The four kinds of samidhi in early T'ien-t'ai Buddhism" in
Gregory 1986, 45-97.
8 JIABS 20.1
L All of the quotes in this section are either from the Pratyutpanna-
(T no. 418, esp. 13.904-909)-the classi-
cal source for the "pratyutpanna samadhi" or "the samad.bi where-
in one finds oneself standing face to face with all the Buddhas of
the present age"-or from the -t1
(1: no. 1511, 13.20-122), in particular section 25 of this
text (T 26.86a-90a), which consists of a commentary on the
pratyutpanna samadhi or the nienjo samadhi. On the one
hand this is rather unusual for Chih-i, who liked to salt and pep-
per his text with quotes from the Lotus Sutra, Mahaparinirva1Ja-
sutra, and Ta chih tu lun. On the other hand it is not all that sur-
prising, since the section is an analysis of a specific topic, the
Constantly-Walking Samadhi, and the Pratyutpanna-samadhi-sutra
and the commentary in the are the pri-
mary sources for this practice.
2. Rather than quoting whole chunks of sutra or sastra passages,
Chih-i tends to summarize or select certain phrases to illustrate
the explanation in the source. Also, rather than outline the con-
tent of the source in its original order, Chih-i tends to jump
around, returning frequently to key passages.
For example, in his opening paragraph (12a21-22), Chih-i gives
as an alternate to "pratyutpanna-samadhi" the "Buddhas
standing [in front of one]" 1LrL, a term found frequently in the
Pratyutpanna-samadhi-sutra (e. g. 903cl3, 904b22, 905a4-5), and
then jumps ahead to a passage at 905cl6-18 to provide three char-
acteristics to show how the Buddhas "stand" in front of one as a
result of one's visualization.
Again, in the section where Chih-i explains the mental (after
discussing the "physical" and the "verbal" D) aspects of this
samadhi (12b24-c28), he relies on a single long passage from the
Pratyutpanna-samadhi-sutra (T. 13.905al-906a11), but jumps from
one part of the section to another, not necessarily in order. For
example, he mentions the idea of being mindful of the Buddha's
thirty-two major marks early in his discussion (12b27), but this
does not appear in the Pratyutpanna-samadhi-sutra until halfway
through the passage (905b16), and many passages that Chih-i
quotes later actually appear earlier in this passage. In all, the
phrases that Chih-i picks up from this passage come to only about
SWANSON 9
10% 'ofits original length in the sutra, and Chih-i's summary must
be filled in somewhat, either with footnotes or additions to the
text, to gain an accurate and full account of what is involved in this
. practice.
There many places where Chih-i refers to whole analogies or
sections of a text with short cryptic phrases that do not make any
sense until one is familiar with the context or the longer version
that is given in the original source. For example, Chih-i's analysis
of the content of this samadhi in terms of "conventional exis-
tence'; f.oc makes reference to six analogies found in a section of
the Pratyuipanna-samiidhi-sutra The first analogy is uti-
lized as follows: "It is just as in a dream one sees the seven jewels
and one's relatives, and rejoices; after awakening one tries to
remember, but does not know where this happened. Be mindful
of the Buddha in this way" (12c8-10). Chih-i'sabbreviatedversion
makes it difficult to understand what the analogy has to do with
being mindful of the Buddha, but the full explanation of the
analogy in the (905a-c) explains how
concentrating one's thoughts on the Buddha is analogous to
dreaming and thus "seeing" treasures, friends, etc. In translating
this passage, either a full explanation must be given in a note, or
enough material must be incorporated into the translation to
make sense of the passage.
7
Again, Chih-i closes this entire section with the following exhor-
tation: ''If one does not cultivate such a method [of samadhi], he
loses immeasurable, valuable treasures, and both people and gods
will grieve. It is as if a person with a stuffy nose sniffed sandalwood
and could not smell it, or like a rustic man who offers [ only] one
ox for a [priceless wish-fulfilling] ma1J-ijewel" (T. 46.13a21-23).
Both of these analogies-the person with a stuffy nose and the rus-
tic man-are references to a series of analogies found in the
Pratyutpanna-samiidhi-sutra (907a-908b). Both are a bit cryptic and
can only be fully appreciated upon reading the full explanation in
the sutra.
8
7. For a full translation of the Tibetan version of this passage see
Harrison 1990, 32.
8. My translation has already been "padded." The translation given above
10 JIABS 20.1
It seems Chih-i assumed that his audience would immediately
recognize and' understand these analogies, much like a modern
audience would understand or evoke a whole range of id.eas and
emotions upon hearing "crying wolf," "finger in the dike," "bark-
ing up the wrong tree," "a material girl," or "Butt-head." The mod-
ern audience, on the other hand, needs help unpacking "a rustic.
man who offers an ox" or "seeing seven jewels and one's relatives
in a dream and rejoicing" in order to make any sense out of them.
At this point one may ask: When Chih-i summarizes or picks up
certain phrases and omits others, does he pick up only what he
thinks is important, or does he assume that his readers or listeners
are familiar with the context and will know how to fill in the details
on their own? Is he deliberately emphasizing certain points, or
does he intend his summary to stand metonymously for the
whole? In some cases, such as the analogies provided at the very
end of the passage, it is obvious that he is just giving a hint and
expects the full content to be known to the listener. In other pas-
sages this is not so obvious. In translating such passages, one must
judge how much of the context and background of the source
material should be provided so that the modern reader will have
sufficient information to understand and interpret the text. In
most cases, a merely "accurate" literal translation captures at best
only the surface meaning, and in the worst case leaves only a jumble
of meaningless, contextless words.
3. Chih-i's quotation in this section, as throughout the Mo-ho chih-
kuan in general, is very accurate. He does not slip into the habit
common among many Chinese Buddhist commentators to mis-
quote his sources. This makes the few places where he departs
from accuracy all the more interesting and significant.
One "inaccurate" quote with interesting doctrinal implications is
is a compromise between' a literal translation and a fully expanded transla-
tion of
[literal] "Like country man for ma1Jijewel offers one ox."
[full translation, including material incorporated from the analogy in
the sutra]: "It is as if a naive and ignorant country man, who does
not realize the priceless value of a magical wish-fulfilling jewel,
offers only a single head of oxen in exchange for it."
SWANSON 11
found near the end of this section (12c22), where Chih-i has "the
mind sees the Buddha" ,C"JMIIl instead of "the mind is the Buddha"
This could well be a copyist's error, since the characters
and :;llk are visually similar. However, the Mo-ho chih-kuan drops a
phrase in ine sutra immediately after the character and picks
up again on the next phrase beginning with ,c". This causes the
Taish6 edition editors to punctuate the phrase so that "Buddha-
mind" becomes a compound (the mind sees the Buddha-
mind), a punctuation and reading that is not possible in the orig-
inal sutra.
A few lines later (12c27) we find the Mo-ho chih-kuan quoting a
verse from the Pratyutpanna-samiidhi-sutra (909a7-S). The first
line has instead of but the meaning is much the
same. The second line has "the mind, without blemish, is called 1;
pure" instead of "the mind is pure, clear 1J)j, and without blemish."
This difference may be due to a mistake in taking down Chih-i's
spoken words, since 1; and IJ)j are homonyms.
Again, a quote (13a6-7) from a verse in the
siistra (26.S6al4-15) uses the compound "covetously attached"
instead of "be tainted with attachment" but once again
with no significant difference in meaning.
In short, unlike some other sections in the Mo-ho chih-kuan,
there are no passages in this section where Chih-i can be shown
consciously or unconsciously to reword and misrepresent his
sources.
4. As with the other three of the four Samadhis (and almost every-
thing else in the Mo-ho chih-kuan), Chih-i interprets this pratyut-
panna samadhi in terms of the threefold truth =:Mi'f -emptiness
conventionality 15:, and the Middle 9=I-giving a "creative" reread-
ing and analysis of the sources. The threefold truth, of course, is
not explicit in the Pratyutpanna-samiidhi-sutra, though Chih-i
would say that the pattern is implicit.
Mter briefly outlining the content of the samadhi, and its physi-
cal and verbal requirements, Chih-i gives an analysis of the mental
aspects of this practice. His analysis is in three parts: contemplat-
ing the emptiness of all things, including the Buddha that is visual-
ized; contemplating the conventional or provisional reality of that
which is visualized; and realizing the Middle Path, that all things
12 JIABS 20.1
are both empty and conventionally real, or, in Chih-i's words,
''These dharmas cannotbe signified [with wordsJ-they are all the
result of thoughts. And even if one [conventionally] est.ablishes
the existence of thoughts, they are ultimately empty and without
being" (12c24-25). Chih-i draws on various passages from the
Pratyutpanna-samiidhi-sutra and creatively reinterprets them in
terms of the threefold truth. For example, Chih-i refers to an
analogy in the Pratyutpanna-samiidhi-sutra (905a27-c5) of a man
who dreams or imagines having sexual intercourse with a woman
far away in Sravasti, in order to illustrate that dharmas, despite
their ultimate emptiness, function in a conventional sense ("he
realizes that she did not [really J come and he did not [really] go
[anywhere], and yet he enjoyed her as if she was [physically with
him] "). This method of extracting the "profound meaning"
(hsuan-i : t : ~ ) of the scriptures is one of the most significant
aspects of Chih-i' s use of scripture-that is, the spelling out of what
Chih-i saw as the implicit teaching of the threefold truth in scrip-
tures that do not explicitly teach the threefold truth pattern.
IV. The analysis of jiieyiivara1Ja
Chih-i taps some unusual sources in a later section of the Mo-ho chih-
kuan (T 46.85b) while discussing klesajiieyiivara1Ja.
9
He first quotes
the Mahiiparinirvii1Ja-sutra, a text Chih-i relies on quite frequently,
but then he relys on the Bodhisattva-bhumi (T no. 1581) and theJu
ta-sheng lun (Introduction to Mahayana, T no. 1634), two texts that
he quotes only infrequently. In the case of the Bodhisattva-bhumi, the
quotes are not literal but rather summaries or just the use of specific
terms. Often it is difficult to tell exactly on which passage Chih-i is
relying, or if he is just summarizing what he takes to be the general
intent of the text as a whole. The quotes from the Ju ta-sheng lun (T
32.45c2-16), on the other hand, are quite accurate, though once
again Chih-i joins together two phrases that are separated by a few
lines in the original text.
In short, this passage is an example of Chih-i relying on texts that,
in general, he quotes relatively infrequently. As expected, his
"quotes" are often "summaries" rather than literal quotations
9. See my article on "Chih-i's interpretation of jiieyiivarar;,a: An applica-
tion of the three-fold truth concept" (Swanson 1983, 51-72).
SWANSON 13
(except in the case of the Ju ta-sheng lun, which he follows quite
closely) .
V. The category of the "four unities" im- in the Fa-hua hsuan-i (T

This passage quotes four scriptures-the Ta chih-tu lun (Paiicavir[iSati-
sahasrika-prajiiaparamita-sutra) , the
the Avataf!lsaka-sutra, and the Vimalakirti-sutra-to provide proof-
texts for Chih-i's teaching of the "four unities": the unity of teaching
n-, the unity of practice 11'-, the unity of persons A -, and the
unity of principle (or "reality") J:1. - .
a. The prooftext for the unity of teaching, that "all dharmas are
included in the Mahayana," is attributed to the explanation of
"prajiia," which for Chih-i usually means the Ta chih-tu lun. I could
not locate the exactphrase in the Ta chih-tu lun, however, except for
a similar passage (T 25.389c) that says "by riding on the Great
Vehicle one attains universal wisdom and turns the dharma-
wheel."
b. The next prooftext on the unity of practice, that "to clearly under-
stand all characteristics of reality ... this is the universal practice
of a bodhisattva," is an exact quote from the
pariprccha-sutra (T 15.37c22-26), though the two phrases in the
sutra are separated by a number of lines.
c. The proof text on the unity of persons, that "one enters the
dharmadhatu without moving from the J eta Grove," is a terse sum-
mary of a passage from the famous chapter on "entering the
dharmadhatu" in the Avataf!lsaka-sutra (T 9.683c-684a), where it is
emphasized that one does not physically, or any other way, actually
go some other place in order to "enter" (or realize) the dharma-
dhatu, that it is not necessary to leave the J eta Grove where Sakya-
muni is preaching in order to attain the realm of perfection, and
that the realm of the Buddha and the realm of ordinary people,
nirvana and samsara, are one.
d. The prooftext on the unity of principle, that "to know all dharmas
in a single thought: this is to sit on the seat of enlightenment
(bodhima'IJda)," is a modified quote from the Vimalakirti-sutra (T
14.543a4-5); the original sutra has "perfecting all wisdom on the
seat of enlightenment."
14 JIABS 20.1
In short, this passage is an example of Chih-i using his regular set
of prooftexts to support a set of categories or ideas, in this case that
of the "four unities."
VI. Prooftexts on "the subtlety of the dharma of mind" (Fa-
hua hsuan-i, T 33.693a23-b2)
Chih-i refers to passages in the Lotus Sutra, Vimalakirti-sutra, and
Avataf!lsaka-sutra to illustrate his concept of the subtlety of the mind.
All of these quotes were accurate, with one exception. The final
quote from the Avataf!lsaka-sutra is given as "by destroying the
minute dust-like obstructions of the mind, the thousands of of
sutras are revealed." Whereas the previous phrases from the
Avataf!lsaka-sutra are quoted verbatim, Chih-i takes liberties in para-
phrasing this puzzling quote. The original (T 9. 624a6-12) reads:
These scrolls of sutras in the great trichiliocosm exist within one
minute particle of dust. All of the minute particles of dust are also
likewise. At one time there was a person who appeared in the world
who achieved penetrating wisdom, completed and perfected the pure
divine eyesight, and perceived these scrolls of sutras in a minute par-
ticle of dust. He then had the following thought: "How can these vast
and great scrolls of sutras exist in a minute particle of dust, yet not
benefit sentient beings? I should diligently use expedient means to
destroy these minute particles of dust and benefit sentient beings." At
this time this person used expedient means to destroy the particles of
dust and extracted these scrolls of sutras to benefit sentient beings.
In the original sutra the word "mind" ,t, is not used, and seems to be
making qucite a different point from that proposed by Chih-i. Chih-
i was quoting this phrase to support his contention that the mind is
subtle, and the phrase that Chih-i "quotes" supports this concept
only insofar as the Avataf!lsaka-sutra in general teaches the identity of .
mind and objects. In that sense the mind and the particles of dust
are one, and "destroying a particle of dust" or "destroying the dust-
like obstructions of the mind" to reveal (the meaning of?) scrolls of
sutras can be interpreted as meaning the same thing. A translation
of Chih-i's paraphrase that would be more consistent with the sutra
passage would read, "Destroying the mind and minute particles of
dust, the sutra scrolls of the trichiliocosm appear."
SWANSON 15
VII. The ten categories of subtlety and the Lotus Sutra (Fa-hua hsilan-
i, T. 33.698a)
In this passage Chih-i quotes only from the Lotus Sutra to give proof-
texts for his t ~ n categories of subtlety, from the subtlety of objects to
the subtlety of religious benefits, the ten categories whose discussion
takes up the bulk of the Fa-hua hsilan-i. The striking thing about this
passage is that all of the quotes are extremely accurate, down to the
exact characters. This illustrates Chih-i's practice of quoting the
Lotus Sutra accurately. The only exception, and striking in its devia-
tion, is the proof text given for the ninth category of subtlety, the sub-
tlety ofthe Buddha's attendants. Chih-i quotes the Lotus Sutra as say-
ing that "[The Buddha] teaches only bodhisattvas, and has no
sravakas as disciples." However, the Lotus Sutra (T. lOb5-6) passage
is slightly but significantly different, and this abbreviated quote is
somewhat misleading. Hurvitz translates the entire context:
.. .1, being King of the Dharma
Universally address the great multitudes,
Having recourse only to the Path of the One Vehicle,
Teaching and converting bodhisattvas,
And having no voice-hearing disciples.
All of you, Sariputra,
Voice-hearers and bodhisattvas alike,
Are to know that this subtle Dharma
Is the secret essential to the Buddhas.10
In the Lotus Sutra the word "only" modifies the "Path of the One
Vehicle," not "bodhisattvas." Thus the sutra says that the Buddha has
recourse only to the doctrine of ekayiina to teach bodhisattvas, not to
teach sravakas, with the implication that he has recourse to other
methods to teach sravakas. It does not unambiguously mean that
the Buddha has no disciples that are sravakas (though the Chinese
of Kumarajiva can be construed in that way, and Hurvitz's transla-
tion follows that line). The wider context makes it clear that the
Buddha is preaching the subtle dharma to all beings, "sravaka and
bodhisattva alike," and the sravakas are included in the group of the
Buddha's disciples. I fear that Chih-i was overzealous in his attempt
10. The phrase in italics could be translated, "to teach and convert bodhi-
sattvas, / and not for [teaching] sravaka disciples." See Hurvitz 1976,46.
-16 JIABS 20.1
to illustrate the "subtlety" of the Buddha's attendants.
VIII. The Four Noble Truths and the Mahiiparinirviir;,a-sutra (Fa-hua
hsuan-i, T 33.700c-701c)
Chih-i's analysis of the four noble truths and its presentation in the
Mahiiparinirviir;,a-sutra is quite complicated,ll but here I will take a
look only at the analysis of the "four noble truths as arising and per-
ishing" !t. ilfX; IZQ ,
Generally Chih-i utilizes the Mahiiparinirviir;,a-sutra accurately,
though once again he tends to paraphrase and / or pick up key
phrases rather than quote word for word. Chih-i also refers to the'Fo
ch 'ui pan nieh p 'an liao shuo chiao chieh ching (T no. 389,
12.1112b24-28), often used in conjunction with the Mahiipari-
nirviir;,a-sutra, to quote: "the causes [of suffering] are true causes,
and there are no separate causes .... The path to extinguish suffer-
ing is the true path." In this case Chih-i uses the character 7.JU ("sep-
arate") instead of ("different") and the character &P ("is") instead
of 1t' ("truly"). This appears to be a result of Chih-i quoting the
sutra from memory rather than a copyist error, since the characters
are not visually similar, and the meaning is much the same.
Finally, Chih-i quotes the Mahiiparinirviir;,a-sutra (T 12.682c7-14) to
say "All ordinary people have suffering, but not the truth, Sravakas
and pratyeka-buddhas have suffering and [know] the truth of suf-
fering." It is curious that Chih-i left out the concluding phrase in the
sutra " ... but not [the knowledge of] the real truth." Perhaps he felt
it superfluous at this point.
IX. Prooftexts on types of Buddha lands in Chih-i's commentary on
the Viirlalakirti-sutra (Manji Zokuzokyo 27.862-873)
Toward the end of his life-after the lectures that became the Mo-ho
chih-kuan and Fa hua hsuan i and after he had returned to Mt. T'ien-
t'ai from the capital-Chih-i composed commentaries on the
Vimalakirti-sutra. Since these texts are among the few of Chih-i's
works written in his own hand, they can provide more direct evi-
dence regarding Chih-i's use of scripture. Let us take a look at a pas-
sage in the Wei-mo ching wen-su where Chih-i is commenting
11. For details see Swanson 1989, 142-44 and 226-34.
SWANSON 17
on the opening chapter of the Vimalakzrti-sutra and giving scriptural
references on the "Buddha Lands" f 5 i l ~ .
First we may note that Chih-i relies to a great extent on his usual
array of primary texts: the Lotus Sutra, Mahiiparinirvii1fa-sutra, and Ta
chih-tu lun, not to mention the Vimalakzrti-sutra itself. The Avataf!t-
saka-sutra is also referred to frequently. There are also scattered ref-
erences to texts Chih-i uses on occasion, such as the Jen wang ching,
the Vajracchedikii-prajiiiipiiramitii-sutra, the Siimiiliidevz-sutra, and the
Contemplation of Samantabhadra Sutra.
It is worth noting that there are numerous references in this sec-
tion to the Mahiiyiinasangraha, a text rarely used by Chih-i. However,
the Mahiiyiinasangraha is not quoted directly, but rather referred to
in general, as in 'The Mahiiyiinasangraha clarifies that there are seven
types of samsara" (865al) and ''The Mahiiyiinasangraha clarifies the
realm of the blossom-king" (865b13).
In contrast, the quotes from the Lotus Sutra are direct and accu-
rate, such as the reference (864b17-18) to the phrase "I will become
a Buddha in another realm" (T. 9.25c17); another reference soon
after (865a3) to a phrase in the same section of the sutra (T.
9.25c14-15) that "after my extinction there shall again be disciples
who, not having heard this scripture ... "; or again soon thereafter
(865b9), a truncated but accurate "[He or she shall also see] this
Saha world-sphere, [its soil made of vai<;lurya;] flat and even; ... a
multitude of bodhisattvas all dwelling in its midst" (T. 9.45b19-21).
The references from the Ta chih-tu lun and Mahiiparinirvii1fa-sutra
fall between direct literal quotation and general reference. For
example, Chih-i attributes to the Ta chih-tu lun the teaching that
"Arhats and pratyekabuddhas enter nirvana, and though they are
not reborn in the triple world, there is a pure land in the transcen-
dent realm wherein they experience the body of dharma-nature"
(865a9-10). A computer search of the Ta chih-tu lun on CD-ROM
failed to turn up this exact phrase, but it appears to be a summary
of a passage toward the end of the text (T. 25. 714a-b) that discusses
how arhats dwell in a pure land and experience the body of the
dharma-nature (and thus fail to attain the further goal of Buddha-
hood because they are attached to Hinayana). The same is true for
another quote soon thereafter, that ''The Buddha of the body of
dharma-nature preaches the Dharma as the Dharmakaya bodhi-
18 JIABS 20.1
sattva, and in this land there are no sravakas or pratyekabuddhas"
(S65bl0-1I), which appears to be a summary from a passage earlier
in the Ta chih-tu lun (T. 25.IS8a-c) that first explains that arhats and
pratyekabuddhas do not have the profound insight and compassion
of a bodhisattva or Buddha, and then proceeds to explain how a
Dharmakaya bodhisattva is transformed into innumerable bodies to
preach the Dharma to sentient beings (ISSclS).
The other references in this section are a mixture of direct quota-
tion and general references. The Jen wang ching is referred to twice
(S65bS-9 and S66a5-6) to quote the same phrase 'Those [on the
stages] of the three levels of erudition and the ten noble stages
=:N+M dwell in [the land of] resultant reward :llt1R, and only the
Buddha alone dwells in the Pure Land" (T. S.S2SaI). The
Contemplation of Samantabhadra is also quoted directly (S66a6-7):
"Sakyamuni is called 'Vairocana Who Pervades All Places,' and his
dwelling place is called Eternal Tranquil Light" (T. 9.392cl5-I7),
though (like the Taisho text and unlike other variant texts) it lacks
the character f\lll for "Buddha" after the name Sakyamuni. On the
other hand, the Avataf(iSaka-sutra is referred to generally to point out
that "it clarifies the realm of Indra" (S65bI2) and "it clarifies ten
types of Buddha lands" (S66bI).
This section, then, gives the impression that Chih-i was quoting
texts from memory rather than referring directly to the texts as he
was composing his commentary. The passages from the Lotus Sutra,
with which he was no doubt familiar and could quote from memory,
as well as other familiar phrases from various sutras, are quoted in
full and with a high level of accuracy. Other texts are referred to.
generally ip terms of broader content or short phrases. Again, the'
mix of texts referred to by Chih-i is much the same as we find in his
earlier works, such as the Mo-ho chih-kuan and Fa-hua hsuan-i. The
indications are that there are no significant differences in scriptural
quotation between works written in Chih-i's hand and the works we
have as a result of lectures recorded and edited by Kuan-ting.
Some General Observations on Chih-i's Use of Scripture
The above examples are typical of Chih-i's use of scripture and could
be multiplied almost endlessly. On the basis of these examples and
my experience in translating large portions of these texts, I would
SWANSON 19
like to offer a few general observations on what I see as some char-
acteristics of Chih-i's use of scripture:
l. First, the .texts that Chih-i quotes or refers to most frequently are
as follows: 12
a. texts quoted very frequently, almost habitually:
- Lotus Sidra (T no. 262=Kumarajiva's translation)
Mahiiparinirvii1Ja-sutra (T no. 374, no. 375)
Vimalakzrti-sutra (T no. 475)
Ta chih-tu lun (T no. 1509)
(includes the PaiicavirrtSati-siihasrikii-prajiiiipiiramitii-sutra)
b. texts quoted relatively frequently
Avata'f!lSaka-sutra (T no. 278)
various Agama sutras
Mahiivaipulya-mahii-saf!tnipiita-sutra (T no. 397)
Middle Treatise (Mulamadhyamaka-kiirikii; T no. 1564)
Ch 'eng shih lun (*Satyasiddhi-siistra; T no. 1646)
c. texts quoted on special occasions, where such texts have partic-
ular relevance (e. g., the Pratyutpanna-samiidhi-sutra in the sec-
tion on the Constantly-Walking Samadhi), or texts quoted
numerous times but relatively infrequently:
Jen wang ching (T no. 245)
Contemplation of Samantabhadra Sutra (T no. 277)
Sr'imiiliidevt-sutra (T no. 353)
Pratyutpanna-samiidhi-sutra (T no. 418)
Maiijusr'i-parip!cchii-sutra (T no. 468)
(T no. 586)
Suvarn1Japrabhiisa-sutra (T no. 663)
Ying lo ching (T no. 1485)
(T no. 1511)
(T no. 1546), etc.
In general, the more frequently a text is cited, the more likely it is
that the text is cited accurately. For example, the Lotus Sutra is
12. This list is made on the basis of my own, as yet uncatalogued, impres-
sions from working on a translation of the Mo-ho chih-kuan and Fa-hua hsiian-
i, as well as the cumulative list of texts cited in the Mo-ho chih-kuan as given
in the Makashikan in yo tenkyo saran, ed. Chugoku Bukkyo KenkyUkai,
1987.
'20 HABS 20.1
most frequently quoted exactly as it is found in Kumarajiva's trans-
lation. It is therefore of special interest when a passage from the
Lotus Sutra is found to be used in a way different from its use in
Kumarajiva's translation. In contrast, the MahiiparinirviirJa-sutra is
also cited frequently, but often not in the form of a "prooftext"
accuracy but in reference to a story or analogy found in this sutra.
Thus, references to the MahiiparinirviirJa-sutra are more likely to
be summaries or short citations needing to be filled out, and more
likely to be given a creative interpretation. Texts cited infre-
quently tend to be less accurate.
Also, when the Paiicavif(tsati-siihasrikii-prajiiiipiiramitii-sutra is
quoted, it seems that Chih-i is usually referring to the Ta chih-tu
lun. Sometimes he will even say, "in the Ta chih-tu lun," but in fact
the passage is a part of the Ta chih-tu lun that is quoting the origi-
nal Paiicavif(iSati-siihasrikii-prajiiiipiiramitii-sutra. There are also
examples of the opposite; Chih-i says he is quoting the sutra but
he is actually quoting the commentary in the Ta chih-tu lun. This
indicates that Chih-i was reading the Paiicavif(tsati-siihasrikii-
prajiiiipiiramitii-sutra through the lens of the Ta chih-tu lun.
2. It is not unusual for Chih-i to quote texts with word-for-word accu-
racy. but it is more often the case that he summarizes or briefly
rewords the original source. Even in the case of the Lotus Sutra,
which is usually quoted word-for-word, it is not unusual for pas-
sages to be summarized or reworded.
3. There are many cases in which Chih-i does not specifically say "in
the sutras";fIB or identify the source he is quoting, but in fact he
is either quoting or summarizing a scriptural source. There is
therefore more "quoting" going on than one might assume from
just picking up the passages identified as from a specific sutra (e. g.,
in the sections on the Four Samadhis).
4. In cases where Chih-i's quote differs from the "original source" (i. e.,
the edition [s] we have in the Taisho canon), almost inevitably the
quote will differ from the source in a way that serves to support
the point that Chih-i is trying to make. This may seem like an
obvious point, but it is significant. It indicates, for one thing, that
the differences between Chih-i's quote and the original source are
more likely due to a conscious or unconscious manipulation or
creative reinterpretation on Chih-i's part, rather than due to copy-
SWANSON 21
ist errors or variant texts.
Of course one must be aware that the sources available to us
today may be different from the texts that Chih-i had at hand, and
that there are dangers to my identification of "original sources" as
the texts we have in the Taish6 edition of the Buddhist canon.
How certain can we be that the MahiiparinirviirJa-sutra prin ted in
the Taish6 edition is the same as the version Chih-i himself
referred to? Until we discover the remains of Chih-i's library, there
is no way to be absolutely certain. However, when we see that for
the most part the quotations by Chih-i of the Lotus Sutra and
MahiiparinirviirJa-sutra, for example, are indeed the same as those
found in the Taish6 edition, we can conclude that Chih-i was refer-
ring to much the same text, and the quotes that are different from
the source as found in the Taish6 canon take on a greater
significance.
Further, when Chih-i's quote differs from the source as we have
it in the Taisho canon, the following possibilities need to be con-
sidered:
a. The text in the Taisho canon is different from the one used by
Chih-i. This is quite possible, and would require complicated
textual histories to sort out. The differences, however, may be
minor-such as a single character or compounds-caused by
errors in transcription. In these cases it is often quite clear what
has happened-the characters are visually similar or homo-
phones and often do not result in a significant change in the
meaning of the text.
b. Texts such as the Mo-ho chih-kuan and Fa-hua hsuan-i are said to
have been oral lectures taken down by Chih-i's disciple Kuan-
ting. One explanation for differences between quotations and
the sources is that since these are lecture notes, the quotes were
written down from memory and not copied directly from man-
uscripts. Still, this does not exclude the possibility of Chih-i's
creative (and perhaps unconscious) rephrasing of scripture in
his memory as he was lecturing, resulting in "wishful remem-
bering." It should not be forgotten, however, that Kuan-ting
edited these notes over a long period of time and certainly had
the opportunity to check the sources. I find the argument
explaining away misquotes as attributable to oral transmission
22 JIABS 20.1
difficult to accept.
c. Chih-i himself had no intention of always quoting scripture
word for word, but was more interested in summarizing or rely-
ing what he felt was the intent of the sutras. This opens up the
whole question of Chih-i's view of "scripture" and his attitude
toward "authoritative texts." (This point is discussed in more
detail below.) In a word, how far should we apply modern stan-
dards of textual critique to figures such as Chih-i?
d. Finally, there is the possibility that Chih-i, while fully aware of
the original form of the quotation, deliberately modified the
quote to be more in line with the understanding and interpre-
tation that he intended to convey. I believe there are cases in
the Mo-ho chih-kuan and Fa-hua hsuan-i where this is the best
explanation of the situation.
5. When Chih-i refers to the same sutra numerous times in the same
section, it is likely that he is referring to the same section in the
sutra. This may seem like a minor point, but it is a useful rule to
keep in mind when trying to identify the source exactly. It also
gives an indication of how Chih-i was working. As scholars, there
are times when we recall a pertinent source or quote to back up
an argument as we are writing a paper, and incorporate it as such
into the paper. There are other times, however, when we are writ-
ing a paper while already referring to a specific passage, e. g., in the
Mo-ho chih-kuan or Lotus Sutra, and we develop our argument along
the lines of the source material. This second method seems to be
used often by Chih-i, and his argument is clarified by reference to
the wider passage that he is relying on.
6. Chih-i often utilizes a kind of metonym by allowing a single word
or phrase to stand for a whole idea or passage. This is common
even in modern Chinese, such as using a short set of Chinese char-
acters to refer to an entire well-known poem, or in the use of the
phrase "E'loi, E'loi, la'ma sabach-tha'ni" (My God, my God, why hast
thou forsaken me?) attributed to Jesus on the cross to evoke the
entire 22nd Psalm. Chih-i often uses only a few characters or words
to evoke an entire verse, parable, analogy, or section of a sutra-see ,
the discussion above on the Constantly-Walking Samadhi. Often
the few words Chih-i uses convey little or no meaning in them-
selves, or can actually be misleading, until they are placed in the
SWANSON 23
context from which they come, or until the entire verse or passage
is known. As mentioned above, in such cases it is the translator's
and interpreter's responsibility to judge how much of the evoked
passages should be incorporated into the final translation to more
accvrately convey what Chih-i was actually trying to say.
7. Finally, as discussed in detail above, one of the most important
uses of scripture by Chih-i-in addition to direct or indirect quo-
tation-is the re-interpreting of the texts in terms of the threefold
truth (emptiness, conventionality, the Middle) to creatively
extract the "implicit" meaning. This could be called the m"!:ior
purpose of the Fa-hua hsuan-i-an exposition of the "profound (or
'hidden')13 meaning" (hsuan-i, x ~ ) of the Lotus Sutra. Almost
every section of the Mo-ho chih-kuan and Fa-hua hsuan-i ends up as
an analysis of the topicvunder discussion in terms of the threefold
truth of emptiness, conventional reality, and the Middle. Chih-i
himself says, "Many sutras contain the meaning [of the threefold
truth] in detail, though the names come from the Ying lo ching
and the Jen wang ching" (T 33. 704c17). It is not an exaggeration
to say that most of Chih-i's mature work is an exposition of the
threefold truth-both doctrinally and in practical terms-as it
appears implicitly and explicitly in the Buddhist scriptures. Thus
by quoting and explaining the texts in terms of the threefold
truth, he is extracting their "profound meaning," and it is on this
13. Stanley Weinstein has pointed out that, "As was typical of the
founders and systematizers ofthe T'ang schools, Chih-i read the scripture
in light of his own religious intuition and experience rather than in the
literal fashion that had prevailed before his time. Whereas the traditional
method of exegesis had been one of 'literal interpretation' (sui-wen chieh-
shih), Chih-i perfected the method of searching out and expounding the
'hidden meaning' (hsuan-i 3 i : ~ ) of the text, which was subsequently
adopted by such eminent T'ang scholar-monks as Chi-tsang, Shan-tao,
and Fa-tsang .... " (1973,284). Although my readings of Chih-i have
confirmed Weinstein's insightful conclusion, I find the term "hidden
meaning" for hsuan-i to be too strong, for it implies that the sutras or texts
themselves have deliberately "hidden" their real meaning. Also, I have
found no indication that Chih-i himself was conscious of exposing a hid-
den meaning, although he does speak of "revealing" rm the true meaning
of the sutras. I prefer the term "profound [or mysterious, esoteric,
implicit] meaning" for hsuan-i .
24 JIABS 20.1
basis that he. chooses which texts to quote, and how to quote (or
summarize) them.l
4
Summary
As I have indicated, Chih-i is not above bending his sources to his
own liking. Still, despite the title of my paper, it is important to
emphasize that generally Chih-i is very "true" to sources, at least in
the sense that his quotations are usually quite accurate (as far as we
can tell with the Taish6 edition texts). Where the quote does seem
to deviate from the source, however, it appears that Chih-i opts to
rely on the "meaning" rather than the letter, and on the perspective
of wisdom gained from his own practice of the Buddha-dharma
rather than on ordinary literal interpretation.l
5
In fact Chih-i suc-
cinctly states his approach to scripture in a passage near the end of
his discussion of the meaning of the "cessation-and-contemplation"
(T 46.26b20-26): "These interpretations are based on [the insights
gained from] contemplation of the mind. It is true that they are not
categories set out in order on the basis of reading thesutras.
However, in order to avoid suspicion and doubt among people, and
to increase and strengthen faith [in the Buddha-dharma], happily [1
can say that] they are in agreement with the sutras. Therefore 1 have quoted
(from sutra passages] as a witness." 0 0

This approach actually makes the "inaccurate" quotes all the more
significant and interesting, because in a positive sense they reflect
most clearly the insightful genius of Chih-i's understanding of the
Buddha-dharma. On the other hand, if Chih-i quotes a sutra as
authoritative backing for his ideas while the sutra clearly says some-
thing else, then we have to consider the possibility of inaccurate or
irresponsible scholarship (or at least he is not justified in resorting to
14. For a discussion of Chih-i's use of apocryphal texts, see my paper on
"Apocryphal Texts in Chinese Buddhism: T'ien-t'ai Chih-i's Use of
Apocryphal Scriptures," (forthcoming).
15. This approach is consistent with the "four reliances," a traditional
Buddhist hermeneutical technique: "Rely on the teaching, not the
teacher; rely on the meaning, not the letter; rely on the definitive mean-
ing, not the interpretable meaning; rely on wisdom, not on [ordinary]
consciousness." See Lamotte 1988, 11-27.
SWANSON 25
scriptural authOlity as moral or theoretical support for his position).
Whether this is done in sixth-century China or twentieth-century
Europe, there are certain practices that by their nature are academ-
ically suspicio1,ls, and we should be aware of these passages. The
question of whether or not Chih-i's "creative interpretation" is "cor-
rect" or "profound" or "insightful" is, of course, a different problem
that involves a different set of value judgments.
Appendix: Thoughts on Working in Two (Or More) Languages
Upon preparing papers on this subject of Chih-i's use of scripture in
Japanese and then attempting to prepare the "same" paper in
English, I learned what soon becomes obvious to anyone who works
in more than one language, namely, that the essay would not simply
"translate" into English. Not only do the words and ideas fail to
carry the same nuances, but I found myself pulled in different direc-
tions by the force of the words and ideas in the different languages,
and the essay would thus develop in a very different way. It seemed
to me that these forces were at work in at least three (often overlap-
ping) levels-that of individual words or terms; that of more gener-
al ideas and their implications; and that of the intended audience.
1. First let us examine the level of individual words. As anyone
working in translation quickly realizes, there are no "exact" equiva-
lents for translating from one language into another, and there is no
one final, correct translation. Each word has multileveled meanings
and implications that can never be carried over into another language.
When a certain word is used, it carries with it layers of historical
development and half-hidden associations that are often uncon-
sciously present even to the original writer, not to mention the
entirely different way that Chinese characters work (with their visual
implications) as compared with alphabetic languages. Let us take a
couple of examples from the Mo-ho chih-kuan.
The character ;@ is used in many different ways by Chih-i, some-.
times technical and sometimes not. In a nontechnical sense it can
be used as the verb "to cross," to go from one place to another,
either physically or mentally. It can imply "penetration," particular-
ly in an adjectival sense of "penetrating insight," or can be used by
itself to refer to this penetrating insight that is achieved through
contemplation. It is used as a translation of "supranormal powers"
26 JIABS 20.1
(abhijiiii) , and it is used as.a technical term for the "common" or
"shared" teachings that are the third of the Fourfold Teachings in
the T'ien-t'ai doctrinal classification system. Finally, the character
itself subliminally implies some sort of movement or progress, and
visually suggests the similar character for "the path" J!!. Of course it
is impossible to translate all these meanings and nuances into
English, and if one aims at "consistency at all costs" and tries always
to translate this character with the same English word, the result will
be nonsense. The translator and interpreter must deal with the
term in its context and interpret accordingly.
Similar comments may be made about the term chih-kuan 11:iII.. It
is not enough simply to identify this compound as a Chinese trans-
lation of samatha-vipafyanii, for Chih-i uses it in ways that the original
may not have been used, and the Chinese characters themselves
offer images and nuances not available in the original Sanskrit. As
Chih-i himself points out (see, for example, T. 46.21b-23c), 11: and
iII. can be used either as verbs or nouns, as both action and object, as
both the practice and the goal attained through the practice. Thus
11: is both the stopping and stilling of delusion, passions, and
obstructive thoughts, and the quiescence that results from such
practice and attainment; iII. is both the contemplation of things cor-
and the insight that results from such contemplation. The
term "cessation" can be used to translate both aspects of 11:, but I still
have not found a satisfactory solution to translating iII., except to usb"
"contemplation" for the active meaning and "insight" for the goal
attained.l6
To give a more pedestrian example, recently I was translating the
Mo-ho chiJl-kuan when I came across the compound in a con-
text in which was discussing the inexpressibility of ultimate
truth (T. 46.21b7), and I proceeded to clumsily yet literally translate
the phrase "the way of discourse is severed." Mter all, my dictionary
told me that the original Sanskrit for this term was * sarva-viida-caryii-
uccheda. The next morning, by coincidence, my eyes fell on a head-
16. Of course I am not advocating arbitrariness. I have heard that in a
paper published on T'ien-t'ai Buddhism recently in mainland China, J.l:.fi
was translated as "stop contemplation." This may be an "accurate" trans-
lation, but it is also wrong.
SWANSON 27
line in the newspaper (Chunichi Shinbun, 14 March 1993, 12) that
read "Kanemaru's Illegal Savings are 'Beyond Words'
heading an article explaining how members of the financial com-
munity were left speechless and shocked by the extent and method
of former political king-maker Kanemaru Shin's stocking up of ille-
gal political contributions and kickbacks. Since then I have run
across the term in many contexts, and I realized that the phrase
(gongo-dodan in Japanese) had found its way into common parlance,
meaning "unspeakable, beyond description, inexcusable, out of the
question, making one at a loss for words, preposterous, abominable"
and so forth, and that the phrase probably meant much the same to
Chih-i and should be understood simply as "beyond description,"
rather than the technically "correct" "the way of discourse is sev-
ered." How many more such phrases are there, I often wonder, that
we translate technically and literally but that actually have a much
more commonplace meaning?
2. At the level of ideas, as with individual words, one is often led in
different directions even if one begins with what are close "equiva-
lents" in English and Japanese. One might start with the English
and Japanese titles I gave to my "same" essay on the current subject.
The Japanese titles and
1.1::. nJ :l3lt .oKJIll O)f'!Jtl) are rather staid expressions of an
intent to present a textual and doctrinal theme, whereas the original
English title of this essay-"Say What!? Chih-i's Use (and Abuse) of
Scripture"-carries quite a different nuance. Also, inJapanese I use
the word kyoten *J:JI!!. while in English I use the word "scripture." The
idea of "scripture" immediately suggests ideas, directions, and impli-
cations different from that of *J:JI!!., though the terms are close
enough to be used to translate each other. By deliberately using the
word "scripture" I drew on a vast background of meaning, feeling,
and nuance associated with the word, some of which have Judeo-
Christian implications that would not be applicable in a Buddhist
context. 17 The use of such words and ideas immediately draws one's
attention and line of argument in a direction that similar words
would not in another language.
The implication for understanding and translating classical or
17. See the essays in Levering 1989.
28 JIABS 20.1
sacred texts is. that surely the same thing was going on when, for
example, Chih-i delivered the Mo-ho chih-kuan. An awareness of this
process may help clarify, for example, sections in which it seems
there is no consistent line of thought, or where there appears to be
ajump in the argument-there may be a nuance in the terminology
or ideas that are forever lost to our consciousness, but which inex-
plicitely determined the flow of the text.
3. Finally, and not unrelated to the above levels, there is the
influence of the intended audience. Preparing a paper on Chih-i
for aJapanese academic audience of Tendai specialists, for example,
and preparing the "same" paper in English for a more general but
Western academic audience, cannot but influence the content and
flow of the material. For example, a paper for the Japanese audi-
ence can assume a certain knowledge of technical terms that one
cannot assume for a Western audience. On the other hand, one can
assume a greater interest among a Western audience for general
hermeneutical issues, or a wider scope of interest in the history of
Buddhism beyond the Sinojapanese developments. This colors not
only the details that one must provide or can avoid, but also the
direction in which one's train of thought will proceed. This leads
me to ask one of those impossible or unanswerable questions: How
different would the content of the Mo-ho chih-kuan be if Chih-i had
given it in Japanese at Otani University at the beginning of the MeiJi
Period? Or if Chih-i would present the Mo-ho chih-kuan in English to
a Western audience today at the Naropa Institute? The answer is:
very different. In a sense the translations and interpretations we
make of the Mo-ho chih-kuan today in our current languages and con-
texts is this "content."
It is clear from the above points that a strict and literal translation
of texts such as the Mo-ho chih-kuan does not do justice to the texts
themselves, and many levels of meaning must be taken into account
to understand the text. A literal translation is flavorless-even if it
succeeds in conveying a surface, uni-leveled meaning :::tfij, it cannot
convey the rich and multilayered flavor ::P* of the original. It is my
belief that the task of the translator and interpreter goes beyond
mechanical word-for-word translation and requires a grappling with
the text, its language, and its conceptual world that results in trans-
lations and interpretations that convey the many layers of the origi-
SWANSON 29
nal, and is in itself rich and multivalent. And this seems to be the
approach that Chih-i took toward his own scriptural sources: a bal-
ance between, on the one hand, a cateful accuracy, and on the other
hand, the out of what he perceived as the deeper ("pro-
found") meaning of the Buddha-dharma that led to the rephrasing,
and at times even misquoting, of his sources.
REFERENCES
Chiigoku Bukkyo Kenkyiikai ed. 1986. Makashikan inya
tenkyo saran Tokyo: Nakayama ShobO.
Donner, Neal. 1976. The Great Calming and Contemplation of Chih-i. Chapter
one: The synopsis (translated, annotated, and with an introduction). Diss.
The University of British Columbia. Ann Arbor: University Micro-
films International.
Donner, Neal, and Daniel B. Stevenson. 1993. The Great Calming and
Contemplation: A Study and Annotated Translation of the First Chapter of
Chih-i's Mo-ho chih-kuan. Honolulu: University of Hawai'i Press.
Gregory, Peter, ed. 1986. Traditions of Meditation in Chinese Buddhism.
Studies in East Asian Buddhism 4. Honolulu: University of Hawai'i
Press.
Harrison, Paul. 1990. The Samadhi of Direct Encounter with the Buddhas of the
Present. Studia Philological Buddhica Monograph Series V. Tokyo: The
International Institute for Buddhist Studies.
Hirai Shun'ei 1985. Hokke mongu no seiritsu ni kansuru kenkyu
Tokyo: Shunjiisha.
Hurvitz, Leon. 1976. Scripture of the Lotus Blossom of the Fine Dharma. New
York: Columbia University Press.
Lamotte, Etienne. 1988. 'The Assessment of Textual Interpretation in
Buddhism." Buddhist Hermeneutics, ed. Donald Lopez. Studies in East
. Asian Buddhism 6, Honolulu: University of Hawai'i Press. 11-27.
Levering, Miriam, ed. 1989. Rethinking Scripture: Essays from a Comparative
Perspective. Albany, NY: State University of New York Press.
Stevenson, Daniel B. 1986. "The Four Kinds of Samadhi in Early T'ien-
t'ai Buddhism." Traditions of Meditation in Chinese Buddhism, ed. Peter
Gregory. Honolulu: University of Hawai'i Press, 45-97.
Swanson, Paul L. 1986. "Chih-i's Interpretation of JiieyavaraI).a: An
Application of the Threefold Truth Concept." Annual Memoirs of the
Otani University Shin Buddhist Comprehensive Research Institute 1: 51-72.
30 JIABS 20.1
----,. 1989. Foundations of T'ien-t'ai Philosophy: The Flowering oj the Two
Truths Theory in Chinese Buddhism. Berkeley: Asian Humanities Press.
----.. forthcoming. "Apocryphal Texts in Chinese Buddhi.sm: T'ien-
t'aiChih-i's Use of Apocryphal Scriptures." Canonization and Decanon-
ization: Contributions to a Theme for Religious Studies, Leiden: E. J. Brill.
Weinstein, Stanley. 1973. "Imperial Patronage in the Formation ofT'ang
Buddhism." Perspectives on the T'ang, eds. Arthur Wright and Denis
Twitchett. New Haven: Yale University Press.
GEORGES DREYFUS
Tibetan Scholastic Education
and The Role of Soteriology
The Hermeneutical and Rhetorifal Dimensions of Commentary
Education is not the mere handing down of knowledge but the active devel-
oping of the person through the internalization of a tradition's content. If
this process begins in the Tibetan monastic education with the acquisition
of basic literacy and the heuristic of memorization, it continues with the
hermeneutical practices aimed at appropriating the content of tradition as a
basis for the cultivation of virtues. In general, hermeneutics can be defined
as the art of interpretation systematically analyzed from a philosophical or
methodological point of view. Tibetan scholastic educational activities are
hermeneutical in that they are reflective interpretive practices that aim to
understand the content of the root-texts used as bases of the educational
process and their commentaries. These root-texts are themselves commen-
taries that are memorized and studied in the light of further commentaries.
The interpretation of commentaries is thus one form that hermeneutical
practice takes in the Tibetan Buddhist tradition. It is not, however, the only
or even necessarily the main one, for a remarkable feature of much Tibetan
scholastic education is the importance of dialectical debates. They sustain
the students in their investigations and lead to an in-depth comprehension
of the tradition, Dialectical debates, together with commentary, represent
the two central aspects of the hermeneutical practices that form the core of
Tibetan scholastic education.
A study of the interpretive practices of a tradition cannot focus, however,
only on the interpreted message. It must also examine the audience to
which this message is addressed and the way in which the author or
transmittor of this message intends to influence its audience. To interpret
means to clarify, explicate, explain, but also to translate, render, and trans-
pose. Interpretation is the work of an interpreter, a go-between, who
mediates between an author and an audience (in some cases himself or her-
self). Interpretation, as Mailloux puts it, "conveys the sense of a translation
pointed in two directions simultaneously: toward a text to be interpreted
31
32 JIABS 20.1
and for an audience in need of interpretation."l Hence, a study of the inter-
pretive practices of Tibetan scholasticism must take into account this double
orientation and consider the semantic as well as the pragmatic oFperforma-
tive dimensions of interpretive practices. Such a study must understand
these practices not only hermeneutically but rhetorically as well. The tradi-
tion is not a pure content but acquires its significance only in relation to the
way in which it is used.
In one sense, it is tempting and not entirely wrong to assimilate commen-
tary to the semantic aspect and debate to the pragmatic or performative
dimension. Nevertheless, as we will see, this distinction is not to
the understanding of either form of interpretive practice. Commentary,
which is our present focus, cannot be understood merely through an exam-
ination of its content. Like other types of text, commentary is not just
descriptive but also performative. The commentator seeks to do something
by writing his text and, more importantly for our purpose, the tradition or
institution that uses his text is also trying to do something through the study
of his words. We could speak here of textual communities, that is, actual
social entities formed around common uses of basic texts and their com-
mentaries. When people engage in common interpretive practices, they
develop a sense of solidarity, of belonging to a distinct community with its
own worldview, ethos and sense of identity. In this way, common inter-
pretive practices provide the focus for further institutionalization and
development of rules. They also become the means through which ne'w
members are introduced to the community.2
Here I examine the pragmatic uses of texts and commentaries in the con-
text of Tibetan Buddhist scholastic education. I approach this education by
examining its curriculum, focusing on the use of texts rather than their
content. In doing so, I follow a comparative approach in order to avoid the
danger of focusing too narrowly on a single tradition, which is then taken
to represent Tibetan tradition as a whole. I examine the curriculum of two
types of institution which include most of Tibetan scholastic education, the
1. S. Maillous, "Interpretation," Critical Terms for Literary Studies, eds. F.
Lentricchia and T. McLaughlin (Chicago: The University of Chicago Press,
1995) 121-134, 121.
2. B. Stock defines textual community as "a group that arises somewhere in the
interstices between the imposition of the written word and .the articulation of a
certain type of social organization. It is an interpretive community but it is also a
social entity," Listening for the Text (Johns Hopkins University Press, 1990)
150.
DREYFUS 33
dGe lugs (pron., ge-Iuk) monastic university exemplified here by Se rwa
(se-ra), which can be described as a debating institution (rtsod grwa), and
the cbmmentarial institution (bshad grwa) exemplified by the rNying rna
(nying-ma) monastery of rNam grol gling (Nam-drOl-ling), which is typical
of non-dGe lugs institutions of higher leaming,3
My examination of the dGe lugs and rNying rna versions of the scholas-
tic curriculum follows a two step approach. I first provide a general com-
parison, following the classical method of delineating similarities and dis-
similarities, between the curriculum of these institutions. In this way I
show the nature of the two types of educational institutions that have domi-
nated the Tibetan scholastic tradition. I then examine the curriculum more
closely by focusing on one of its central topics, the study of the path, and
inquiring into its role in the overall education. I show that in the Tibetan
scholastic traditions this kind of topic, which concerns the practice of medi-
tation' is important less for its direct relevance to meditative practice than
for its contribution to the construction of a universe in which Buddhist
practice becomes meaningful. I conclude by emphasizing the doctrinal
nature of such construction and argue that this reliance on doctrine for the
elaboration of a religious universe is one of the main characteristics of
scholastic education.
The Structure of the Curriculum of a dGe lugs Institution
Se rwa is typical of the great institutions of higher learning that have consti-
tuted the intellectual strength of the dGe lugs tradition. Founded in 1419 by
'Jam chen chos rje Gam-chen-chO-jay), one of Tsong-kha-pa's (dzong-ka-
ba) main disciples, it became a very large monastery in Tibet with more
than ten thousand monks in the 1950s, possibly up to a third of them taking
part in scholarly activities. It is now relocated in Bylakuppe, in South-
India, not too far from Mysore, where it is becoming large again (well over
three thousand) due to a recent influx of new refugees from Tibet.
3. Although there are minute differences between the scholastic institutions of
the three contemporary non-dGe lugs traditions, they all have the same commen-
tarial model of education and are quite similar. This similarity is not accidental,
for they all derive from the scholarly revival initiated by gZhan phan (Zhan-
phan) toward the end of the nineteenth century in the context of the non-sectarian
(ris me d) movement initiated by 'Jam mgon kong sprul Gam-gon-kong-trul,
1813-1899) and 'Jam dbyangs mkhyen tse'i dbang po Gam-yang-kyen-tse-wang-
po, 1820-1892).
34 JIABS 20.1
The Se rwa curriculum
4
does not differ substantially from that of the
other dGe lugs institutions of higher learning. The dGe lugs curriculum in
its different version largely consists of the study of five texts (po ti lnga),
which summarize the exoteric aspects of the tradition, and the study of
tantric texts, patticularly those pertaining to the Guhya-samaja (gsang ba
'dus pa) cyc1e.
5
This curriculum can be divided in three parts.
1) The fIrst preliminary part is devoted to the mastery of the techniques and
basic concepts necessary to the practice of debates. During this period,
which can be as short as one year and as long as four or five years, monks
are trained in the art of debate through the study of the Collected Topics.
They are also introduced to the basic logical and epistemological notions
that they will use throughout their studies. The texts used are textbooks
(yig cha), specifIc to the college within the monastic institution.
-Collected Topics (bsdus grwa) in three parts
-Types of Mind (blo rigs)
-Types of Evidence (rtags rigs)
This preliminary study is often completed by an introduction to the study of
doxography, which examines Buddhist and non-Buddhist tenet systems,
and a Paths and Stages (sa lam) text, so that the students have a good idea
of these aspects of the tradition. This part of curriculum is a preparation for
the main part, the study of the five treatises. It aims at developing reason-
ing abilities. It also provides the student with the basic philosophical
vocabulary required for the rest of the studies, but does not aim to bring to
students any in-depth comprehension.
4. The slight variations in the curriculum between the two scholastic colleges
(Byas and sMad) of Se rwa are irrelevant here. .
5. Sources on the curriculum of the three monastic universities are limited.
Geshe Sopa, Lectures on Tibetan Religious Culture (Dharamsala: Tibetan
Library, 1983) 41-3 and A. Wallace, The Life and Teaching of Geshe Rabten
(London: Allen and Unwin, 1980) 47-9 are the main sources on the state of
monastic education in Tibet. My presentation is also, and perhaps mostly, based
on my stays in these monasteries where I observed monastic education as it has
been reconstituted in exile in India. It is also based on countless conversations
with older monks who constantly referred to the state of monastic life in tradi-
tional Tibet.
DREYFUS 35
2) The second and central part is the study of the five great exoteric texts. It
is subdivided into two phases: a) The main part which consists of the study
of three texts that are considered to summarize the main aspects of non-
tailtric Buddhism as understood by the dGe lugs tradition:
-Abhisamayalarr;kara
6
(Ornament of Realization) attributed to Maitreya
-Candrakirti's Madhyamakavatara
7
(Introduction to the Middle Way)
-Dharmaldrti's PramalJO-varttika
8
(Commentary on Valid Cognition)
Together with Naglirjuna's Treatise of the Middle Way, 9 which is studied in
the light of Candraldrti's Introduction, these texts provide the doctrinal and
philosophical core of the dGe lugs tradition. They are considered the most
important texts and studied with great care for a period of six to ten years.
The students start with the Abhisamayalarr;kara (henceforth the Orna-
ment), which is studied for four to six years. This text provides an under-
standing of the Buddhist and more particularly Mahayana worldview
together with a detailed analysis of the path, as we will see shortly.
Dharmaldrti's Commentary, which present an extensive view of Buddhist
logic, epistemology and philosophy of language, is studied together with
the-Ornament, during special sessions (one or two months every year).
This text is very important, for it provides the philosophical methodology
for the whole curriculum. After being already well trained, students are
ready to examine what is considered the most profound topic of the studies,
Madhyamaka philosophy. Through the study of these three plus one texts,
the students are introduced to the sharp philosophical mode of thinking
particularly valued by this tradition. Sometimes, monks who are keenly
intent on leading the heremitic life leave the monastery after finishing the
study of these three texts. Although they could still benefit from further
studies, they are considered well trained and able to start on their meditative
career.
6. Abhisamayalarr;kara-nama-prajiiaparamitopadeSa-sastra-karika, shes rab
pha rol tu phyin pa'i man ngag gi bstan bcos mgnon par rtogs pa'i rgyan zhes
bya ba tshig le'ur byas pa, P: 5184.
7. Madhyamakavatara-nama, dbu ma la 'jug pa zhes bya ba, P:5262.
8. tshad ma rnam 'gre! gyi tshig le'ur byas pa, P:
5709.
9. Prajfia-nama-mala-madhyamaka-karika, dbu ma rtsa ba'i tshig le'ur byas
pa shes rab ces bya ba, P: 5224.
36 JIABS 20.1
b) The auxiliary and concluding part of the exoteric curriculum brings
more maturity to the already philosophically well trained students through
the study of the last two treatises:
-Vasubandhu's (Treasury of Abhidharma)
-Gunaprabha's Vinaya-siitra
11
These texts bring to the students a grasp of some of the doctrinal and prac-
tical backgrounds of Buddhism. The study of the Abhidharma enriches the
students' understanding of the Buddhist worldview and the kind of spiritual
perspective that this world enables. The study of the Vinaya completes the
monastic curriculum by training the students in the intricacies of monastic
discipline and the collective organization of the monastic order. Thus both
texts are important but contribute little to the kind of intellectual sharpness
that the tradition, and Tibetan scholars in general, particularly value. Hence,
they are thought to be less important, though their studies take a long time
(from four to eight years). The reason for this extended period is due to
several considerations. The amount of textual material is large, but the main
reason is to keep students, who are by then advanced scholars, in residence
so that they themselves become teachers and share their knowledge before
leaving the monastery. It is only after the completion of these studies th,at
students are allowed to stand for the different levels of the title of Geshe
(dge bshes), which brings to an end the exoteric part of the training.
3) Finally, the last part of the studies concerns the esoteric domain of the
tantras. Tantras are not included in the official curriculum of monastic uni-
versities such as Se rwa. Monks who finish their studies and become
Geshe are required to spend some time in a separate college devoted to the
study and practice of tantra. This does not mean that these monks have not
studied tantra before, for almost all of them have, but such a study is con-
sidered private and hence not part of the official curriculum.
The Curriculum of a Commentarial Institution
rNam grol gling monastery, or more specifically its commentarial school, is
typical of the non-dGe lugs institutions of higher learning. rNam grol gling
monastery is the exiled version of dPal yul (pa-yill) monastery, which is
10. chos mngon pa'i mdzod, P:5590.
11. Vinaya-sutra, 'dul ba'i mdo tsa ba, P: 5619.
DREYFUS 37
one of the six great monastic centers of the rNying ma school founded in
1665 by Rig 'dzin kun bzang shes rab (rig-dzing kun-zang shay-rab).
rNam grol gling monastery, which has over a thousand monks, is also relo-
cated in Bylakuppe, a couple of miles from Se rwa. Its commentarial
school, which was started at the beginning of the 1970s, is part of the
monastery but is distinct from it. There are over three hundred students in
the school, which is by now the largest institution of its type in the exiled
community in India. It is quite representative of the style of education
adopted by the three non-dGe lugs Tibetan Buddhist traditions.
In examining the curriculum of the rN am grol gling commentarial school,
it is important to remember that the institution we are examining is different
from a dGe lugs monastery such as Se rwa. Whereas in the latter scholastic
studies are central elements of the monastic routine, in the rNying ma and
other traditions studies are carried on in special institutions that are linked
with the monastery but remain separate, often physically set apart. In rNam
grol gling, the commentarial school (bshad grwa) lies next to the monastery
but has its own administration, kitchen, and temple, though ultimately it is
part of the dPal yul monastery as well. The curriculum of this institution is
centered around the study of thirteen great texts (gzhung chen bcu gsum).
It can also be divided into three parts: a preliminary, a central part (the study
of the thirteen texts themselves), and esoteric tantra studies.
1) The preliminary part, which lasts one year, focuses on two texts: Padma
dbang rgyal's (pe-ma-wang-gyel, fourteenth century) Treatise Ascertaining
the Three Types ofVow
12
and Santideva's Introduction to the Bodhisattva's
Deeds.
13
These texts, which are not counted among the thirteen great texts,
are studied with the help of literal glosses and combined with a few
auxiliary texts teaching grammar and history. During this period students
are introduced to basic Buddhist ideas, Mahayana practices, as well as the
three sets of vow bodhisattva and tantra) to which Tibetan
practitioners usually commit themselves. At this early stage central tantric
concepts already are introduced. For example, the difference between
sl1tras and tantras, a topic formally discussed by dGe lugs scholars only
after they have completed their exoteric studies, here is taken as a pre-
liminary of the whole curriculum.
12. sDoms gsum mam par nges pa'i bstand bcos.
13. Santideva, B bodhicaryavatara,byang chub sems dpa'i spyod pa fa 'jug pa,
P: 5272. S. Batchelor, trans., A Guide to the Bodhisattva's Way of Life
(Dharamsala: Library of Tibetan Works, 1979).
'38 JIABS 20.1
2) The second part is centered on the study of the thirteen great texts. It
can be divided into two phases: a) The lower exoteric course, which last for
three years, 14 begins to expose the students to the different aspects of the
classical exoteric tradition as they are found in the most important Indian
Buddhist treatises. Students I have interviewed often describe Madhya-
maka philosophy as the main topic of these three years. This subject is
examined through the following three of the thirteen texts:
-Nagarjuna's Treatise of the Middle Way
-Aryadeva's Four Hundred Stanzas 15
-Candraldrti's Introduction to the Middle Way
To these three texts several other texts are added. A particularity of this
curriculum is its emphasis on Ornament of the Middle
Way, 16 which is studied together with its commentary by Mi pham rgya
mtsho (mi-pam-gya-tso, 1846-1912). Like in the first phases of the
curriculum, these texts are studied with their commentaries, either literal
glosses, often composed by gZhan phan, or more substantial explanations,
often by Mi pham. The other aspect emphasized during these three years is
the study of the Abhidharma, through an investigation of the following two
of the thirteen texts:
-Asanga's Abhidharma-samuccaya
17
(Compendium of Abhidharma)
-Vasubandhu's (Treasury of Abhidharma)
Together with these other texts such as Mi pharo's Entrance Gate for the
Wise, an introduction to the methodology of scholastic studies that rather
closely follows Sa skya Pal).qita's (1182-1251 C. E., henceforth Sa pal).)
text on the same subject. The fourth year is also occupied by the study of
Buddhist logic and epistemology on the basis of the main text of the
Tibetan tradition on this subject:
14. The prospectus for the rNam grol-gling institute includes the introduction in
the lower sutra course, which thus lasts for four years. It divides the curriculum
in three parts: lower sUtra course, higher sUtra course, and tantra course. My
own division in three plus one parts is made for the sake of comparison with Se
rwa's curriculum.
15. Cattul;Sataka-sa.stra, bstan be os bzhi brgya pa, P: 5346.
16. Madhyamaka.lal1Jkara-pafijika, dbu ma'i rgyan gyi bka' 'grel, P: 5286.
17. Abhidharma-samuceaya, chos mngon pa kun las bstus pa, P: 5550.
DREYFUS 39
-Dhannaldrti's Prama1}avarttika (Commentary on Valid Cognition)
This text is studied together with Mi pham's word commentary. Through-
out this part of the curriculum, a variety of other auxiliary topics (grammar,
composition, poetics, history) are also examined. One of the particularities
of the rNam grol-gling's approach, is the limited role played by the study of
logic and epistemology. This is quite different from the dGe lugs tradition,
which prides itself on its mastery of DharmakIrti's thought. It also con-
trasts with the Sa skya emphasis on the use of Sa paI)'s Treasure as a
primer of Buddhist logico-epistemological studies. IS By the end of the first
four years, students have a sound command of Buddhist philosophy as
well as a good overview of the general structure of the Buddhist tradition.
b) This knowledge is developed by the higher exoteric course, which
lasts for two full years, provides students with an understanding of the
Mahayana tradition, its view oLthe path and result. This course focuses on
the five treatises attributed to Maitreya:
- Mahayanottaratantra
19
(The Superior Continuum)
-Abhisamayalan;kara (Ornament of Realization)
-Mahayana-sutralan;kara
20
(Ornament of the Mahayana Sutras)
-Madhyanta-vibhanga
21
(Differentiation of the Middle and the Extremes)
-Dharma-dharmata-vibhanga
22
(Differentiation of Phenomena and
[Ultimate 1 Nature)
The course is completed by a study of monastic discipline on the basis of
the study of the following two texts, which are the last of the thirteen texts.
18. Sa-gya Par;H;lita, Treasure on the Science of Valid Cognition (tshad ma rigs
gter), Complete Works of the Great Masters of the Sa sKya Sect, vol. 5 (Tokyo:
Toyo Bunko, 1968) 155.1.1-167.1.6. The curriculum of the Sa skya College in
Rajpur (India) includes this text in its list of basic curricular texts.
19. Mahayanottaratantra-sastra, theg pa chen po'i rgyd bla ma bstan bcos,
P:5525.
20. Mahayana-satralan;kara-karika, theg pa chen po'i mdo sde'i rgyan gyi
tshig le'ur byas pa, P:5521.
21. Madhyanta-vibhanga, dbus dang mtha' mam par 'yed pa, P:5522.
22. Dharma-dharmata-vibhGliga, chos dang chos nyid rnam par 'byed pa,
P:5523.
40 JIABS 20.1
-Pratimok$a-siitra (the only teaching of the Buddha on the list of thirteen)
-Gunaprabha's Vinaya-siitra
In this way, students complete the exoteric part of their studies. They have
a sound understanding of a variety of points of view in Buddhist philoso-
phy and a good grasp of numerous aspects of the Buddhist path. They are
ready to move to [mal part of the curriculum.
3) The third part is the esoteric curriculum, the study of tantras. In the
seventh and eighth years, general presentations of the tantric path are exam-
ined. The study focuses on the Guhya garbha tantra, which plays basically
the same role in the rNying rna tradition as the Guhya samaja in the dGe
lugs tradition. The main texts are:
-Yon tan rgya mtsho's (yon ten gya tso) commentary on 'Jigs med gling
pa's Gik-may-ling-pa, 1729-1789) Treasury of Qualities (yon tan
mdzod)23
-Mi pharo's commentary on the Guhya-garbha tantra
24
-rDo grub chen's (do-grub-chen) commentary on the Guhya-garbha
tantra
25
This study is completed by an introduction during the ninth year to the view
of the Great Perfection, the main standpoint of the rNying rna tradition.
The study is theoretical and introductory and focuses on kLong chen rab
'byarns pa's (long-chen-rab-jam-ba, 1308-1363) two trilogies:
-the Trilogy of Self-Liberation (rang grol skor gsum )26
-the Trilogy of Resting (ngal gso skor gsum), particularly the Resting [in]
the Mind as Such (sems nyid rang grol)27
23. Yon tan rin po che'i mdzod kyi 'grel pa zab don snang byed nyi ma'i 'od zer,
(Gangtok: 1969).
24. gSang 'grel phyogs bcu'i mun sel gyi spyi don 'od gsal snying po.
25. dPal gsang ba'i snying po'i rgyud kyi spyi don nyung ngu'i ngag gis mam
par byed par rin chen mdzod kyi Ide mig, Collected Works, vol. 3 (Gangtok:
Dodrup Chen Rinpoche, 1974).
26. Rang grol SkOT gsum, Gangtok: Sonam Kazi, Ngagyur Nyingmay Sungrab,
vol 4.
27. Ngal gso SkOT gsum, Gangtok: Dodrup Chen Rinpoche, 1973.
DREYFUS 41
Together with these works, other tantric texts, particularly Mi pham's com-
mentary on the Eight Words of Practice (sgrub pa bka' brgyad) and 'Jigs
med gling pa's work on the stage of development, are examined. In this
way, students are given a solid grasp of the world of tantras, which is, as
we will see, one of the goals of this education.
Comparing Curriculums: The Organization of Knowledge
If we compare the curriculums of Se rwa and rNam grol gling, we can see
similarities and differences. There is no point here in listing all the relevant
features of our comparison. Rather, let me make a few remarks, starting
with the similarities. One of the most important features of Tibetan
scholastic traditions is the way they organize knowledge on the basis of
.,.
root-texts and their commentaries. As we know, this is not a Tibetan
invention but derives from the methodology used by both Hinduism and
late Indian Buddhism. In traditional India, topics of learning are discussed
on the basis of a root text explicated by further commentaries, including a
teacher's oral explanation. Even considerations of secular topics follow this
model. For example, aesthetics is discussed in relation to the Natya siistra,
a basic text that provides the reference point for the whole field. Similarly,
in Tibet the study of grammar, for instance, proceeds by commenting on
basic texts, in this case the gsum cu pa and the rtags 'jug pa, two grammat-
ical treatises that are said to have been composed by Thonmi Sam bhuta
(seventh century) upon his return from India. Even the study of medicine
is organized around the study of basic texts, the four medical tantras (rgyud
bzhi), which are first memorized and then commented upon. Hence, com-
mentary is central not just to religious traditions, but to the way in which
knowledge is organized in these cultures.
We could even go a bit further and draw a partial contrast between mod-
ern ways of organizing knowledge by disciplines and traditional Indian or
Tibetan reliance on commentary. Modern cultures mostly rely on an
anonymous and abstract organization of knowledge through disciplines
structured around "groups of objects, methods, their corpus of propositions
considered to be true, the interplay of rules and definitions, of techniques
and tools." 28 This is quite different from the Indian and Tibetan commen-
tarial mode of organization which is based on the principle of explication of
a pre-given meaning found in basic texts, which are called root texts (rtsa
28. M. Foucault, "The Discourse on Language," The Archeology of Knowledge
(New York: Harper, 1969, 1972) 215-237, 222.
42 JIABS 20.1
ba, mula). These texts are most often versified, that is, written in kiirika
(tshig le'ur byed pa) or mnemonic verses. In the Hindu traditions, these
texts are called sutras, the aphoristic summaries of a tradition's scriptural
basis, following the methodology developed in Patafijali's grammatical
tradition. For example, the meaning of the is summarized by the
Brahmasutra, which is in tum further explained by commentaries. Such
texts are not written to be picked up and read by anybody, but are intended
to serve as the basis of further oral and written commentary. They would
be read in relation to a bhCi$ya or a VItti Cgrel ba), a commentary often
written by the author of the root text. Those in tum could be supplemented
by a vyCikhyCi or /ikii egrel bshad),29 a more detailed gloss used to supple-
ment the fIrst commentary.3
o
Tibetan curriculums are similarly structured. The root-texts' that are
memorized and studied in the exoteric part of the curriculum are all, with
one or two exceptions, Tibetan translations ofIndian treatises (bstan beos,
sCistra), All of the five or thirteen texts listed above, with the exception of
the Pratimok$a-sutra, fIt in this category)! This extended use of commen-
tary is fairly unique in the Buddhist world. Certainly, other Buddhist tradi-
tions use commentaries but the Tibetan reliance on commentary is stronger
than in most other Buddhist traditions, which tend to rely more on the study
of the direct teachings of the Buddha and less on later commentaries. For
example, both Chinese and Theravada Buddhisms tend to emphasize the
study of the direct teachings of the Buddha as they are contained in their
versions of the canon.
32
Monks in these traditions study the words of the
29. A brief examination of the Tibetan catalogues of the bstang gyur suggests
that the Tibetan translation of these terms is far from systematic. The word
bshad pa isused to translate a vyakhya as well as a a bha$ya. See P: 5555 and
5565. /
30. L. Gomez, "Buddhist Literature: Exegesis and Hermeneutics," Encyclopedia
of Religion, vol. 2 (New York: Mcmillan, 1987) 529-540, 532.
31. One exception is the inclusion in the Sa skya curriculum of Sa pal)'s Trea-
sure, That this exception is also a treatise is quite revealing of the role of treatise
in the Tibetan scholastic tradition.
32. On the different canons, see W. E. Clark, "Some Problems in the Criticism
of the Sources of Early Buddhism," Harvard Theological Review 18.2 (1930):
121-147. For the Pali canon, see K. R. Norman, pali Literature (Wiesbaden:
Harrassowitz, 1983) and S. Collins, "On the Very Idea of a Pilii Canon," J oumal
of the Pali TextSociety 15 (1990): 89-126. On the Chinese canon, see K. S.
Chen, Buddhism in China (Princeton: Princeton University Press, 1964) 365-
386.
DREYFUS 43
Buddha "more often and their commentaries directly explicate those. This is
obviously not to say that these traditions do not rely on commentaries. For
example, Theravada Buddhism relies on commentaries such as
Buddhaghosa's Path of Purification and Chinese schools tend to emphasize
texts such as the Awakening of the Faith.
33
Nevertheless, monks and
scholars do tend to devote significant efforts to the study of the teachings
attributed to the Buddha as a normal part of the curriculum. Theravadins
tend to read the main silttas as contained in the Majjima Nikiiya or the
Dfgha Nikiiya, whereas Chinese monks often focus their study on a central
sutra such as the Varjarcchedikii, the Lotus or the Avatan:zsaka. 34
The Tibetan curriculum is structured quite differently. Although Tibetans
do read and study the Buddhist sutras, the exoteric teachings that purport to
be Buddha's words, they tend to put less (this is a matter of degree)
emphasize on the words of the founder and more on the systematic study of
their content. All the five or thirteen texts used in the exoteric studies, with
the exception of the are Indian treatises (bstan bcos,
siistra). They are the root-texts that are memorized and explained by fur-
ther commentaries. These treatises do not purport to be the direct words of
the founder but to clarify aspects of his message. They offer systematic
presentations of the founder's teachings in order to facilitate the compre-
hension and practice of followers. Although these texts are not part of the
bkd gyur, the collection of the Buddha's teachings available in Tibetan,35
they are nevertheless canonical, since they are included in the bstan gyur,
the translated treatises. The thirteenth century polymath Bu ston (bu-don)
brings out the authoritative and commentarial nature of such treatises,
defining them as "works that explain the meaning of the Buddha's word, are
33. Bhikkhu NyaI).amoli, The Path of Purification of Bhadantiicariya
Buddhaghosa (Boston: Shambala, 1956, 1976) and Y. Hakeda, The Awakening
of Faith (New York: Columbia University Press, 1967).
34. Bhikkhu NyllI).amoli and Bhikkhu Bodhi, The Middle Length Discourses of
the Buddha (Boston: Wisdom, 1960, 1995); M. Walshe, Thus I Have Heard
(London: Wisdom, 1987); E. Conze, trans., Vajracchedikii Prajiiiipiiramitii
(Rome: Ismeo, 1957); L. Hurwitz, Scripture of the Lotus Blossom of the Fine
Dharma (New York: Columbia University Press, 1976); T. Cleary, The Flower
Ornament Scripture (Boulder: Shambala, 1984).
35. P. Harrison, "A Brief History of the Tibetan bKa' 'gyur," Tibetan Literature,
ed. 1. Cabezon (Ithcaca: Snow Lion, 1995), 39-56.
'44 JIABS 20.1
in accordance with the path for the attainment of emancipation, and are
composed by someone with a nondistracted mind."36
We may wonder about such a choice of curricular material, wqich seems
to be unique in the history of Buddhism. This is not the place for an elabo-
rate exploration of the scriptural background of Tibetan Buddhism which
would be required to answer such a question in any detail. Suffice it to say
that historically the form that Buddhism has taken in Tibet partly derives
from the Indian models that existed at the time (eighth to twelfth centuries)
when Buddhism was adopted by Tibetans. The emphasis on treatise can
also be seen as a way to deal with the tremendous complexity of the
canonical material. In general the Buddhist canon is enormous. The bkd
gyur contains more than a hundred volumes of the teachings that purport to
be Buddha's direct words. Moreover, these teachings are not only numer-
ous, but they often explicitly contradict each other. Confronted with this
mass of teachings, Tibetans have tended to be selective and systematic.
They have prefered the systematic treatment of the material found in the
canonical treatises to the more inspirational but less organized material
found in the bkd gyur.
This organization of the curriculum reflects the unabashedly classical ori-
entation of Tibetan scholastic traditions, their regard for the lost antiquity of
high Indian Buddhist culture. The great Indian treatises, which form the
basis of the curriculum, are considered to be classical by all the schools of
Tibetan Buddhism. Their scholastic educations look on these texts from a
past period (fourth to eighth century C. E.), a period often described as the
"golden age of Indian civilization," as their models in relation to which their
contemporary achievements are measured. For Tibetan scholars, such texts
are classical in the full sense of the word, which is explained by Gadamer
in this way:
~
The "classical" is something raised above the vicissitudes of changing times
and changing tastes. It is immediately accessible, not through that shock of
recognition, as it were, that sometimes characterizes a work of art for its con-
temporaries and in which the beholder experiences a fulfllled apprehension of
meaning that surpasses all conscious expectations. Rather when we call some-
thing classical, there is a consciousness of something enduring, of signifi-
36. Bu ston, lung gyi snye ma, 5, quoted in J. Cabezon, Buddhism and Lan-
guage (Albany: Suny University Press, 1994) 45.
DREYFUS 45
cance, that cannot be lost and that is independent of all the circumstances of
time-a timeless present that is contemporaneous with every other present)7
The great Indian treatises have this timeless and normative status. They are
the obligatory reference points for later reflections. They are the "great
texts" revered by Tibetan scholiasts. They provide the basis and model for
the education of Tibetan scholars, who take them as setting the standards
against which contemporary achievements are measured.
Comparing Curriculums: Commentary vs. Debate
On the side of differences, a striking feature is the number of texts and the
time devoted to the study of each of them in rNying rna and dGe lugs cur-
ricular models. Whereas in the dGe lugs curriculum of Se rwa, only five
texts are studied during a period of fifteen to twenty years, rNam grol gling
monks study at least thirteen texts in half that time. The number of texts is
much greater when we include the tantric ones, which are not counted
among the thirteen texts, and additional the texts covering auxiliary topics.
We may wonder about the reason for this difference. Does it reflect a dif -
ference in the content of the education?
It is true that there are differences in the number of topics covered by the
two curriculums. The auxiliary topics of grammar, poetry, history, etc., are
not covered in the Se rwa curriculum and neither are the tantras, which are
studied privately in the tantric colleges. For the most part, however, the
content of the two curricular models is similar. Both curriculums cover the
same five main topics, albeit quite differently. If we group the different
texts into areas of study, we can then discern five main areas: Madhyamaka
philosophy, logic and epistemology, the study of the path, monastic
discipline and tantra. Let us leave the last topic aside, since it is not
37. H. G. Gadamer, Truth and Method (New York: Crossroad, 1992) 288.
Such a description is adequate only from a phenomenological point of view. It
describes the attitude of the participants in the tradition toward certain texts, but
does not provide an adequate analysis of the cultural reality of these texts.
Despite what Gadamer seems to suggest, there is no necessity in classical texts,
for tradition is contingent. Textual choices come and go and what is considered
classical by one age is forgotten by the next. Tibetan education provides exam-
ples of such changes. In the study of logic and epistemology, DharmakIrti's
Pramiir,ta-viniscaya was first chosen but later replaced by his Pramii1}.a-viirttika
under Sa pru;rs implilsion. Since then, Tibetan scholars consider this latter text as
the classical expression of Buddhist logico-epistemological tradition.
.46 nABS 20.1
officially part of the Se rwa curriculum and is supposed to be studied
privately, and examine the ways the two curriculums cover the first four
exoteric topics. ,
For each topic the Se rwa curriculum tends to focus on a single text,
which is then supplemented by further commentaries and textbooks. The
only exception to this practice is found in the study of the path which. is
done through two texts: the Ornament of Realization attributed to Maitreya
and Vasubandhu's Treasury of Abhidharma (which I would also count as a
study of the path). Even here, however, the textual overlap is only partial,
since the former covers the Mahayana path whereas the latter covers the
Basic Buddhist path. Thus, each topic is really examined through a single
text. By contrast, the rNam grol gling curriculum covers each main area by
examining several of the relevant texts. For example, when the Mahayana
path is covered all five treatises attributed to Maitreya are examined. Simi-
larly, when the Abhidharma is studied, both Vasubandhu's and Asanga's
texts are examined. Thus, the number of texts studied for each area varies,
although the four main areas are similar.
Thus, it is clear that the main difference between rNying rna and dGe lugs
models is not one of content but of educational style or pedagogy. What
we have here are two quite distinct models of scholastic studies. The
dialectical style of the dGe lugs tradition exemplified by the Se rwa curricll-
lum focuses on a few texts and emphasizes the practice of dialectical
debates as one of (and possibly the) central method of education. Whereas
in traditional Indian Buddhism debate seems to have been an occasional
skill used mostly in public, the dGe lugs tradition emphasizes its pedagogi-
cal use as a way to master texts and develop a spirit of inquiry. This peda-
gogical role for debate has led the dGe lugs tradition to focus on dialectical
questions rather than on the more textual and commentarial aspects of
Indian Buddhism. As a consequence, this tradition has tended to limit th'1
textual basis of its studies.3
8
It has also sometimes neglected, especially in
the three monastic universities, the practice of higher literary skills.
The rNying rna tradition, as exemplified by the rNam grol gling curricu- .
lum, on the other hand, is more textual. It emphasizes commentary over
debate, and offers a more rounded education which combines literary as
well as dialectical aspects. Contrary to dGe lugs institutions, which rely
38. The dGe lugs tradition is often praised by outsiders for its dialectical depth
but criticized for its limitations in knowing the fundamental Indian treatises.
Thus dGe lugs scholars are sometimes characterized as having a "limited
[textual] vision" (mthong bya chung ba).
DREYFUS 47
overwhelmingly on the practice of debate, non-dGe lugs scholastic institu-
tions are more moderate in their use of debate as a scholastic pedagogy.
Debate is a limited though important part of their curriculum and does not
constitute the central methodology, as in the dGe lugs institutions. In that,
the non-dGe lugs institutions may be closer to the Indian tradition where
debate seems to have taken place mostly for public performance or in actual
confrontations with other schools.
These two educational traditions are associated with two institutional
forms: the debating institution (rtsod grwa) of the dGe lugs tradition, as in
Se rwa, and the commentarial institution (bshad grwa), as in rNam grol
gling. These two types of institution and the traditions associated with
them have a long history, which we cannot examine at this point in any
detail. Briefly, however, the model of commentarial institutions in Tibet
can be traced back to Sa-pal)., who transformed the Sa skya tradition into
one of the main Tibetan scholarly schools in the thirteenth century. Sa pal).
stressed the role of study in monastic training and proposed a model of
intellectual inquiry which was in many respects close to classical Indian
ideas. Such a model is based on the harmonious combination of three
practices: exposition ('chad), composition (rtsom) and debate (rtsod), as
explained by Sa pal.1's own Entrance to the Gate for the Wise (mkhas pa la
'jug pa'i sgo ).39 In this text, Sa pal). greatly emphasized traditional Indian
commentarial categories as well as their literary background. He stressed
the importance of grammar and semantics as basic scholarly skills and the
relevance of Indian poetics to commentarial practice.
The debating tradition grew out of the scholarly activities of the famous
translator rNgog 10 tsa ba (ngok-lo-tsa-wa, 1059-1109). Despite his
belonging to the bKa' gdams pa (ga-dam-ba) tradition, which in its origins
looked askance at the study of philosophy, rNgog was deeply interested in
scholarly studies, which he promoted in Tibet. Under his influence, Tibetan
Buddhism in general and the bKa' gdams pa tradition in particular became
more philosophically oriented. Under his impulse, the monastery of gSang
phu ne'u thog (sang-pu-ne-wu-tok), founded in 1073 by his uncle rNgog
legs pa'i shes rab (Ngok-lek-bay-shay-rab, one of AtIsa's direct disciples)
started to develop as an active intellectual center. Its importance further
increased with the work of Phya pa chos kyi seng ge (cha-ba-ch6-gyi-seng-
gay, 1182-1251), who brought about important developments due to his
39. See D. Jackson, The Entrance Gate for the Wise (Wien: Arbeitkreis fur
Tibetische und Buddhistische Studien, 1987).
.48 JIABS 20.1
acute and original intellect. 40 Phya pa is credited with settling the form of
debate practiced by Tibetans. It is under his influence that gSang pu
became the center of a tradition that was going to differ from the more
classical Indian model later imitated by Sa-pal).. .
Gradually, the education offered by the gSang pu tradition
41
spread
throughout the Tibetan world. Later scholastic centers such as sNar thang
(nar-thang), Zha lu (sha-lu), and bDe ba chen (de-wa-chen) adopted a cur-
riculum similar to that of gSang pu. It is in these centers that Tsong kha pa
and his main disciples received their basic scholarly training. Conse-
quently, the dGe lugs school adopted the gSang pu tradition with its philo-
sophical views, curriculum, and methods of study. From the fifteenth to
the seventeenth century, a close link existed between the three dGe lugs
monastic universities around Lhasa and gSang pu.
42
After this period,
gSang pu lost its importance as a center of study and was supplanted by the
three dGe lugs monastic universities, which became the dominant scholastic
establishments in central Tibet.
We may begin to understand better the curricular organization of the two
models of Tibetan scholastic education. We realize their important sirnilari-
ties and their more subtle variations, as well as the complex histories that lie
behind them. But our effort of comprehension cannot stop here, for we
need to understand the content of the curriculum. To do so I could describe
the content of each text, but it would be hard to avoid the tedium 6f a
scholastic laundry list. Hence, rather than survey the content of the whole
curriculum, let me focus on a single aspect, the study of the path, in order to
40. L. van der Kuijp describes Phya pa as a non-sectarian thinker mostly asso-
ciated with the bKa' gdams pa. "Phya-pa Chos-kyi-seng-ge's Impact on Tibetan
EpistemologIcal Theory," Journal of Indian Philosophy 5 (1978): 355-369, 3 5 7 ~
41. It should be clear that this label is a simplification, for both rNgog's and Sa-
paIJ.'s traditions coexisted at gSang pu. The monastery was divided between
bKa' gdams colleges, which followed Cha-ba's tradition, and Sa skya colleges,
which probably followed Sa-paIJ.'s model of education. Rong ston (rong-don,
1367-1449), for example, who was one of the foremost proponents of Sa-pal).'s
tradition, taught extensively at gSang pu. See D. Jackson, "introduction," in
Rong-ston on the Prajfiiipiiramitii Philosophy of the Abhisamayiila11J.kara
(Kyoto: Nagata BunshOdo, 1988). Nevertheless, the name is convenient in view
of the later connection between the dGe lugs school and the bKa' gdams ele-
ments at gSang pu.
42. See S. Onoda, Monastic Debate in Tibet (Wien: Arbeitkreis fur Tibetische
und Buddhistische Studien, 1992) 13-36.
DREYFUS 49
clarify some of the central topics, goals and concerns of Tibetan monastic
education.
The Place of the Study of the Path
If we look at the two types of curriculum and the number of texts studied
and years spent on them, we can see that by far the greatest amount of
effort is devoted to the area of studies which I have termed the study of the
path. In the dGe lugs curriculum, this topic is examined through at least
two texts: the Ornament and the Abhidharma. Even Candrakirti's Intro-
duction is largely concerned with the path as well. The importance of the
topic is clear also in the number of years spent on each of these texts, par-
ticularly on the former, which is studied for four or five years at Se rwa
through an elaborate textual examination always combined with lengthy
debates. It is studied with Tsong kha pa's Golden Garland, 43 and rGyal
tshab's (gyel-tsap, 1364-1432) Ornament of the Essence of Commen-
taries
44
as well as with the textbooks of the college. In this topic, the text-
books are important because they allow the students to cover topics that are
not explicitly covered by the Ornament. Students, who have alreadyexam-
ined the Abhidharma topics in their study of the Ornament, examine them
again when they study Vasubandhu's commentary on the Abhidharma for
43. Tsong kha pa, Extensive Explanation of the Treatise of the Ornament
Together with its Commentaries, a Golden Garland of Good Sayings.(bstan
beos mngon rtogs rgyan 'grel pa dang beas pa'i rgya eher bshad pa legs bshad
gser gyi phreng ba, Bylakuppe, India: Sera Monastery. Block). The use of this
book in the dGe lugs tradition has given rise to a lot of controversies. Despite its
being authored by the founder of the tradition, many dGe lugs scholars prefer to
rely on rGyal tshab's work or on textbooks. This choice is often questioned by
thinkers outside of the dGe lugs tradition who snear at the refusal of many dGe
lugs scholars to use the book of their founder. dGe lugs scholars, however,
justify their choice by the fact the Golden Garland was written when Tsong-
kha-pa was thirty one and had not yet reached his maturity. Hence, it cannot be
taken as reflecting a mature dGe lugs standpoint, they argue. There is some truth
to this. Tsong-kha-pa's large work appears to be a compendium of commonly
accepted opinions concerning the Ornament and it reflects a variety of views,
which are not all compatible with Tsong-kha-pa's later views. Nevertheless, it
contains also some insightful explanations and several dGe lugs teachers hold
that it is impossible in this tradition to claim to know the Ornament and its
literature without mastering the Golden Garland.
44. rGyal tshab, Ornament of the Essence of Commentaries (rnam bshad sny-
ing po rgyan; Varanasi: Pleasure of Elegant Sayings Press, 1980).
.50 nABS 20.1
two to four more years. Thus, altogether dGe lugs students may spend
close to ten years examining the path.
In the rNying rna curriculum of rNa..'1l grol gling, the time devoted to the
study of the path explained in the exoteric literature is shorter since the
overall exoteric curriculum does not take more than six or seven years.
Nevertheless, the topic is covered in considerable detail. True to its textual
methodology, the rNying rna tradition exposes the students to this toplc
through the study of many texts: at least three of the five treatises attributed
to Maitreya are clearly devoted to the study of the path and so are the two
Abhidharma commentaries as well as Santideva's Introduction to the Bod-
hisattva's Deeds, which is used as an introductory text.
One may wonder why this topic of the path is covered so extensively in
both types of curriculum? To those who are experts in a Buddhist tradition,
the answer to such a question is self-evident. The path (lam, mtirga) is the
central notion of the tradition. As expressed by Buswell and Gimello, the
path "incorporates, underlies, or presupposes everything else in Buddhism,
from the simplest act of charity to the most refined meditative experience
and the most rigorous philosophical argument. The study of marga directs
attention ... to a general pattern of discipline encompassing both the whole
life of the individual and the corporate life of the whole Buddhist commu-
nity."45 Scholars of Buddhism know that the study of a particular fonnula-
tion of the path plays a central role in a Buddhist tradition. It is the struc-
ture around which a Buddhist tradition organizes its practices, its main
doctrinal teachings, its central narratives, etc.
For those who have little expertise in a Buddhist tradition, this focus on
the path may appear alien, requiring the substitution of the well known
terms of religious studies with arcane Buddhological jargon. We should
first notice, however, that the Buddhist literature dealing with the path is
extremely Jtequent throughout the Buddhist world. Many other c l a s ~ i c a l
Indian treatises, such as those attributed to Asariga himself, fit in this class.
In Tibet, there is a whole literature expounding this topic: the numerous
commentaries on the Prajfiti-ptiramitti literature, the studies of Stages and
Paths (sa lam gyi mam bzhag) of the sutra and the tantra, the texts devoted
to the structure of the path in the traditions of the Great Seal and of the
Great Perfection. Outside of India and Tibet, such texts are. also
widespread. In Theravada, Buddhaghosa's Path of Purification is only the
45. R. Buswell and R. Gimello, introduction, Paths to Liberation (Honolulu:
Hawaii University Press, 1992) 6.
DREYFUS 51
most famous example of an extensive literature. Similarly, such texts have
also played an important part in Far Eastern Buddhism, as evinced by the
importance of Chih-i's (538-597) Mo-ho-chih-kuan.
46
The impression of unfamiliarity further dissipates when we begin to real-
ize that this classical Buddhist standpoint can be recast in terms of an
emphasis on practice. Too often religious traditions are defined in terms of
creed, an approach that is far from being as universal as it may seem.
Although such a view has some applicability to Buddhism, I would argue
that it is basically inappropriate to a tradition that emphasizes practice as its
central focus. This does not mean that doctrines, symbols or narratives are
irrelevant to Buddhism, as Buddhist scholars know, but that they need to be
understood in terms of how they relate to actual practices.
When we realize that the idea of the path is the way in which Buddhism
expresses its pragmatic and soteriological emphasis,47 we begin to under-
stand why students spend so many years in studying the structure and
result of the path. We have yet to understand, however, the way in which
such studies relate to actual practices. It may be tempting to assume that
texts dealing with the path directly relate to actual practices, in particular to
the meditative practices that are normatively speaking central to the tradi-
tion. I would like to argue that this assumption is warranted, however, only
to a very limited extent. I would further argue that although practice is cen-
tral to Buddhist traditions and the various treatments of the path are meant
to address this pragmatic emphasis, it is a mistake to assume that teachings
on the path necessarily reflect an experiential standpoint.
Recently, R. Sharf has argued in the same sense. His view is that some
modem Buddhist scholars and contemporary Buddhist practitioners mis-
takenly regard the literature describing the structure and results of the path
in experiential terms. Sharf says:
In fact, it is difficult to imagine how somebody could mistake this kind of
religious literature for "expressions" or "reports" of personal experiences; they
are first and foremost scholastic compendiums, compiled by monks of
46. See J. McRae, "Encounter Dialogue and the Transformation of the Spriritual
Path in Chinese Ch'an," Paths to Liberation, eds. R. Buswell and R. Gimello
(Honolulu: Hawai University Press, 1992) 339-370.
47. The soteriology normatively emphasized by Buddhist traditions is best
described, following J. Z. Smith's useful distinction, as utopian rather than loca-
tive. "The Wobbling Pivot," Map is not a Territory (Chicago: University of
Chicago Press, 1993) 88-103.
52 nABS 20.1
formidable learning who were attempting to systematize and schematize the
confused and often conflicting descriptions of practices and stages found scat-
tered throughout the canon. 48
For Sharf, it is a categorical mistake to assume that the literature dealing
with the path is either a reflection of Buddhist practice or a direct
preparation for it. My point here is not to discuss Sharfs arguments, which
address the presentations of many traditions and thus may have to be
nuanced. Nevertheless, I believe that his view is quite appropriate in the
case of the Tibetan presentations of the exoteric path that are central to
scholastic education, especially the presentations derived from the works
attributed to Maitreya and AsaIiga. Let me elaborate this point, before
making a few broader concluding remarks.
The Study of the Path and Buddhist Practice
Among the canonical works concerning the exoteric path, the one that
stands out is the Ornament, which is attributed by Tibetan scholars to the
celestial Bodhisattva Maitreya. This work is studied for often up to six
years in dGe lugs institutions and, although less time is devoted to it in
non-dGe lugs institutions, it remains a central reference of the Tibetan pre-
sentations of the path. Thus, it constitutes an ideal testing ground to see
whether Sharfs view applies to the Tibetan presentations of the path.
The Ornament is a commentary on the Prajfiii-piiramitii-sutra, the main
canonical source of the teaching of emptiness. The primary concern of this
commentary is not, however, to explain this teaching but to delineate the
stages of the path from the Mahayana standpoint, a subject taught only
implicitly in the sutra, according to the Tibetan tradition. Tibetan scholars
describe the topic of the Ornament as the stages of realization that a r e ~ t h e
hidden meaning of the sUtra (smdo'i sbas don mngon rtogs kyi rim pa).
The Ornament summarizes its own content in this way:
The perfection of wisdom (prajfiii-piiramitii) has been proclaimed by way of .
eight themes: 1) the wisdom knowing all modes, 2) the wisdom knowing the
paths, 3) the wisdom knowing all [phenomena], 4) the full practice of all
48. R. Sharf, "Buddhist Modernism and the Rhetoric of Meditative Experience,"
Numen 42 (1995): 228-283, 238.
DREYFUS 53
aspects,- 5) the culminating stages of practice, 6) the gradual practice, 7) the
instantaneous practice, 8) the dharma-body. 49
Each of the eight chapters of the Ornament addresses one of the eight
themes (dngos po, padiirtha). There is no point here in analyzing these
eight themes. Suffice it to say that the Ornament describes the structure of
the Mahayana path through the four practices (sbyor ba bzhi, catviira/:t
prayogii/:t) or realizations (mngon rtogs, abhisamaya). These four realiza-
tions (chapters 4-7) take as their objects the first three themes (chapters 1-
3), the three wisdoms of the Buddha. The result of this fourfold practice is
the dharma-body of the Buddha and his special attainments (chapter 8).
I suggested earlier that the importance of the path in Buddhist tradition
reflects a pragmatic orientation on the part of the tradition, which under-
stands what would be called in English religion more as a matter of practice
than of creed. It is tempting to infer from this that since it teaches the
Mahayana path, the Ornament must bear a direct relation with actual
Mahayana meditative practices. It is also tempting to infer the since this
text explains the Mahayana path, those who study it intensively, as Tibetan
scholars do, must be interested in this text for practical reasons. These
assumptions are, however, unjustified. Although practice is central to
Buddhist traditions and the Ornament relates to this pragmatic emphasis, it
is incorrect to assume that teachings on the path necessarily reflect an expe-
riential standpoint. In order to understand a text we cannot look just at its
content and deduce from this its application; rather, we must consider the
ways in which such a text is used by the textual communities in which it is
embedded.
49. shes pha rol phyin pa ni I dngos po brgyad kyis yang dag bshad Imam kun
mkyen nyid lam shes nyid I de nas tham cad shes pa nyid Imam kun mngon
rdzogs rtogs pa dang I rtse mor phyin dang mthar gyis pa I skad cig gcig
mngon rdzogs byang chub I chos kyi sku dnag de rnams brgyadll
prajffaparamita-$.tabhi!; padarthai!; samudfritiil sarviikarajffatii (1) margajffata
(2) sarvajffata (3) tata!;11 sarvakarabhisa7rJ,bodho (4) miirdhaprapto (5)
'nunpiirvika!; (6) I (7) dharmakiiyas (8) ca te '.f.tadhii II
E. Obermiller and Th. Stcherbatsky, Abhisamayala7rJ,kara-nama-prajfiiipiira-
mitopade.sa-sastra, The Work of Bodhisattva Maitreya (Leningrad: Bibliotheca
Buddhica, 1929; reedited Osnabrock: Biblio, 1970), stanzas 1:3-4, P: 5184, Ka,
1.a-I5.b, 1. This work has been translated by E. Conze, Abhisamayala7rJ,kara
(Roma: ISMEO, 1954). For a still unmatched study of the content of this work,
see E. Obermiller, Acta Orientalia 11 (1933): 1-100. Translation, which is mine,
is based on this last work.
54 JIABS 20.1
In discussing the ways this text is used by Tibetan traditions, we may
want to keep in mind the fact that the Ornament is used differently by the
two main Tibetan scholastic traditions characterized above. In the non-dGe
lugs commentarial institutions, the Ornament is studied for its content, the
eight themes, which are explainedthrough seventy topics (don, artha). In
this way, students learn about the four realizations, the bodies (sku, kaya)
of the Buddha as well as a number of elements of the Mahayana path such
as the mind of enlightenment (byang chub kyi sems, bodhicitta). Non-dGe
lugs traditions do not focus exclusively on the Ornament, but complete this
study of the path by examining the other texts attributed to Maitreya as well
as AsaIiga's and Vasubandhu's Abhidharma texts. .
dGe lugs monastic universities proceed differently. They take'the Orna-
ment as the central text for the study of the path, treating it as a kind of
Buddhist encyclopedia, and read it in the light of commentaries by Tsong-
kha-pa, rGyal tshab and the authors of textbooks. Sometimes a single
word of the Ornament is taken by commentaries more as a pretext for elab-
orate digression than as an object of serious textual explanation. Several
dGe lugs colleges, such as the Byas (jay) College of Se rwa, recognize this
situation and consider these topics as special (zur bkol). 50 They are studied
in relation to the Ornament but apart. In this way, most of the topIcs rele-
vant to the Buddhist path, whether from a Mahayana perspective or from a
more general basic Buddhist standpoint, are covered in the course of
studying this one text. The summarizing commentaries of the textbooks,
particularly the General Meaning (spyi don), are here helpful in offering
synthesized presentations of all the relevant topics. In this way, students
are introduced to a variety of topics and perspectives, despite the limitations
of their textual basis. f
When We look at the ways in which both these Tibetan scholastic tradi-
tions use the Ornament, we see very little practical relevance, despite some
claims by members of the traditions themselves. Among the topics either
directly covered by the Ornament or studied in relation to it, few appear to
have any direct relation to practice. Let us flrst look at the central themes of
the text. Among the eight topics the first three, the three wisdoms of the
Buddha, are not meant to be practiced directly. They are taken as the object
50. In some colleges, some of the central topics such as tranquility or the dis-
tinction between interpretable and definitive teachings are considered separate
topics. They have special texts devoted to them and in Se rwa Byas are studied
apart, usually the year before fInishing the Ornament. The other colleges do not
have a special time devoted to them, but they do have special texts.
DREYFUS 55
of the path, which consists of the four practices. Similarly, the last theme,
the dharma-body of the Buddha is not directly relevant to practice but is the
goal of practice. The central form of practice presented by the Ornament is
the four practices or realizations, particularly the practice of all the aspects
(rnam rdzogs sbyor ba), the topic of the fourth chapter. In fact, this is the
central topic of the text and may have been an actual practice in which all
the different aspects of the three wisdoms are summarized in a single medi-
tation called the meditation summarizing the three wisdoms (mkhyen gsum
bsdus sgom). This is not the place to explain this highly technical topic
which would take us into the stratosphere of Tibetan scholasticism. What
is relevant for our purpose is that this practice seems to be realistic. It does
not involve any extraordinary feat, as in the case of the miraculous qualities
of the Buddhas and Celestial Bodhisattvas, but can be implemented by any-
body interested in doing so.
But, and this is the important point, no teacher I have ever met, seems to
have practiced this meditation or even to have been clear on how to do so.
Non-dGe lugs curriculums do study this practice but few seem to have a
convincing understanding of this topic, even at the textual level. As far as
the students I interviewed, they seem to have gotten very little out of the
study of this part of the text. Among dGe lugs scholars, there is probably a
better understanding of the topic at the theoretical level. Nevertheless,
nobody I encountered seems to be clear about the ways to practice this text.
Thus, it is clear that in the Tibetan scholastic traditions, the central themes
of this text are not practiced. What about the other auxiliary topics, those
that are briefly presented by the text or those that are studied through other
texts?
It may seem that some of the less central topics studied have direct practi-
cal applications. For example, the mind of enlightenment (byang chub kyi
sems, bodhicitta) is studied in the first chapter. Similarly, the single-pointed
concentration that leads to the attainment of tranquility (zhi gnas, samatha)
is studied in great detail. Concentration is studied with considerable care
for several months, and in certain colleges such as 'Bras spung sGo mang
(dre-bung-go-mang) and Se rwa Byas is considered a special topic (zur
bkol). Thus, topics such as the mind of enlightenment or concentration,
which are of practical importance, are studied at great length. Moreover,
teachers do point out the practical importance of studying them. Are these
not signs that these texts are used for practice?
Although it is tempting to assume here an experiential relevance, the real-
ity appears to be quite different, for the study of these topics remains
mostly confined to the theoretical domain. Students do not devote much
56 JIABS 20.1
time to the study' of the aspects of these topics that are of direct relevance to
actual meditation. For example, in the study of concentration, the nine
stages leading to tranquility, which are of practical use, are not given much
attention. Similarly, the two methods for generating the mind of enlighten-
ment, 51 which are central to the Gradual Path (lam rim) literature, are barely
mentioned. The real focus is theoretical. The mind of enlightenment is not
studied here as an attitude to be developed but in function of its role in the
overall Mahayana path. Similarly, the study of concentration focuses on
the attainments of the four absorptions (bsam gtan, dhyiina) and the four
formless concentrations (gzugs med, arupa).52 These are standard f ~ r m s
of Buddhist practice which have been and are practiced in certain Buddhist
traditions. Nevertheless, they are rarely practiced in the Tibetan tradition.
When monks become really serious about practice and start the type of
extended retreat that would enable them to aim for such attainments, they do
not practice the four absorptions or the four formless concentrations, but
focus on the tantric path. There, the attainment of tranquility is discussed
for which special methods are introduced, 53 but the attainments of absorp-
tions and formless concentrations play little role.
Thus, the conclusion seems hard to escape. Despite claims to the con -
trary sometimes made by members of the tradition, the study of the "Orna-
ment and other texts similarly presenting the exoteric path seems to have
little relation to experience within the context of Tibetan scholastic tradi-
tions. We may then wonder why Tibetan scholars spend so much time
studying these topics? Are they taken in by their own claims? Or do they
just keep studying texts that had an experiential relevance in an earlier time,
which is now lost? I have obviously little to say about the historical back-
51. The tradition of the Gradual Path often speaks of two methods to develop
the mind of enlightenment: the fIrst, the seven causes and effect, is based on con-
sidering the debt we owe all sentient beings for their having been our mothers
and having had countless other kindness. The second, exchanging self and oth-
ers, focuses on the equality of self and others and proposes an exchange of one's
attitudes toward oneself and others. See Geshe Rabten, The Essential Nectar
(London: Wisdom, 1984) 305-66.
52. The dGe lugs views on this topic have been well presented by L. Zahler,
Meditative States (London: Wisdom, 1983), and Geshe Gediin Lodra and J.
Hopkins, Walking Through Walls (Ithaca: Snow Lion, 1992). For a detailed
Theravlida view on the topic, see Nlinamoli, The Path of Purification, 1.84-478.
For an easier view, see A. Sole-Leris, Tranquility and Insight (Boston:
Shambala, 1986) 56-73.
53. See D. Cozort, Highest Yoga Tantra (Ithaca: Snow Lion, 1986) 55-6.
DREYFUS 57
ground of this last question, for the way in which these texts were used by
Indian Buddhists is outside the purview of my inquiry. It is important to
remember, however, that understanding the practices of a tradition as left-
overs of a meaningful past that has lost its relevance is inadequate. People
engage in the lengthy study of such texts not out of habit but because they
find it meaningful. But what is the meaning that Tibetan scholru::s find in a
text such as the Ornament?
Worldview and the Study of the Ornament
I would like to suggest that the answer is not to be found in experience but
in what could be described as the formation of a worIdview. The discus-
sion of the exoteric path is central to Tibetan traditions not because it pro-
vides practical guidance but because it provides for the construction of the
kind of meaningful universe that Buddhist practice requires. This explana-
tion of the role of the Ornament follows a venerable tradition in the West-
ern academical study of religions, which proposes that religion is a way to
understand the universe and cope with the limits that it imposes on humans.
Some of the formulations ofthis view, such as those of Tylor and Frazer,
are by now thoroughly discredited. They were clearly wrong in presenting
religion as a kind of primitive science aiming at the explanation of natural
phenomena. Even more recent and relevant formulations of this view are
still problematic in that they reflect too closely the theological background
out of which they come. Weber, for example, holds that the religions of
salvation are based on a theodicy of suffering and happiness. 54 Similarly,
Geertz argues that religion is a model both of and for human existence. It
enables humans to bear existential problems such as suffering or evil by
placing these experiences within a meaningful framework. 55 Although not
without merit, these views in which the Protestant influence is transparent
fit Buddhism only imperfectly, for the latter is based on the rather optimistic
idea that humans can overcome suffering. Hence, the idea of acceptance,
which is central to Weber, Geertz and many modern scholars of religious
studies, is problematic in a Buddhist context. Nevertheless, it is certainly
not wrong to argue that a religion such as Buddhism seeks to enable its
followers to cope with suffering and the other limits of human existence.
54. M. Weber, "The Social Psychology," From Max Weber, eds. H. Gerth and
W. Mills (New York, 1958) 271-275.
55. C. Geertz, "Religion as a Cultural System," The Interpetation of Cultures
(New York: Basic Books, 1973) 87-125,100-5.
58 JIABS 20.1
In a Buddhist perspective, this coping with suffering, which is the goal of
the tradition, has several dimensions. First and foremost, Buddhist tradi-
tions hold that only sustained religious practices can effectively help
humans to diminish and eventually overcome suffering. Such liberative, or
to use J. Z. Smith's words,56 utopian practices involve a whole range of
soteriological practices. Most of them have little to do with meditative
experience and pertain to what is usually called merit making. In this cate-
gory, we can include not only most traditional lay practices such as giving
to the monastic order but also most of the monastic practices as well. In
particular, the scholastic studies examined here are understood by partici-
pants as a form of merit making. This type of Buddhist practice fornis the
core of much actual Buddhist practice. It should not be considered at odds
with so-called higher meditative practices, but, on the contrary, as continu-
ous with them. Merit making is part of the liberative or utopian dimension
of the tradition. In some ways, the value that monks find in monastic stud-
ies derive from their being meritorious. Studying a text such as the Orna-
ment is intrinsically valuable. It is in and of itself virtuous.
Nevertheless, this intrinsic virtuous quality of Tibetan scholastic studies
is not their main value. Normatively speaking, the main value of
one of the two types of activity in which Buddhist monks are supposed to
engage, is in their leading to the development of virtues such as inner calm,
attention and inquisitiveness that will in turn enable the practitioner to be
successful in the higher meditative practices. For there is no doubt that,
from a normative standpoint, meditative practices are considered by most
Buddhist traditions as the ultimate means of freedom. In considering these
higher practices, however, it is a mistake to overemphasize the experiential
dimension. Although Buddhist meditations involve experience, thistis not
their only pr even most relevant feature. From a Buddhist perspective,
meditations are fIrst and foremost ethical practices that seek to develop cen-
tral virtues such as detachment and compassion.
Moreover, ethical practices do not exist independently of larger cultural
frameworks in relation to which they make sense. In particular, Buddhist
practices require a cosmological framework in which the virtues that are
being developed and the practices used for this purpose make sense. Bud-
dhist practices and virtues are supposed to have immediate effects on the
basis of which Buddhist teachers often argue for the cogency of their tradi-
tions. But the immediate benefIts that one derives from certain practices are
56. Smith, "The Wobbling Pivot."
DREYFUS 59
not enough to support the kind of intensive commitment necessary to their
implementation. Humans do not live just by quick fixes but need to decide
on long term goals and means to reach them. They need a narrative through
which they know what to do and become persuaded that they are on the
right track. They also need to be able to bring a sense of closure to such a
narrative, to find a standpoint toward which their efforts are aimed and
from which they make sense. Such a standpoint can be found only in a
certain type of universe. To construct such a universe of meaning is one of
the main goals of the study of the Ornament and other related texts in
Tibetan scholastic traditions.
This universe of meaning is the one familiar to students of Buddhism. It
is explained by the basic teachings of Buddhism such as the four noble
truths and dependent origination, supplemented by the Mahayana sl1tras.
The four noble truths provide the kind of existential analysis of human
existence, as impermanent, suffering and no-self, that can provide the basis
for spiritual practices. These basic teachings also indicate the possibility of
liberation and the path that can lead to such a goal, thus forming a universe
in which the practices recommended by Buddhist traditions become mean-
ingful. The universe of meaning constructed by the Ornament and other
related texts is not, however, just that of basic Buddhism, for it is a
Mahayana universe, where the goal of practice is less self-liberation than
universal salvation. This is the universe of the Mahayana sl1tras in which
bodhisattvas strive to become Buddha through the practice of the perfec,-
tions (phar phyin, piiramitii).
To develop such a view of the world, students go through a number of
topics which pertain either to basic Buddhism or to the Mahayana tradition.
They study the basic teachings mentioned above, including the four truths,
the analysis of mental factors, the difference between concentration and
insight, the form and formless absorptions, etc. In the dGe lugs debating
institutions, these topics, (with which students of other Buddhist traditions,
particularly Theravlida), are familiar, are studied in the textbooks and the
commentaries, which take the Ornament as a pretext for exploring the
Buddhist universe. This is in accordance with this tradition's emphasis on
debate and the concordant tendency to keep the textual basis of studies lim-
ited. In the non-dGe lugs commentarial institutions, such a study is done in
relation to other texts such as those of the Abhidharma.
Students also study the central topic of the Mahayana tradition, the
structure of the Mahayana path, the central topic of the Ornament. Related
topics such as the development of the mind of enlightenment, the nature and
role of the perfections (phar phyin, piiramitii), or the conflicting views on
60 nABS 20.1
Buddha-nature (bde gshes snying po, tathagata-garbha) are examined at
great length. Students also study the divisions and sub-divisions of the
paths, the stages of the Mahayana path, the qualities obtained at each of
these stages, and the final results to which they lead. 57 In this way, the stu-
dents form a coherent picture of the path and the universe in which this path
makes sense.
In the non-dGe lugs commentarial institutions, particularly at the rNam
grol gling monastery, this Mahayana picture of the world is in tum supple-
mented by the study of the tantric path. Right from the beginning, students
are introduced to the tantric dimensions of Buddhist practice. The universe
of meaning constructed here is not just Mahayana, but tantric as well .. Stu-
dents are made aware that the path and the goal are esoteric and that the
exoteric texts figure as introductions to the real path, which is tantric.
These texts are meant to be supplemented by the tantric description of the
path. Thus the last three years out of a total of nine years of study are
devoted to a detailed study of the tantric tradition.
But here again, it would be a mistake to take this tantric curriculum as
reflecting a practical orientation. Students do not receive practical instruc-
tions on how to meditate. Such instructions are provided only after stu-
dents have begun their actual meditative career. Moreover, such instruc-
tions are mostly given only in private or during optional periods of retreat.
Hence, the tantric instructions contained in the curriculum of commentarial
institutions are not intended to provide practical guidance but theoretical
models that support the construction of a universe in which tantric practice
is meaningful. The particularity of the rNying rna curriculum is not that it is
more practical, but that the universe that it constructs is tantric rather than
based purely on the exoteric aspects of the tradition. Thus, the
with dGe lugs curriculum is real but does not concern the actual practices of
either tradition.
The practices of the Tibetan traditions are quite similar, although not
identical. What differs is the rhetoric used to present such practices and the
ideological contexts thus created. In the dGe lugs model, the universe and
the path to which students are introduced theoretically are exoteric and the
actual tantric practices they later engage in are understood to fit into such a
framework. Even while describing actual tantric practices dGe lugs texts
tend to emphasize the primacy of the exoteric narrative of spiritual progress
57. For a brief overview of the literature, see: J. Levinson, "The Metaphors of
Liberation," eds. Cabezon and Jackson, Tibetan Literature, 261-274.
DREYFUS 61
thereby bringing the legitimacy of the classical exoteric model to their eso-
teric practices. In the rNying rna model, the universe to which students are
introduced doctrinally is mostly tantric and the exoteric teachings are taken
as supporting this construction. The actual practices that students later
engage in fit easily into the narratives of spiritual progress derived from
these tantric models. Members of the tradition sometimes find it harder,
however, to justify their practices in reference to the classical Indian model.
Scholasticism and the Construction of Meaning
It is in this ideological and theoretical perspective that the Ornaments dis-
cussion of apparently practical topics must be understood within a Tibetan
context. Topics such as the mind of enlightenment or the attainments per-
taining to the form and formless realms are important not because they
directly prepare for meditations but because they support the elaboration of
a universe in which Buddhist practice makes sense. The Ornament and
similar texts are, for Tibetans, not reports on or direct preparations for
Buddhist practice, but rhetorical representations of the meaningful universe
envisaged by the tradition. They provide students with a meaningful out-
look, which may support further practices, but which has no direct rele-
vance to them.
This construction of a universe of meaning is not something unique to
Tibetan scholastic traditions. Most religious traditions, however, do not
take the doctrinal and intellectualist approach adopted by Tibetan scholasti-
cism. Rather, they emphasize the role of myths and rituals in achieving
such a goal. In the Tibetan scholastic traditions such dimensions obviously
exist but they seem less important than in non-scholastic traditions. Myths
are obviously present but they seem to play a less important role in the
construction of meaning than the doctrinally based narratives. The central
narratives are not derived from the concrete teachings of the founder or the
biographies of the central figures, but emerge from abstract doctrines. This,
I suggest, is a partiCUlarity of scholasticism as a religious phenomenon.
To be successful, this construction of a meaningful universe and the path
that transcends it must become self-evident, so that students feel confident
in their practices. The steps along the path must appear to them as concrete
stages in relation to which Buddhist practice makes sense. This concrete-
ness should, however, be understood in relation to the process of reification
through which it is constructed. The map provided by the Ornament litera-
ture does not refer to some self-evident mental states existing independently
of textuality. The stages described by the Ornament are not set in stone.
Rather, they are constructed symbolic objects that acquire the solidity nec-
62 JIABS 20.1
essary to inspire and sustain people in their actions. They are best charac-
terized, following Burke's term, as symbolic actions, that is, as representa-
tional forces that attempt to influence their audience.
58
Thus, far from being
a kind of guide to Buddhist practice or a description of spiritual experi-
ences, the Ornament provides the Tibetan tradition with the framework that
makes a narrative of spiritual progress possible and introduces an element
of closure without which the commitment required by Buddhist practices
cannot be sustained.
58. K. Burke, A Rhetoric of Motives (New York: Prentice Hall, 1950) 22-3.
ANDREW HUXLEY
Studying Theravada Legal Literature
Two years ago the Journal devoted an entire issue to the subject of Bud-
dhism and Law. I have two excuses for this early return to the subject.
Firstly, I shall deal with areas that were untouched in 1995. I describe the
legal literature produced by the pre-modem Thai, Khmer and Laotian cul-
tures and discuss the strange absence of written Sinhalese legal works.
Secondly, as an encouragement to potential researchers, I append a selective
bibliography which emphasises recent scholarship and the discoveries of
the last twenty years. This article was originally intended to be published
alongside von Hiniiber's A Handbook of Pali Literature. 1 He has provided
a very useful list of addenda to the Critical Pali Dictionary'S numerical list
ofPali texts. 2 I have considered whether to do the same. I have decided
that it would be premature to do so: much more research is needed before
we can make an authoritative list of surviving legal texts. Meanwhile, I
must warn you that none of the information in that portion of the CPD's list
headed 2.9.23 "Law" is reliable.
Few outside the specialist field of Theravada legal history are aware that
the last twenty years have been a Golden Age of discovery. The number of
manuscripts now available in book, microfilm and digital form has
increased a hundred-fold since 1977. But the number of scholars examining
these primary sources remains pitifully small: I know of about half a dozen
colleagues around the world. A similar inbalance between human
resources available and the shelves of manuscripts to be studied afflicts all
branches of Buddhist studies and accounts for much of Buddhism's charm
I apologise to readers who are offended by my grudging use of diacritics. My
aim has been to add diacritics only to Pali words in the titles of books, where
omitting them might make the books harder to locate. Otherwise, because there is
no generally agreed system for transliterating Southeast Asian languages and
Sinhalese, I have omitted all tone markers and diacritics. I should warn Indolo-
gists that several of the books I mention have macaronic titles: typically the first
few words are Pali and the last word Burmese or Thai.
1. von Himlber 1996. See the foreword, v-vi.
2. Ibid., 256-7.
63
64 nABS 20.1
as compared with such over-cultivated fields as the Shakespearian corpus.
But in the field of Southeast Asian Buddhist law this disparity has reached
Churchillian proportions: never in the field of human scholarship have so
few scholars struggled to master so much fresh material. Since about 1977
scholars from Thai, Laotian and universities have boldly gone
where their European predecessors dared not venture. Sponsored by
Japanese research funds and equipped with the microfilm camera, the xerox
machine and the scanner as tools of their trade, they have winkled
manuscripts out from forgotten comers of libraries, from private collections
and, above all, from the book chests of provincial monasteries. to
striking effect. In 1975 only one law text from the Tai Yuan culture of
northern Thailand was in print. By 1989 one-hundred-and-thirty-two such
texts had been discovered and made available on microfilm. Up to 1991
only one Mon dhammathat (the legal genre that provides norms for village-
level dispute settlement) had been printed. Now eleven additional Mon
dhammathats have been published in a handsome Mon-English edition.
Before 1985 only three or four of the Burmese kings' rajathat of
State, one or two of which promulgate something like a royal code o(good
citizenship) were in print: 1990 saw the publication of the tenth and final
volume of Than Tun's magisterial Burmese-English collection .. Before
1993 our knowledge of Siamese legal literature was limited to the Three
Seals Code of 1805, which is an edited edition of that portion of the
Ayuttbayan palace legal archives which survived a particularly destructive
Burmese raid. Pitinai Chaisaengsukkul has now copied and cat<ilogued
about seven-hundred legal manuscripts from central and southern Thailand.
Meanwhile Sarup Ritchu has revealed that the Southern Kingdom based on
Nakhon Sri Thammarat had its own distinctive tradition of law "texts.
According to rumour, we shall soon be able to say the same about the Tai-
Shan and Tai-Khamti cultures.
This explosion of legal source material has coincided with a shift in aca-
demic fashion towards topics for which the Theravada legal literature is our
prime source. Interest (particularly in western history departments) is
switching from "dates and dynasties" towards the structures of everyday
life. Topics such as the economics of the rice field, the relationship
between the sexes and the complexities of social status are working their
way to the top of the agenda. Since the chronicles concentrate on palace life
and since popular story-telling was set in a never-never land derived from
the Jataka, the legal literature is our only source for the details of everyday
life. You're interested in freshwater fishing? Here's a description of the
twenty different types of rod and net used by 18th century Burmese fish-
HUXLEY 65
ennen.3 You want to know how much a rhinoceros horn sold for in Pnomh
Penh in 1853? You can have prices for six different grades and an explana-
tion of how to tell them apart.
4
What would you do in central Laos if you
saw a star-shaped piece of bamboo bark by the side of the trail? The Code
de Vientiane tells us that this warns of an animal trap nearby.5 Whether
you're studying the population statistics of Burma or pre-modem Southeast
Asian marital relationships, it helps to know what they "knew'" about the
causes of barrenness in a couple:
The cause of conceptual gloom is prior bad kamma or doom, or a virulent
worm has enfeebled his sperm, or abortions have damaged her womb. 6
In short an enormous amount of new, potentially interesting source material
awaits anyone who can read "late" Pali or one of the Southeast Asian ver-
naculars. The Theravada legal literature has at last been recognised as too
important to be left to us legal historians. I am delighted by these develop-
ments but would add a single note of caution for non-lawyers. You cannot
treat a 17th century law text from Cambodia as if it were a 20th century law
text from Connecticut: they embody different assumptions about the func-
tions of writing down law. A list of these differences and further wise
advice about the use of pre-modem, non-western legal materials may be
found in a twenty page book review by Bernard Jackson available in any
law library. 7
LAW TEXTS IN SRI LANKA
Sri Lanka has not shared in the exciting discoveries of the last twenty years.
Indeed after two hundred years of unsuccessful search for Sinhalese Bud-
dhist legal literature, it would seem that it is non-existent. This claim has
two aspects, both of which are controversial. The frrst-that no Sri Lankan
legal literature has survived---depends on our assessment of a mysterious
text called the Niti Nighanduva. The second-that there never was any Sri
Lankan legal literature-is harder to prove but is of great theoretical inter-
3. Richardson 1847, 375; Manugye dhammathat XII 3. Full citation of works
included in the bibliography are given in an appendix.
4. Leclere 1898,2:313-347 Kram Prohmotont.
5. Raquez 1902, 424; kamphi phra thammasat buhan.
6. U Gaung 1909, 2: 238; Manucittara, an obscure 18th century verse
dhammathat.
7. Jackson 1975.
66 JIABS 20.1
est. We lawyers think that no literate culture would forego the blessings of
written law:
... production of codes, the use of written fonnulae in legal proceedings
... Some, if not all, of these consequences will certainly follow a
widespread use of writing. 8
Yet quite possibly Serendip, the fortunate island, has avoided producing
legal literature throughout its history. Since 150 BC Ceylon has been as lit-
erate and cosmopolitan as anywhere else in Eurasia. If Sri Lanka, which
harboured ships from Greece and Arabia to the west, China and the Spice
Islands to the east, did remain lawless, it was surely not through ignorance
of the concept of written law. Rather, it was because the needs that written
law meets elsewhere in the world were nonexistent or otherwise satisfied in
Lanka. The Southeast Asian experience indicates that Theravada belief and
practice can coexist with written law for the laity. Sri Lankan Theravada
has one distinctive attribute which is lacking in Southeast has
learnt to coexist with a modified version of the Indian caste system. The
entire Buddhist population of Ceylon is classified by caste, in
Southeast Asia caste is only of concern to kings and untouchables at either
extreme of the social ladder. Is there a sense in which caste as a form of
social organisation filled up the niche which written law might have
occupied?
The Mahavamsa contains a few legal references which have been cited as
proof that certain kings wrote down law. I am very grateful to Steven
Collins for reexamining them with an eye to their legal implications. Ipse
his unpublished translations o(the key passages. King Udaya I (797-801
AD) "had cases which were rightly decided set down in writing and kept in
his palace, because he feared that decided cases might be reopened [through
bribery]" (49.20-1). This would appear to be a register of judgements, a
royal depositary of judicial records. "Legal literature" is a flexible term, but
I would prefer to restrict it to texts which achieve some public circulation
outside the palace archive. General Ayasmanta, who acted as regent in the
early 13th century "organised and separated the four varnas which had
become mixed, and being intent on the good, had a treatise written contain-
ing questions about the dhamma" (80.41). The word "dhamma" need not
indicate a written collection of legal norms. The sentence makes good
enough sense as a description of two different ways in which Ayasmanta
8. MacConnack 1976, 229-248, 233.
HUXLEY 67
behaved meritoriously: he reformed social organisation and he sponsored a
book on religion. A 15th century poem, the Giva Sandesaya talks of "High
Ministers who having learnt the art of deciding cases after obtaining a
knowledge of ~ e various laws of the kings, lovingly look after the entire
people of Lanka ... " 9 I read this as further proof that a royal legal archive
was kept in the palace, to which library tickets were reluctantly issued in the
name of such of the king's functionaries who could establish "a need to
know." In the 16th century the island was divided into three Theravada
kingdoms, which succumbed in tum to European force. All we know
about law in the more cosmopolitan coastal kingdoms of Kotte and
Sitawake, which fell early to the Portugese, comes from ethnology: a 17th
century account of law in these Buddhist port kingdoms was written by de
Costa and has been translated into English.
lO
But note the frustrating
remark made by L. Prera in Colombo in 1744, which hints that some
Sinhalese legal literature was known to him: "I have never seen such an
oath. Nor have I seen such prescribed in any Sinhalese book on laws,
though I have read many." 11 Stronger evidence that Ceylon once had a
iegalliterature comes from Burma. An 18th century list of nine old dham-
mathats known in southern Burma includes "The Sinhalese Edition, com-
piled in Sri Lanka"l2 while another Burmese list of Collected Royal Judge-
ments in manuscript includes the Elaya Kyemin Pyatton, named after the
famous Tamil invader of Sri Lanka.!3 Since neither work has survived, we
can only guess what these references mean. Pyatton can refer to folk-tales
as well as to law, so the latter text might have been a popular historical
romance based on the rise and fall of King Elaya. One text of Sri Lankan
law that definitely did circulate in southern Burma was Parakramabahu 1's
Katikavata-his supplemental vinaya code. Could this possibly have been
known as "The Sinhalese Dhammathat"?
We have a surviving text, the Niti Nighanduva, which purports to be a
code of Buddhist law. Whether it should be regarded as genuine depends
on the verdict we reach in the case of R v John Armour. The Niti Nighan-
duva conveniently fulfilled 19th century needs and expectations. English
administrators charged with the task of governing the kingdom of Kandy
after 1815 needed a written statement of applicable law: the Niti Nighan-
9. Jayasekare 1969,273.
10. At the order of Sir Alexander Johnstone, Chief Justice (1809-1817). It is
held in the Public Record Office, London (C.O. 54/123-124).
11. Quoted in Kotelawale 1987, 112-120.
12. Huxley 1996, 93-5.
13. Forchhammer 1885,76. Htin Aung 1962, 38.
68 JIABS 20.1
duva was as a descriptien ef law "as it existed in the last days ef
the Kandyan kingdem." Was it written fer an English readership by seme-
ene who, knew what was required? Jehn Armeur stands accused ef ferg-
ing the Niti Nighanduva. He was well placed to, act as a middle-man
between Kandyan culture and the English celenists: having been expelled
frem his father's heuse in Celumbe fer distributing anti-Christian literature,
he married a Sinhalese weman and lived with her, native-style, in Kandy.
He wrete and speke better Sinhalese than any ether Eurepean and had
already published articles en the tepic ef Kandyan law in English. Specific
decumentary evidence against him exists: a manuscript has survived which
leeks like a reugh draft fer the fITst two, chapters ef the Niti Nighanduwi. It
was prebably written by a Eurepean, since it uses Reman numerals, and its
Sinhalese script resembles Armeur's Sinhalese handwriting. But dees all
this add up to, fergery? Assuming that Armeur was the auther, I prefer to,
regard his efferts as a repackaging exercise. He wrote dewn the bral wis-
dem ef his infermants (the menks and bureaucrats ef Kandy) in ferm
which the English autherities weuld find mest palatable. The epe1)fng per-
tiens ef the werk, centrasting "the dispensers ef justice" who, are interested
enly in written law with "the Kings ef Lanka who, knew what was legal and
what was net," make it fairly plain that it was written after the British cen-
quest in 1815. Armeur's sins, if any, are these ef "inauthentic cultural
appropriatien" rather than fergery. The language ef pelitical correctness is
mere apprepriate than the language ef criminal law. By adepting a
Sinhalese first persen veice fer ene ef his expesitiens ef Kandyan law,
Armeur ceuld be accused ef having ripped eff a marginalised cultufe. But
if we cendemn him new fer falsely pretending to, be a Sri Lankan, we ceme
dangereusly clese to, the judgement ef his centemporaries that Armeur had
"gene native." The Niti Nighanduva ferced the British to, take Kandyan law
serieusly:/'had it net existed the westernisatien ef Sri Lankan law weuld
have gene further faster. Objectively, therefere, Jehn Armeur was a friend
rather than an enemy to, Kandyan culture. The Niti Nighanduva is an
attempt, by an Englishman, to, invent Sinhalese legal literature. This leaves
eur inquiry in an unsatisfactery state. Genuine legal literature may er may
net have ence existed. But nene ef it has survived.
THE EARLY DEVELOPMENT OF

That legal literature is absent frem Sri Lanka but abundantly present in
Theravada Seutheast Asia suggests that the latter culture is net a carben
HUXLEY 69
copy of the former. Because Southeast Asian literature makes a virtue of
its fidelity to Sri Lankan models, this truism needs to be emphasised. By
treating the Mahavihara tradition of Ceylon as their touchstone of ortho-
doxy, Southeast Asian authors invite us to conclude that they merely imitate
the Sinhalese sages. Legal literature is the largest and most important of the
non-Sinhalese gemes and therefore the best argument for the proposition
that intelligent Theravada life can exist outside Ceylon. Early western
commentators took Southeast Asian claims at face value and assumed that
Southeast Asians were too unoriginal and stupid to have written the law
texts themselves. This view is no longer tenable: it is virtually certain that
the legal literature has a local origin. When and where in Southeast Asia
did it emerge, and to what extent did it incorporate written material from
elsewhere?
Epigraphic evidence will help answer the first question. The inscriptions
of Angkor reveal sufficient about classical Khmer law that we can eliminate
it from our inquiries: there is no evidence of any Buddhist legal influence in
Cambodia until the 17th century. Inscription 38 from Sukhothai, which
dates either from the middle or the end of the 14th century, makes several
clear references to the law contained in dhammathat and rajathat: it shows
that Buddhist legal literature was well-established in central Thailand before
1400. King Klacwa's Edict on Theft, preserved in several inscriptions from
Pagan, takes us two-hundred years further back. Promulgated on May 6th,
1249, it shows typically Buddhist legal features: it contains earnest discus-
sion of the interaction between kamma and royal punishment and quotes
directly from the Tipitika. On epigraphy alone we could conclude that
Buddhist legal literature was invented in Pagan in the 13th century and
spread eastwards into the Tai and Khmer cultures. But the manuscript evi-
dence tells a different story: several early law texts are ascribed to particu-
lar cities of first millenium Southeast Asia. In my previous contribution to
the journal I explained why I regarded these claims as plausible and why I
preferred to think of Buddhist legal literature as having developed slowly in
various Southeast Asian cities through the last half of the first millenium.
Pagan was a clearing house rather than an innovator. It catalysed existing
developments by assimilating the legal traditions of different cities into a
standard model dhammathat which commended itself to the new Thai
monarchies of the late 13th century. My reconstruction of the development
of legal literature highlights the role of port cities like Mrohaung and
Nakhon Sri Thammarat which traded directly with India. Recent work by
Mayoury Ngaosyvathn and Michael Vickery prefers to highlight the Tai-
Chinese connection. When the Tai moved into Laos from their homelands
70 nABS 20.1
on the borders of Vietnam and China they may have brought legal and gov-
ernmental texts with them. Such texts would at first inevitably reflect the
Chinese approach to law and government but gradually Chinese. influence
was displaced by the increasingly confident Theravada culture.
The legal literature drew on at least five sources. Two of these, the Bud-
dhist and Hindu traditions of written law, came from India. One; the impe-
rial approach to organising government, came from China. And two, the
oral custom of the irrigators occupying the vallies and that of the
burn hoe cultivators occupying the hills, were indigenous to Southeast
Asia. The content of the rules on marriage, divorce and the relative status
of the sexes (which vary little as between Thai, Burmese and Khmer "law
texts) are taken from the hoe cultivators of the hills. The rules on land-
ownership, debt-slavery and adoption would have developed along with
Southeast Asian irrigation from the Buddha"s lifetime onwards. Readers of
this journal will be most interested in the flrst category, the Buddhist influ-
ence on legal literature. References to the canon in the legal literature would
take an entire book to catalogue, but I can enumerate the cO:rnnlonest
canonical tropes in two short lists. The Southeast Asian legal literature
very often refers to the following seven stories: the tale of Mahasammata
and the origins of kingship (D iii 80), of the Cakkavatti king and his loss of
the celestial wheel (D iii 58), of Devadatta and his Faustian ambition and
punishment (V ii 202), of Vessantara, the paragon of generosity (Jat.
#547), of Mahosadha, the wisest of the king's wise counsellors (Jat. #546),
of Vidhura, the bureaucrat's bureaucrat (Jat. #545) and of King Adftsa-
mukkha, who would not follow the letter of the law if that led to injustice
(Jat. #257). And the following seven lists are very conunonly incorporated
in the law texts: the three kinds of slave (V iv 224), the four agati or prin-
ciples of justice (J i 176; J iv 105), the four factors which determine
a thing's market value (V -a viii 64), the flve duties of marriage-partners (A
iii 36), the seven kinds of wife (A iv 91; J ii 347), the ten types of female
under male protection (V iii 139) and the twenty-five kinds of theft (V-a
viii 60). The ratio of 4:2: 1 between Vinaya literature, the Jatakas and the
Suttas in this last list can be generalised: the Vinaya pitaka contributes most .
to the legal literature, while the Abidhanuna pitaka is scarcely represented at
all. It is a great mistake to think of the Vinaya as concerned solely with the
minutiae of monastic dress and food: one third of Samantapasadika is
devoted to the discussion of just three out of the two-hundred-and-twenty-
seven rules in the Patimokkha-those concerning theft, murder and unlaw-
ful sex. In the excessive attention he pays to these three issues
Buddhaghosa shares a general legal mind-set. But Vinaya influence on
HUXLEY 71
legalliteiature goes beyond direct quotation into m-eas of genre, training and
legal reasoning. The Vinaya literature was copied, taught and debated by a
specialist group of monks. Such vinaya-dhara monks were the first legal
specialists in Southeast Asia. Their techniques of training and argument
acted as a model for the authors of dhammathat and rajathat and, in some
parts of Southeast Asia, for the emergence of a distinct lay legal profession.
For a brief period following the fall of Pagan the Buddhist kingdoms of
Southeast Asia must have shared a homogeneous legal literature. Then
distinct sub-regional styles developed. From the 13th to the 19th century
the political map of Buddhist Southeast Asia -the royal cities, the dynasties
that controlled them and the amount of land which they effectively con-
trolled-underwent constant, apparently random, change. All attempts to
generalise about the region risk getting bogged down in detail. Happily the
surviving legal literature divides itself into three distinct sub-regions each of
which has developed the Pagan legal inheritance in a different direction.
The western sub-region comprises the kingdoms of Burma. Speakers of
the Burmese language, such as the kings of Ava, Toungoo, Sagaing and
Mandalay, predominate but the land around the mouths of Burma's three
great rivers is Mon territory, and was often ruled by independent Mon-
speaking kings. There were also speakers of Pyu and Arakanese, lan-
guages closely related to Burmese. The Pyu were pioneers of Buddhism
and city life during the fIrst millenium, but lost their cultural distinctiveness
thereafter. The Arakanese kept an independent kingdom on the narrow
coastal strip along the Bay of Bengal until shortly before the European
invasions. The northern sub-region is the land of the Tai. Their two most
important kingdoms were Lanna, centred on Chiang Mai in northern
Thailand, and Lan Xang, centred on Vientiane in Laos. From the 14th
century onwards Theravada Buddhism diffused from these capitals into the
smaller surrounding kingdoms-Keng Tung, the Sipsongpanna (in present
day Yunnan), the Shan States and the piedmonts of northern Burma and
southern Assam. Far to the north-west, in the main valley of Assam, and
far to the east, in the Dien Bien Phu region of Vietnam, were Tai kingdoms
which never converted to Buddhism. The chronicle of the Assamese Tai,
the Ahom Buranji, includes a short legal document which allows some
speculations about the non-Buddhist input to Tai legal literature. The east-
ern sub-region is also dominated by Tai-speakers, but the Siamese king-
doms of Ayutthaya and Bangkok have a more bureaucratic and interna-
tional flavour. Whether they derived this from the remnants of the Angkor
bureaucracy, from independent Khmer cities in central Thailand, or from
regular maritime trading contacts with southern China, is a matter of debate.
72 JIABS 20.1
Also included in the eastern region are khmer-speaking Cambodia, and the
"Southern Kingdom" in the Malay Peninsula centred on Nakhon Si
Tharnmarat. The northem region is land-locked and mountainous. Western
and eastern regions both had ports enjoying access to the international
trading routes. Siam and Phnomh Penh entered into this world of trade
wholeheartedly, while the kings of Upper Burma approached it with some
suspicion. Nonetheless by the early nineteenth century western and eastern
regions had developed equally strong state institutions.
There is an important distinction between studying law texts as literature
and as sources of law. In the former case the literary historian looks at
what gets written while in the latter case the comparative lawyer looks at
how what is written gets used. The literary approach, when applied to
Buddhist Southeast Asia, emphasises similarities. Each of the three sub-
region recognises the same genres of legal literature and identifies each
genre by more or less the same technical term. Dhammathat (or some
close variant such as thammasat or dhammasat) is the nearest equivalent to
legal codes, rajathat (or some close variant such as rachasat or yazathat)
describes the local equivalent of legislation and pyatton (or the 10ca1lan-
guage equivalent of this Burmese word) describes a collection of prece-
dents. While the literary historian will regard the three regions as one, a
comparative lawyer will point out that each sub-region uses the genres dif-
ferently. Each has established its own pecking order between the types of
legal literature, reflecting three quite different sub-regional approaches to
dispute settlement. Law in the western kingdoms developed in a dini'ction I
call "common law Buddhist," with professional lawyers and an interest in
case law. The eastern kingdoms adopted a more bureaucratic approach
which I call "state centred Buddhisf': they regarded law as the preserve of
the king and his small group of legal functionaries. The northern kingdoms,
which had achieved impressive state 6rganisations by the 15th century,
never recovered from the Burmese i n ~ a s i o n s of the mid 16th century.
Their legal approach relied heavily on monks, both as authors of texts and
as settlors of disputes. I call this "small town Buddhist law," a phrase
which hints both at their comparatively undeveloped traditions of the state
and their simple but fervent piety. In the following pages I shall examine
the three genres of legal literature in turn. Implicit in what I say is that liter-
ary history and comparative law feed back into each other. How legal lit-
erature is used, in other words, has a long term effect on what legal litera-
ture contains.
HUXLEY 73
THE DHAMMATHAT GENRE
Dhammathat is characterised both by its contents and by its claim to
authority. The contents of a dhammathat text should give a general account
of law for popular consumption, much as works like "Everyman His Own
Lawyer" and the "Penguin Guide to Law" do for us. Its appeal to authority
should bypass the throne and the government: dhammathat rules should be
obeyed because they are as old as human society, or because they are uni-
versally acknowledged as correct or because they are implicit in the
Buddha's dhamma. If a text is written by the king or relies on his name for
authority, I would prefer to label it as rajathat. But, since kings will try to
associate themselves with any source of legitimacy that happens to be lying
around, these categories are constantly blurred. For example, the oldest of
the Cambodian codes translated by Leclere, Kram Sanphea Thipdey
(1618), is a self-contained code aspiring to generality. On grounds of
content I am happy to call it a dhammathat, even though its preamble tells
us that it was promulgated bya king of Pnomh Penh. The king, I strongly
suspect, coopted an existing dhammathat by adding his preamble because
he wanted some of the legitimacy associated with the dhammathat text to
rub off on him and his throne. Aroonrut Wichienkeeo has shown the same
process at work in Lanna. Forty of the Lanna legal manuscripts include
Thammasat in their title but many of these compound the word with a royal
signifier: there is a "Royal Thammasat," a "Thammasat of King Kuna," a
"Mangrai Thammasat" and even a "Thammasat Rajathat." Assigning these
hybrids to their proper genre must depend on personal judgement. For my
part, if the text emphasises continuity with the wisdom of the ancients, I
regard it as a dhammathat. If it stresses the beneficial innovations of the
king, I regard it as rajathat.
Starting with the western sub-region, between twenty and forty different
Burmese dhammathats have survived in whole or part. Around a quarter of
these are in Pali (either prose or verse) and about a third of the remainder
are in Burmese verse, including works by two of Burma's most famous
poets: in 18th century Burma some dhammathats were written to show off
artistic skills rather than to reduce friction in society. We know the names,
and in some cases the authors and first lines, of another thirty or so lost
works. I gave a general guide to this body of literature in my previous
contribution to the journal. Two Arakanese prose dhammathats survive,
one of which, Kyannet, is in Pali and contains rules on sexual inequality
and the validity of wills which are quite different from the Burmese main-
stream. A detailed study of Kyannet would reveal exactly how different
the legal traditions of Arakan are from those of Burma. We have a dozen
74 JIABS 20.1
Mon dhammathats in prose (the best known of which has only survived in
Burmese translation) and one in verse, a macaronic of the Pali, Mon and
Burmese languages. Though there is no way of dating them, two of the
prose dhammathats give the impression of antiquity and seem to me to be
among the half dozen oldest surviving Southeast Asian dhammathats.
Turning to the northern sub-region, in Laos the kotmay thammasat Khun
Borom, the surviving text of which was compiled in 1503 from even earlier
material, bears a dhammathat title. Mayoury Ngaosyvathn argues that this
is the oldest of the Northern Tai law texts and that it has influenced Lanna
legal literature. The "Code de Vientiane" (probably written in the 17th cen-
tury) is also a dhammathat: it should properly be known as khamphi phra
thammasat buhan. Finot mentions a Pali manuscript with Laotian transla-
tion called the kotmai thammasat, 14 but I do not know if this has survived
Indo-China's turbulent 20th century. The earliest of the forty Lanna dham-
mathat texts can be dated to 1472 but many of these texts invoke the
authority of King Mangrai who founded Chiangmai at the end of the 13th
century. Further research is needed to establish whether Lanna legal litera-
ture is as old as it claims. None of the one-hundred-and-thirty-two Lanna
law texts is written in Pali or in verse. Three law texts from the
Sipsongpanna have recently been described. Their substantive rules reflect
Tai custom elsewhere, but their literary expression borrows from Chinese
genres. In the eastern sub-region the dhammathat has been completely
swamped by the rajathat genre. In the Three Seals Code of Siam and the
closely connected Cambodian literature, a Pali dhammathat has shrunk in
length and function to become an index and legitimiser to the
The remaining portions of the dhammathat explain, making much use of
Pali legal technicalities, what law consists of and why law should be
obeyed, but the details of what law is are left to be expounded by rajathat. It
appears that the process of moving substantive rules from a dhammathat to
a rajathat context was still going on in 1805; Burnay notes internal evi-
dence in the Three Seals Code that the compilers had access to a longer text
of the dhammathat than they chose to reproduce. is
The dhammathats' function is to provide guidance for unofficial dispute
settlement. They are written in the paddy field rather than the king's palace,
and view law from a village perspective. Hence murder and rape give rise
to financial penalties only: if the king wants to punish a rapist, that is his
business-a matter of rajathat. And hence the rules for mortgaging land are
14. Finot 1917, 136.
15. Bumay 1939, 163.
HUXLEY 75
described in detail, but neither the king's Land Registry nor the king's
charges levied on land-transfer are mentioned. The procedural rules
emphasised are those which allow an irrigation community to regulate its
own disputes without calling on the royal enforcement procedures. Dis-
putants are to be shamed or admonished into a settlement. Failing that, they
are offered a range of non-fatal ordeal procedures which tempt the defen-
dant into trying his luck, and allow a defeated litigant to lose the case with-
out losing too much face. In such unofficial proceedings the rules in the
dhammathat give an agreed normative background against which the
disputants can negotiate. But extra-legal considerations of relative wealth,
power, popularity and status loomed large in determining the outcome. A
disputant who wanted the matter settled by strictly legal criteria could turn
to the official courts, the domain of raja that staffed by the king's appointees,
but there too he would be disappointed. These institutions were usually
more concerned with raising revenue through bribes and extortionate court
charges than with administering justice. In one respect the Burmese
dhammathats differ from the rest: they alone contain rules about the pay-
ment and regulation of the legal profession. These rules, which tend to
favour the lawyer against his client, are testimony to the crucial role played
by Burmese lawyers in copying and editing the dhammathats.
THE RAJA THAT GENRE
Rajathat are orders proclaimed by the king either covering a specific topic
or giving general guidance to the judges of the royal courts. Some Laotian
rajathats are apparently held by the National Library in Bangkok, but have
not been studied or published. The Lanna literature is, as I have said,
poised between rajathat and dhammathat but one or two royal orders on
specific, narrowly focused issues have been preserved.1
6
However it is
Burma and Siam, the strong, centralising states of the 17th and 18th cen-
tury, which have preserved large collections of rajathat. At the head of
these two national traditions stand the two rajathat inscriptions of 1249
(Pagan) and 1397 (Sukhothai) which I mentioned earlier. Since these
inscriptions are extremely old and indisputably genuine, they have given
rise to much analysis and debate. 17 The gemes which they initiated have, in
manuscript form, swollen to enormous size. Over the last decade the
Burmese rajathats have been collected, edited and published by Professor
16. Sommai Premchit 1986, #20 and #56.
17. Than Tun 1959b; Huxley 1991; Prasert na Nagara and Griswold 1992;
Vickery 1978.
76 nABS 20.1
Than Tun. S i i u n e ~ e rajathat from Ayuttbaya and Bangkok are preserved in
King Rama 1's Three Seals Code of 1805. The sheer bulk of this
manuscript is hard to grasp. Compare it with the French Code of the same
year-Dhammaraja Napoleon's Code Civile. Both were compiled in 1805
by imperially appointed committees. of experts led respectively by Sunthon
Wohan and Jean Portalis. Both were intended to function as social glue
after decades of invasion and dynastic turmoil. Both were intended to be
authoritative, in the sense that judges could not consult any sources earlier
than the Code itself. Portalis' committee had to distil a compromise
between the old wisdom of the ancien regime and the new revolutionary
laws into a packet small enough to fit in every citizen's pocket. Sunthon's
committee had to reestablish continuity with the royal archives of
Ayuttbaya, the bureaucratic traditions of the past, which the Burmese had
nearly succeeded in destroying. Thus the Three Seals Code is about three
times the length of the Code Civile. We can think of it as ari internal
bureaucratic reordering of the palace archive rather than as a single com-
posed text. Four years earlier, a court official had versified thetKotmai
I .
Lilit, a selection of Ayuttbayan laws dealing with the procedure for hearing
appeals. IS This poem was intended to circulate as a text, to be read and
admired by anyone educated enough to appreciate its elegant verse. By
contrast the completion of the Three Seals Code seems more like a revision
of the entire Siamese legal data base than a correction of a single text. The
three copies of the [mal manuscript are, when seen in this light, mere back-
up copies kept within the governmental buildings of Bangkok. Each of
them was marked as royal property and guarunteed as genuine by the
eponymous three seals of the Interior, Defence and Finance Depiutments.
To what extent were the public able to consult the Three Seals Code?
Lingat spe<:;ulated that the third copy, kept within the offices of the
Lukkhun, Was "soon multiplied into innumerable copies" 19 and James Low,
writing in 1825 about Bangkok legislation in general, agrees: "The king
orders copies to be given to his officers, and it is through these officers that
the people procure copies."20 But when the printing press offered a cheap
and efficient medium for transmitting the Three Seals Code, it was rejected:
Nai Mot's attempts in 1849 to print part of the Three Seals Code met with
extreme royal displeasure. Evidently the Bangkok Bonapartes wished to
retain control over the spread of legal knowledge. Secrecy was preferred to
18. Terwie11983, 94, 98.
19. Lingat 1931, 10.
20. Low 1847, 395.
HUXLEY 77
promulgation until 1862, when Bradley was allowed to print his edition of
the Code. 1862 conveniently marks the start of the modernisation process,
the neo-colonial transplantation to Bangkok of European legal genres.
Lingat describes the archival reorganisation of 1805 as a step towards
legislation and an important further concentration of state power in the
hands of the Bangkok kings. He talks of earlier reorganisations of the
archives of Ayuttbaya, and favourably contrasts the Siamese practice with
that of the Burmese kings. Perhaps he exaggerates these points a trifle
21
but his fundamental point is well-taken: the Siamese kings periodically
reorganised their archives so as to distinguish personal and ephemeral royal
orders from those which were general and permanent. Than Tun's collec-
tion of Burmese rajathat demonstrates what a royal archive looks like which
has not undergone such pruning: the rare general and permanent order,
such as King Badon's order of 28-1-1795, is buried in a mound of personal
appointments and occasional dispositions, such as:
Nga Pu is appointed Chief of Workers who use Curtains to cover Un-
sightly Things from the Royal View. (ROB 3-3-1806)
Vickery, after subjecting the dates supplied by the Three Seals Code to a
close examination, suggests that earlier archival reorganisations had taken
place under the Ayuttbayan kings who reigned 1593-9, 1611-22 and 1633-
43. But the more the archives were reorganised, the less we can use them
at face value as historical evidence. The problems are evident: a law which
the Three Seals C<?de dates at 1527 refers to another law dated 1690!
Much of Siamese history has been based on a naive belief in the accuracy
of the dates given in the preambles to the Three Seals Code: these dates
urgently need confirmation from some other source. Perhaps Pitinai
Chaisaengsukkul's recent discoveries, which include both pre-1805
manuscripts of Ayutthayan rajathats and post-1805 manuscripts that are
textually independent of the Three Seals Code, contain the key to unlock
these problems. They will also cast light on how far the rajathat leaked out
of the archives into literature. Than Tun describes this process of leakage in
Burma:
These orders. . . were kept in the Palace Archives known as Shwe Daik .
They are now all lost. Fortunately, a minister or an assistant minister
21. Legislation is not the only way in which S. E. Asian kings may assert con-
trol over legal literature.
-78 JIABS 20.1
would have copies of these orders made for his own use and of
these survive. 22
In the Burmese case Than Tun has had to reconstruct the Palace archive
(destroyed when the British sacked Mandalay in 1885) from the copies of
rajathats that leaked into personal and monastic libraries. In the Siamese
case, the Three Seals Code committee had to reconstruct the Palace archive
(partially destroyed when the Burmese sacked Ayutthaya in 1767) from
ministerial copies. Pitinai's discoveries suggest a wider-spread leakage of
the texts of Siamese rajathats from the capital into distant towns, and from
the high ministers down to subordinate local bureaucrats and officers. Sec-
tions of the Palace archive have taken on a new life as independent texts. If
analysis of his discoveries shows that the written law available in
Ayutthayan provincial towns was in rajathat rather than dhammathat form,
we shall have a paradoxical legal picture: Ayutthaya's dispute
will prove to have been based on royal '1egislation" despite the kings' best
efforts to prevent people acquiring knowledge of that legislation!
THE PYATTON GENRE
I use the Burmese term pyatton to describe a collection of precedents,
understood in the non-technical sense of "information" rather than the tech-
nical sense of "binding authority." The American judge Richard Posner
gives a useful analysis of this distinction: he argues that in disciplines
where methodology is weak, helpful information will tend to be gathered in
the shape of anecdote, example and analogy.23 If we think of pyatton as
meaning "helpful legal information" we can understand why the geme
includes such different types of text: "pyatton as anecdote" can be a collec-
tion of mythical judgement tales, "pyatton as example" can be a theoretical
analysis of law and society and "pyatton as analogy" can be a collection of
law reports. The Burmese had to use complex circumlocutions to distin-
guish these three subgemes. Manugye talks about sorting out a jurisdic-
tional muddle between occasional and fulltime royal advisers "by the two
omens, namely the decisions of former kings and embryo Buddhas" (my
"pyatton as anecdote") "and decisions given in modem times in accordance
with these precedents and dhammathat"24 (my "pyatton as analogy").
22. Than Tun 1984-90,1, vii.
23. Posner 1988.
24. Richardson 1847,151; Manugye VI.6.
HUXLEY 79
Judgement Tales
Pyatton can mean a collection of folk stories describing the exploits of a
clever judge. A scriptural model for such collectiol1s is provided by the
Great Tunnel Jataka (#546) describing the career of a local boy who makes
good because of his skills at dispute settlement. Within this frame are
crammed many stories illustrating his courtroom successes each one of
which could be used as the plot for a puppet show, the theme for a story
teller or the precedent for a judge. Such judgement tales are particularly
popular in Southeast Asia. "Poor boy vindicates his rights in court by
clever argumenf' is nearly as popular a theme as "poor boy wins princess
with magician's help." The 547 Jataka of the Canonical collection were
supplemented by further Southeast Asian collections written in Pali which
have spread throughout the region under such titles as The Fifty Jataka and
The Chiang Mai Jataka. In these collections judgement tales are mixed
indiscriminately with romances, fairy tales and edifying sermons. Collec-
tions devoted exclusively to judgement tales are popular in the vernacular
languages: the Burmese Decisions of the. Princess Leamed-in-the-Law and
the Khmer Stories of Judge Rabbit are well-known examples. Maspero
reports a popular belief that the latter collection was written by
Buddhaghosa-which demonstrates how non-canonical stories seek to bor-
row authority from canonical sources.
25
The next stage is for the judge-
ment tales (both canonical and extra-canonical) to infiltrate dhammathat and
rajathat, at which point an interesting phenomenon occurs: when used to
illustrate a legal point, rather than an ethical moral, the judgement tales
prove to be highly adaptable. Different lawyers may use the same tale to
illustrate different conc1usions.
26
The judgement tales are Theravada legal
literature's main distinctive feature but have received little scholarly atten-
tion. Much of the material is already available in translation: this is a prime
area in which an ambitious graduate equipped with the latest literary theo-
ries could make her mark.
Theoretical analysis
By arranging some judgement tales into a particular order and by providing
a commentary thereto, the authors of pyatton collections could give a self-
conscious jurisprudential analysis of law and society. The Manugye
dhammathat orders nineteen judgement tales into a meditation on the legal
aspects of state formation. Twelve stories tell of the cowherd's clever
25. Maspero 1929,362.
26. Huxley 1996, 107-110.
80 JIABS 20.1
judgements within his village, then seven stories recount his clever han-
dling of appeals once he has been appointed judge at the royal court. The
seven tales date to the 11th or 12th century and show us how T,heravada
Buddhism was originally preached de haut en bas from palace to village.
The twelve tales were, I have argued, brought together in this form in the
18th century immediately after the fall of the Restored Toungoo dynasty to
show that village wisdom is, in the long run, preferable to the ideas emanat-
ing from the palace.
27
Two other works use the same technique to give a
theoretical analysis of legal reasoning though, jUdging by the corrupt state
of their text, they were only of marginal interest to their intended audience.
Included in a Nan text of the Mangraisat we have The Traditions of King
Mahosot which arranges judgement tales into a twelve fold analysis of
ways in which the judge can find the truth. From Burma we have the first
two questions of the Maharajathat which must originally have offered an
eight fold analysis of judicial fact-finding. I have analysed the Nan text in
some detail, because it is one of the very few occasions when the l ~ g a l l i t
erature reflects on itself: here, if anywhere, Theravada jurisprudence is to be
found. 28
Law Reports
Thirdly, pyatton can refer to non-fiction as well as fiction: a pyatton can be
a collection of real-life decisions by flesh and blood judges. At the highest
level of the court heirarchy, judgements were automatically preserved in the
palace archives.
29
I do not count such legal data-bases as literature, unless
an editor has prepared and circulated a selection of these law reports. Such
manuscript collections have survived from all three regions. The oldest,
Khlong Jea Phaya Kuena, contains a selection of verdicts given by Kings
Kuna and Tilokaraja from 14th and 15th century Lanna. That the four
manuscripts of this collection are substantially identical indicates that the
text was regarded as an important source of Chiangmai traditions. From
17th century Cambodia comes Chbap Tumnim Pi Bauran, a collection of
fifty cases concerned with lese majeste collected by the king's aunt.
30
Early
19th century Burma produced the Yesagyo Hkondaw Pyatton, a collection
of forty seven judgements dealing mainly with small-town quarrels. These
works give invaluable glimpses of actual dispute settlement, which some-
27. Huxley 1997,21-39.
28. Huxley n. d.
29. As we learn from Maharajathat p. 184 and ROB 2-9-1785.
30. Leclere 1898, 1:123-162.
HUXLEY 81
times differed markedly from what the other law texts lead us to expect. A
Cambodian litigant who inadvertently farts while taking the oath loses his
case. A Burmese husband is allowed the right (unknown in Burmese
dhammathats but mentioned in Laotian legal literature ) to sell his adulterous
wife into prostitution. Why-given that noone suggests that Theravada law
had a system of binding precedent-were these collections of law reports
collated and preserved? The natural assumption is that information about
how wise judges of the past handled disputes would be helpful to judges of
the present. In more technical terms, such collections could be quoted as
persuasive, but not binding, authority in a later case. This seems plausible,
but remains unproven, since none of the judgements which have been pre-
served quote previous "law report pyatton" as authority.
Defining Legal Literature
By restricting legal literature to the three genres of dhammathat, rajathat and
pyatton I have used a very narrow definition which does not reflect any
Southeast Asian classification. Judging b),\ the convoy evidence (the texts
which appear alongside the legal literature in manuscript) no sharp distinc-
tion was made between law, morality and good behaviour. Law texts are"
often bound together with Jatakas and other works on ethics and politesse.
A larger question is whether legal literature should include texts about
kingship and Vinaya. Those of us who think that Buddhist society divides
itself neatly into the three units of laity, sangha and king will want to treat
law for the king and law for the monk alongside law for the laity. Pre-mod-
ern Southeast Asia would not have disagreed with this analysis: in Burma
(the only culture for which we have reliable information about authorship)
we find the same person writing dharnmathats and vinaya-tikas (as for
example the Taungpila Sayadaw in the 17th century) or rajathats and rajaniti
(as for example the Maungdaung Sayadaw in the 18th century). Hence I
have added some rajadhamma and Vinaya references to the bibliography.
A SELECT BIBLIOGRAPHY
What follows is a guide for the interested student, rather than an exhaustive
bibliography. In particular, I have left out most of the excellent work writ-
ten in Japanese. For all-inclusive listings of relevant material, see Ishii ed.
(1978), A Preliminary Bibliography for the Study of Customary Laws of
Southeast Asia and Taiwan, Hooker ed. (1986), Laws of South-East Asia,
Volume 1 The Pre-Modem Texts 497-533 and Marikar (1978), A Bibliog-
raphy of the Books and Periodicals on the Laws of Sri Lanka (Ceylon).
82 JIABS 20.1
The Most Recent Surveys
To fmd out more about legal literature in a particular Theravada culture, you
should first consult the following articles, which are the most up to date
available in English. For Sri Lanka, see Tambiah (1968) Sinhala'Laws and
Customs, ch. 4. Though thirty years old, this account of Kandyan Bud-
dhist law has not been bettered. For Burma and Thailand, see Okudaira
1986, The Burmese Dhammathat and Ishii (1986), The Thai Thammasat.
Both may be found in Hooker ed. 1986. Unfortunately this volume bears a
stratospheric price-tag and thus is less popular with librarians than its con-
tents deserve. For a more iconoclastic approach to the Burmese dham-
mathats and the Three Seals Code see Huxley (1996), Thai, Mon and
Burmese Dhammathats-Who influenced Whom? and Vickery (1996), The
Constitution of Ayutthaya-an Investigation into the Three Seals Code.
These are in Huxley ed. (1996), Thai Law: Buddhist Law. In the same
volume are the best available introductions to the legal literature of Lanna
and Laos: Aroonrut Wichienkeeo (1996), Lanna Customary I;aw and
Mayoury Ngaosyvathyn (1996) An Introduction to the Laws of Khun
Borom. For Cambodia, nothing has been written during the last pundred
years to replace Leclere (1898), Codes Cambodgiens except for a few
pages in Khing Hoc Dy (1990), Contributions a fhistoire de la litterature
Khmer. If you can find a copy, Zhang Xiaohui, Xu Zhongqi and Zhang
Xisheng (1990), Explorations in the Laws of the Dai Nationality in West
Yunnan gives the first general account of the Sipsongpanna (Tai-Lu)
codes.
Classics of the Secondary Literature
On Sri Lanka: Derrett (1956), The Origins of the Law of the Kandyans
suggests that Sinhalese "Buddhist" law is essentially South Indian, with a
small admixture. Lingat (1989), La Fonction Royale a Ceylan
examines the role of the Buddhist kings of Lanka. On Burma:
Forchhammer (1885), The .Jardine Prize: An Essay was the first scholarly
European investigation into the field and should be treated with care: half
of what Forchhammer said was brilliantly perceptive while half was hope-
lessly inaccurate. E Maung (1951), The Expansion of Burmese Law is a
series of lectures by Burma's finest legal historian (and quondam Judge of
the High Court, Rangoon). Than Tun (1959a), The Legal System in Burma
AD 1000-1300 is based on a thorough knowledge of the Pagan inscrip-
tions. On Siam: one scholar towers above all the others in reputation and
productivity. Out of his very many published works, I recommend these
four as a good starting point: Lingat (1937), Vinaya et Droit Zai'que, Lingat
HUXLEY 83
(1950), The Evolution of the Conception of Law in Burma and Siam,
Lingat (1951), La Conception du Droit dans EIndochine Hinayaniste and
Lingat (1952), Les Regimes Matrimoniaux du Sud-Est de fAsie. While
Lingat always painted the big picture, his colleague, concentrating on details
of the textual tradition, has left a sounder foundation for future scholarship:
Bumay (1939), Materiaux pour une Edition Critique du Code de 1805.
His project has been carried forward by Vickery (1984), Prolegomena to
Methods for using the Ayutthayan Laws as Historical Source Material.
On Cambodia: Leclere (1898-9), Recherches sur les Origines Brahmani-
ques des Lois Cambodgiennes. It was this article, along with Forchhammer
1885 which led Max Weber into regarding Buddhist Southeast Asian law
as "a law of Hindu origin modified in the direction of Buddhism." On
Lanna: Wyatt et al. (1984), Symposium on Societal Organisation in main-
land Southeast Asia prior to the Eighteenth Century reacts to the first few
legal texts from Chiangmai to be discovered and published.
Doctoral Theses
The following four theses were produced by Doctoral students attached to
the Law Department, SOAS: Shwe Baw (1955), The Origin and Devel-
opment of Burmese Legal Literature; Kyin Swi (1965), The Judicial Sys-
tem in the Kingdom of Burma; M-)L. S. Jayasekera (1969), The Sources
and Development of the Customary Law of the Sinhalese up to 1835; M. B.
Voyce (1982), The Legal Aspect of Early Buddhist Vinaya. The first two
are held at the library of the Institute of Advanced Legal Studies, London;
the last two are held at Senate House Library, London. The library of the
University of Paris holds Mayoury Ngaosyvathn (1975), Changement et
Continuite de la Justice: Cas du Laos.
Legal Literature in Translation
The two key works of Sinhalese legal literature are both translated:
LeMesurier and Panabokke (1880), Niti Nighanduva is the nearest we have
to the tradition explaining itself in its own words. Ratnapala (1971), The
Katikiivatas translates and annotates the vinaya proclamations of the Sri
Lankan kings. A surprising amount of Western legal literature from
Burma, Arakan and Monland-up to a third of the total-is available in
English: the key work is U Gaung A Digest of the Burmese Buddhist Law
concerning Inheritance and Marriage being a Collection of Texts from
Thirty Six Dhammathats in separate Burmese (1898; 1899) and English
(1902; 1909) editions. U Gaung is often referred to by the ministerial title
of "the Kinwunmingyi" which he bore under the last two kings of
'84 nABS 20.1
Mandalay. His Digest contains excerpts from many manuscripts, including
two from Arakan and Bunnese works that have subsequently disappeared.
Additional passages from some of the same dhammathats are translated in
Jardine ed. (1882-3), Notes on Buddhist Law. The second volume of Shwe
Baw 1955 doctoral thesis translates all of Kaingza's Maharajathat, an
innovative law text from the early 17th century. Richardson (1847), Tl:ze
Damathat, or the Laws of Menoo, a parallel translation of the Manugye
dhammathat, was reprinted often throughout the 19th century, ensuring that
Manugye became the best known of the Burmese dhammathats.
Forchhammer (1892), King Wageru's Manu Dhammasattham gives what
purports to be a 13th century Mon original in Bunnese and English trarisla-
tion. Sangermano (1833), A Description of the Burmese Empire gives a
forty page summary of an otherwise unknown dhammathat which is
closely connected with Numbers 1, 2, 5, 8 and 10 of the Mon dhammathats
in: Nai Pan Hla (1992) Eleven Mon Dhammasat Texts. This beautifully
produced edition reproduces the Mon manuscripts as well as translating
them. For Bunnese rajathats see the ten volumes of Than Tun 1984-90 The
Royal Orders of Burma AD 1598-1885 which has become the. essential
tool for pre-colonial Bunnese historiography. It gives the full Buimese text
of each order, coupled with an English language summary. Htin Aung
(1962), Burmese Law Tales translates some law reports from the Yesagyo
Hkondaw pyatton, and retells several judgement tales from the oral tradi-
tion. Bandow (1881), The Precedents of Princess Thoodamma Tsari
translates the best known Burmese collection of judgement tales. Gray
(1886), The Niti Literature of Burma translates three of the most popular
ethical works. Taw Sein Ko (1893), A Preliminary Study of the Kalyani
Inscriptions of Dhammaceti translates a pious 15th century Mon king's
grapplings with a knotty problem in the Vinaya.
The Northern law texts have been much less well served. Only three of
the Lanna discoveries are translated: Aroonrut Wichienkeeo and Gehan
Wijeyewardane (1986), The Laws of King Mangrai. This very useful vol-
ume transcribes and translates a Chiangmai text contained in a manuscript
from Nan, which suggests that some at least of their legal literature was
shared between the Middle Mekong kingdoms. The only Laotian text
translated is the "Code de Vientiane" khamphi phra thammasat buhan in
Raquez (1902), Pages Laotiennes. An interesting portion of one of the
Sipsongpanna (Tai-Lu) legal codes has been translated in I. Bain (1989),
The Correct Attitude and Approach for handling Legal Cases. Leclere
(1903), Contes Laotie!1-S contains a chapter of contes judiciaires, and some
helpful comparisons with the Bunnese and Cambodian judgement tale col-
HUXLEY 85
lections. Sila Viravongs (1970-1), Thao Sieo Savat-Conte ludiciaire Lao
and Brengues (1904), Contes ludiciaires Laotiens translate more Laotian
judgement tales. The Ahom laws are in chapter two of Barua (1935),
Ahom Buran}i. Turning to the Eastern texts, the Three Seals Code has been
very badly served by translators. If you want to consult it seriously, you
must leam Thai. However, non-Thai speakers can sample its charms from
the portions translated into French in Lingat (1931), EEsclavage Prive
dans le Vieux Droit Siamois, Lingat (1961), Le Delit de Voisinage
Malijique and the extensive French paraphrases in Lingat (1964) La
Preuve dans fancien Droit Siamois. By contrast, all of the Cambodian
legal literature is available in translation, thanks to Adhemard Leclere's
exemplary work: Leclere (1898), Codes Cambodgiens. Khmer judgement
tales are translated in Leclere (1895), Cambodge, Contes et Legendes,
Midan (1933), Histoires du luge Lievre -Recueil de Contes Cambodgiens
and Anonymous (1970), Recueil de contes Khmers-Le luge Lievre. The
Cambodian works on good manners are translated in Jenner and Pou
(1975-81), Les Cpap, ou "Code de Conduite" Khmers. Written at the
beginning of this century in the midst of Thailand's transition to modernity,
Vajirananavarorasa (1969-83) The Entrance to the Vinaya expounds the
traditions of Thailand's strict-observance lineage.
Legal Literature in Southeast Asian languages
Several important pre-colonial Burmese texts have been published in
Rangoon and Mandalay. Among them are Maung Tetto's 1878 edition of
Myat Aung's Manoo Woonana, a Pali verse dhammathat written in 1772
by an ambitio}!s ex-monk, the anonymous (1870), Mahapanna-kyaw
pyatton which includes ten pages of Shin Kyaw Thu of Hanthawadfs
pyatton written in 1625 and U Nirodha's ed. (1899), Vinaya samaha vinic-
chaya kyan a collection of twenty-four vinaya treatises written by Burmese
sayadaws, each of which had circulated in manuscript during the 18th and
19th century. At p.121-9 of Okudaira 1979, whose title can be translated
from the Japanese as An Outline of the Origin, Development and Research
on the Dhammathats, is a catalogue of sixty-five legal manuscripts held by
the University Central Library, Rangoon and sixty-six more at the Rangoon
National Library. But note that Burmese libraries have been closed to non-
Burmese students for the last thirty years. Some of this material is available
in microfIlm as catalogued in List of Microfilms Deposited in the Centre for
East Asian Cultural Studies: Part 8, Burma. I understand that Cornell also
holds much of this in microfilm in its ''Luce Collection" but fm not aware
of any published catalogue. Pe Maung Tin's two articles Burmese
86 nABS 20.1
Manuscripts in the British Museum and in the Bodleian Library, Oxford
list ten legal manuscripts held in England. Two relevant Burmese Pali
works are in print: Jaini (1983), Zimme Pafifiiisa is an edition of the
Burmese collection of Fifty lataka while Maung Tin (1914) has edited
Riijiidhiriijaviliisini, a florid 18th century Pali work about kingship by the
1st Maungdaung sayadaw. Turning to northern legal literature, the discov-
eries in Lanna over the last twenty years are catalogued in Sommai
Premchit (1986), Lan Na Literature: Catalogue of 954 Secular Titles
available in Microfilm. Much of this has been transcribed into central Thai
and discussed in Prasert na N agara, Pitinai Chaisangsukkul and Aroonrut
Wichienkeeo (1989) twelve volume work, whose title translates from Thai
as Basic Research on the Ancient Lanna Law: An Analysis of its Legal
Structure and Texts as Inscribed in Palm Leaves from Time Immemorial.
Details of the Laotian law texts which have been printed in Vientiane may
be found in Mayoury 1996. A Pali work on kingship and go:yemment
called the Lokavinaya or the Dhanafijaya liitaka was known in most Mid-
dle Mekong capitals. A version of it from Kengtung has been edited: J aini
(1986), Lokaneyyappara1)a11J,. (
Details of the various printed and digitised editions of the Three Seals
Code are given in Ishii 1986 and Vickery 1996. Relevant materials held in
Bangkok libraries are catalogued in Burnay (1930), Inventaire des
Manuscrits juridique Siamois. The seven-hundred newly discovered cen-
tral and southern Thai legal manuscripts are catalogued in Sarup Ritchu
(1996), Local Laws of Southern Thailand and Pitinai Chaisaengsukkul
(1996), Newly Discovered Source Material. The manuscripts described by
Low (1847), On the laws of mu'ung Thai or Siam are, I am reliably
informed, still held in the library of the Royal Asiatic Society, London,
although t4ey do not appear in its catalogue. A Traditional family law text,
Ayudhya-style writing was exhibited at the Australian National University
Library in July 1987. Twenty Cambodian legal manuscripts which Leclere
deposited with the Bibliotheque Nationale de Paris are catalogued in Khing
Hoc Dy 1990.
HUXLEY 87
BIBLIOGRAPHY
Anonymous. 1870. Mahapanna-kyaw pyatton. Rangoon: the Hanthawaddy
Press.
--'----,. 1970. Recueil de contes Khmers-Le Juge Lievre Tome 4. Fasc. 4
Serie de Culture et Civilisation Khmeres. Phnomh Penh: Edition de l'Institut
Bouddhique.
----. 1976. List of Microfilms Deposited in the Centre for East Asian
Cultural Studies: Part 8, Burma. Tokyo: Centre for East Asian Cultural Stud-
ies, The Tokyo Bunko.
Aroonrut Wichienkeeo, and Gehan Wijeyewardane. 1986. The Laws of King
Mangrai. Canberra: The Richard Davis Fund and the Australian National
University.
Aroonrut Wichienkeeo. 1996. "Lanna Customary Law." Thai Law: Buddhist
Law Essays on the Legal History of Thailand, Laos and Burma. Ed. A.
Huxley. Bangkok: White Orchid Press. 30-41.
Bain, I. 1989. "The Correct Attitude and Approach for Handling Legal Cases."
Thai-Yunnan Project Newsletter 4: 25-6.
Bandow, C. 1881. The -Precedents of Princess Thoodamma Tsari. 6th ed.
Rangoon: American Mission Press.
Barna, G. 1930. Ahom Buranji: From the Earliest Times to the End of Ahom
Rule. Calcutta: Assam Government.
Bechert, H., Khin Khin Su, and Tin Tin Myint. 1979. Burmese Manuscripts.
Wiesbaden: Franz Steiner Verlag. No. 42 of their catalogue at 1:53-4 is a 19th
century dhammathat.
Brengues, J. 1904. "Contes Judiciaires Laotiens." La Revue Indochinoise 2:
124-8.
Burnay, J. 1930-2. "Inventaire des Manuscrits juridiques Siamois." Journal of
the Siam Society 23 (1930): l35-203; 24 (1931): 29-79; 25 (1932): 127-52.
----. 1939. "Mat6riaux pour une Edition Critique du Code de 1805." Jour-
nal of the Siam Society 31: 155-168.
Derrett, J. 1956. 'The Origins of the Law of the Kandyans." University of
Ceylon Review 14: 105-150.
E Maung. 1951. "The Expansion of Burmese Law, a Series of Lectures."
Rangoon: Royal Printing Works.
Finot, L. 1917. "Recherches sur la Litterature Laotienne." Bulletin de f Ecole
Frangaise de fExtreme Orient 17: 5-224.
Forchharnmer, E. 1885. The Jardine Prize: an Essay. Rangoon: Government
Press.
----,. 1892. King Wageru's Manu Dhammasattam: Text, Translation and
Notes. Rangoon: Government Printing.
U Gaung (alias the Kinwunrningyi). 1902-9. Translation of a Digest of the
Burmese Buddhist Law concerning Marriage and Inheritance, being a Col-
lection of Texts from Thirty Six Dhammathats. Vol.1 1902; vol. 2 1909.
Rangoon: Superintendent of Government Printing.
88 JIABS 20.1
Gray, J. 1886. TheNiti Literature of Burma. London: Trubner's Oriental Series.
Htin Aung. 1962. Bunnese Law Tales, The Legal Element in Burmese Folk-
Lore. Oxford: Oxford University Press.
Hinuber, O. v. A Handbook of Pali Literature. Berlin: Walter de Gruyter, 1996.
Huxley, A. 1991. "Sanction in the Theravada Buddhist Kingdoms of S.E. Asia."
Receuils de la Societe Jean Bodin 58.4: 335-370.
----. 1996. "Thai, Mon and Burmese Dhammathats-Who Influenced
Whom?" Thai Law: Buddhist Law Essays on the Legal History of Thailand,
Laos and Bunna. Ed. A. Huxley. Bangkok: White Orchid Press. 82-131.
----. 1997. "The Village Knows Best: Social Organisation in an 18th
century Burmese Law Text." South East Asia Research 5.1: 21-39.
----,. N. d. "The Traditions of Mahosadha: Legal Reasoning from
Northern Thailand." Forthcoming in Bulletin of the School of Oriental and
African Studies.
Ishii, Y., ed. 1978. "A Preliminary Bibliography for the Study of Customary
Laws of Southeast Asia and Taiwan." Bulletin of the National Museum of Eth-
nology. Tokyo 3.2: 781-833,
---. 1986. "The Thai Thammasat (with a Note on the Lao Thawnasat)."
Laws of South-East Asia, Volume 1 The Pre-Modem Texts. Ed. M. Hooker.
Singapore: Butterworths and Co. 143-203
Jackson, B. 1975. "From Dharma to Law." American Journal of Comparative
Law 23: 490-512.
Jaini, P. 1986. Lokaneyyappara1Jal1J. Pali Text Society Text Series 175. London:
Routledge and Kegan Paul.
----. 1983. Pafifiiisa-Jiitaka or Zimme Pafifiiisa Pali Text Society Text
Series 172. Vol. 2. London: Routledge and Kegan Paul.
Jardine, J., ed. 1882-3. Notes on Buddhist Law. By the Judicial Commissioner,
British Burma. Eight parts, issued separately. Rpt. as one volume 1903.
Rangoon: Government Printing.
Jenner, P., and S. Pou. 1975-81. "Les Cpap, ou 'Code de Conduite' Khmers."
Bulletin de rEcole Franqaise de CExtreme-Orient 62 (1975): 369-94; 63
(1976): 313-350; 64 (1977): 167-216; 65 (1978): 361-40; 66 (1979): 369-94;
70 (1981): 135-94.
Khing Hoc Dy. 1990. Contributions a rHistoire de la Litterature Khmere Vol-
ume 1: Litterature de repoque "Classique" (XVe-XIXe siecles). Paris: Editions
L'Harmattan.63-8.
Kotelawale, D. 1987. "Realms of Law: Judicial Procedure of the Low Country
Sinhalese during Dutch Rule in Sri Lanka." Modem Sri Lanka Studies 2: 112-
120.
Leclere, A. 1898. Les Codes Cambodgiens (pub lies sous les auspices de M.
Dourner, Gouverneur general de rIndochine franqaise. 2 vols. Paris: E.
Leroux.
----,.1903. Contes Laotiens et Contes Cambodgiens. Paris: E. Leroux.
HUXLEY 89
----. 1898-9. "Recherches sur les Origines Brahmaniques des Lois Cam-
bodgiennes." Nouvelle Revue Historiqu6S de Droit franr;ais et etranger Sept. /
Oct. (1898): 609-656; Mai / Jun. (1899): 265-288.
LeMesurier, C., and T. Panabokke. 1880. Niti Nighanduva, or The Vocabulary
of Law as it EJCisted in the Last Days of the Kandyan Kingdom. Colombo:
Government Printer. .
Lingat, R. 1931. L'Esclavage Prive dans Ie vieux lois Siamois. Paris: Les Edi-
tions Domat-Montchrestien.
----. 1937. "Vinaya et droit laique: Etudes sur les conflits de la loi
religieuse et de la loi laique dans l'Indochine hinayiste." Bulletin de fEcole
Franr;aise de fExtreme-Orient 37: 415-477.
----. 1950. "Evolution of the Conception of Law in Burma and Siam."
Journal of the Siam Society 38: 9-31.
----.1951. "La Conceptiondu Droit dans l'Indochine Hinayaniste." Bul-
letin de f Ecole Franr;aise de f Extreme-Orient 44: 163-187.
----.1952. Les Regimes Matrimoniaux du Sud-Est de fAsie. Tome 1: Les
Regimes Traditionnels. Paris: E. de Boccard.
----,. 1961. "Le Delit de V oisinage Malefique dans Ie vieux droit Siamois."
Journal Asiatique 249: 63-84.
----.1964. "La Preuve dans l'ancien droit Siamois." Receuils de la Societe
Jean Bodin 17: 397-418.
----. 1989. "La Fonction Royale a Ceylan." Royautes Bouddhiques. Ed.
G. Fussman and E. Meyer, preface by L. Dumont. Paris: Ecole des Hautes
Etudes en Science Sociales.
Low, J. 1847. "On the Laws of mu'ung Thai or Siam." Journal of the Indian
Archipelago and Eastern Asia 1: 327-429. At p. Ivii of vol. 2 of the journal
Low reveals that his article was written in about 1825.
MacCormack, G. 1976. "Professor Gluckman's Contribution to Legal Theory."
Juridical Review (1976): 229-248
Marikar, 1. 1978. "A Bibliography of the Books and Periodicals on the Laws of
Sri Lanka (Ceylon)." Colombo Law Review 4: 130-143.
Maspero, G. 1929. Un Empire Colonial Franr;ais: rIndochine. Vol. 1. Paris:
Editions G. van Oest.
Maung Thettoo, ed. 1878. Manoo Woonana. Rangoon: Superintendent of Gov-
ernment Printing.
Maung Tin, ed. 1914. "Rajadhirajavilasini." Journal of the Burma Research
Society 4: 7-21.
Mayoury Ngaosyvathn. 1996. "An Introduction to the Laws of Khun Borom."
Thai Law: Buddhist Law Essays on the Legal History of Thailand, Laos and
Burma. Ed. A. Huxley.Bangkok: White Orchid Press. 74-81.
Midan, P. 1933. "Histoires du Juge Lievre-Recueil de Contes Cambodgiens."
Bulletin de la Societe des Etudes Indochinoises 8.4:I-VI, 1-116.
Nai Pan Hla. 1992. Eleven Mon Dhammasat Texts. Bibliotheca Codicum Asi-
aticorum 6. Tokyo: The Centre for East Asian Cultural Studies for Unesco. A
less opulent paperback version preceded it: Nai Pan Hla. The Significant Role
"90 JIABS 20.1
of the Mon Version Dharmasastra. Tokyo: Institute for the Study of the
Languages and Cultures of Asia and Africa. 1991.
U Nirodha, ed. 1899. Vinaya samaha vinicchaya kyam. Vol. 1: 1899 vol. 2:
1901 vol. 3: 1902. Mandalay: Mandalay Times Press. .
Okudaira, R. 1979. "An Outline of the Origin, Development and Research on
the Dhammathats" (main article inJapanese). Tonan Ajia Kenkya (South East
Asian Studies) 17: 99-130.
----. 1986. "The Burmese Dhammathat." Laws of South-East Asia, Vol-
ume 1 The Pre-Modem Texts. Ed. M. Hooker. Singapore: Butterworths and
Co. 23-142
Pe Maung Tin. 1924. "Burma Manuscripts in the British Museum." Journal of
the Burma Research Society 14: 221-246. Lists nine law texts.
----. 1925. "Burma Manuscripts in the Bodleian Library, Oxford." Journal
of the Burma Research Society 15: 145-7. Lists a dhammathat in Burmese
verse.
Pitinai Chaisaengsukkul. 1996. "Newly Discovered Source Material from 34
Central and 14 Southern Provinces of the Siamese Kingdom." ',Thai Law:
Buddhist Law Essays on the Legal History of Thailand, Laos and Burma.
Ed. A. Huxley. Bangkok: White Orchid Press. 42-62.
Posner, R. 1988. "The Jurisprudence of Skepticism." Michigan Law Review 86:
827-891.
Prasert na Nagara, Pitinai Chaisangsukkul, and Aroonrut Wichienkeeo. 1989.
Wichai phu'nthan ru' ang kotmai Lanna boran: wikhro' rabop khrongkan sang
lae nu'aha botbanyat thi chara nai bailan (Basic Research on the Ancient
Lanna Law: An Analysis of its Legal Structure and Texts as Inscribed in Palm
Leaves from Time Immemorial). 12 vols. Bangkok: Mahawitthayalai
Thammasat.
Prasert na Nagara, and A. Griswold. 1992. Epigraphic and Historical Studies
(Bangkok: the Historical Society under the Royal Patronage of H. R. H.
Princess Maha Chakri Sirindhorn). Study #4.
Raquez, A. 1902. Pages Laotiennes. Hanoi: Schneider. The "Code re
Vientiane" is summarised on pp. 402-418. This version has been reprinted,
without illustrations, in Bulletin des Amis du Royaume Lao 1970-2: 23-6; 75-
88; 94-9; 123-31; 70-76.
Ratnapala, N. 1971. The Katiklivatas: Law of the Buddhist Order of Ceylon
from the 12th to the 18th Century. Miinchen: Kitzinger, Miinchen Studien zur
Sprachwissenschaft.
Richardson, D. 1847. The Damathat, or the Laws of Menoo. Moulmein: Ameri-
can Baptist Mission Press.
Sangermano, V. 1885. A Description of the Burmese Empire. 1st English ed.
Rangoon: 1885. Trans. W. Tandy with a preface by John Jardine. 5th ed.
Santiago de Compostela: Susil Gupta, 1966.
Sarup Ritchu. 1996. "Legal Manuscripts from Southern Thailand." Thai Law:
Buddhist Law Essays on the Legal History of Thailand, Laos and Burma. Ed.
A. Huxley. Bangkok: White Orchid Press. 62-73.
HUXLEY 91
Sommai Premchit. 1986. Lan Na Literature, Catalogue of 954 Secular Titles.
Bangkok: Chulalongkom University Bookstore.
Tambiah, H. W. 1968. Sinhala Laws and Customs. Colombo: Lake House
Investments. C:h. 4.
Taw Sein Ko. 1893. "A Preliminary Study of the Kalyani Inscriptions of
Dhammaceti." Indian Antiquary (January 1893): 11-17; (February 1893): 29-
53; (April 1893): 8 5 ~ 8 9 .
Terwiel, B. 1983. A History of Modem Thailand 1767-1942. St. Lucia: U.
Queensland Press.
Than Tun. 1959a. "The Legal System in Burma AD 1000-1300." Burma Law
Institute Journal 1: 171-184. Rpt. without footnotes in Essays on the History
and Buddhism of Burma by Professor Than Tun. Ed. P. Strachan. Isle of
Arran: Kiscadale Press, 1988. 69-82.
---. 1959b. "A History of Burma." New Burma Weekly (March 1, 1959):
ch.4.
---. 1984-90. The Royal Orders of Burma, AD 1598-1885. 10 vols.
Kyoto: Centre for Southeast Asian Studies.
Vajirananavarorasa, Krotn Phraya. 1969-83. The Entrance to the Vinaya. Vinaya
Mukha. 3 vols. Bangkok: Mahamakut Rajavidyalaya Press.
Vickery, M. 1978. "A Guide through some Recent Sukhothai Historiography."
Journal of the Siam Society 66: 182-243.
----. 1984. "Prolegomena to Methods for using the Ayutthayan Laws as
Historical Source Material." Journal of the Siam Society 72: 37-59.
----,. 1996. "The Constitution of Ayutthaya-an Investigation into the
Three Seals Code." Thai Law: Buddhist Law Essays on the Legal History of
Thailand, Laos and Burma. Ed. A. Huxley. Bangkok: White Orchid Press.
132-194.
Sila Viravongs. 1970-1. "Thao Sieo Savat-Conte Iudiciaire Lao." Bulletin des
Amis du Royaume Lao 3-6: 59-67,79-82,191-9.
Wyatt, D., et al. 1984. "Symposium on Societal Organisation in mainland South-
east Asia prior to the Eighteenth Century." Journal of Southeast Asian Studies
15: 233-270.
Zhang Xiaohui, Xu Zhongqi, and Zhang Xisheng. 1990. "Explorations in the
Laws of the Dai Nationality in West Yunnan." Proceedings of the 4th Inter-
national Thai Studies Conference, Yunnan.
PETER SKILLING
The Advent of Theravada Buddhism
to Mainland South-east Asia
In the present paper I examine evidence for the school-affiliation of the
early Buddhism of mainland South-east Asia, in the first millenium of the
Common Era.! Is the evidence sufficient to establish that this school was
the Theravada, and, if so, when and from where did it arrive in the
region?
For the Theravada of Ceylon-or more precisely, for the Mahavihara
school of the Theravada-we have the history as presented in the two
famous chronicles, the Dfpava11Jsa and Mahiiva11Jsa. Information may
also be gleaned from references to historical events embedded in the
commentaries of Buddhaghosa and others, from inscriptions in Old Sin-
hala and Sanskrit, from archreological and iconographical evidence, and
from Chinese sources-in some cases fIrst hand, such as that supplied by
the redoutable pilgrim Fa-hien. Altogether, we have at least in broad out-
line a continuous history of Theravada in Ceylon from its inception up to
the present day.
Outside of Ceylon, the history of Theravada is obscure. For mainland
India we have almost no information at all. There are some-but not
many-references to Theravadin doctrines in the works of other schools,2
but the historical information-such as that provided by inscriptions or by
the Chinese pilgrims Hsuan-tsang and I-ching-is at best sketchy.
For the South-east Asia of the early period we do not have any histori-
cal records comparable to those of Ceylon: no indigenous chronicles,
This is a revised version of a paper given at the Ecole franlj:aise d'Extreme-
Orient, Phnom Penh, 6 July 1996. The title was inspired by Luce 1974.
1. That is, I do not discuss the Buddhism of peninsular and insular South-east
Asia, or that of Campa (the coastal regions of present-day central and southern
Vietnam). In none of these areas is there any early evidence for Theravada
Buddhism.
2. See Skilling 1987, 1993a and b, and 1994 for some examples from Tibetan
sources.
93
94 JIABS 20.1
whether in Pllli, Sanskrit, or in vernaculars survive. The few extant his-
torical inscriptions do not give us any continuous history, and Chinese
reports tell us little about the type of Buddhism practised on the mainland.
pali Inscriptionsfrom Burma and Siam
The main evidence for the of early Buddhism in South-
east Asia comes from Pali inscriptions. These are known from two mam
areas: the Pyu kingdom of SrIlq;etra in the vicinity of Prome in the lower
Irrawaddy valley of Burma, and the Mon kingdom of Dvaravati in the
Chao Phraya basin of Siam.3 The inscriptions from Burma are engraved
on gold plates (fashioned in imitation of palm-leaf manuscripts), a silver
reliquary (stupa), terracotta tablets, and stone slabs. The inscriptions
from Siam are engraved on stone dhammacakkas, octagonal pillars, stone
slabs, and clay tablets and reliquaries. The script used in both cases is
similar, and may be described as a variety of the South Indian Pallava
script. 4 The inscriptions are dated to the 5th to 7th centuTIes CE,
the Siamese inscriptions to the 6th to 8th centuries: that is, they are
broadly contemporary. 5
(1) Inscriptions from the region of
-the ye dhamma hetuppabhava verse (Vinaya Mahavagga, 140.28-29);
-the iti pi so bhagava formula (cf. Dhajagga-sutta, SN I 219.31-33);
-the svakkhato bhagavata dhammo formula (cf. Dhajagga-sutta, SN I
220.1-2);
3. In this paper I set aside the historical questions (of, for example, chronology
and geographical extent) attached to the names of these two kingdoms, and
(with not a little reluctance) use the names as a conventional shorthand.
4. The script of the Pyu inscriptions has in the past been variously described as
Kadamba, Telegu-Canara, or Grantha: for a welcome reappraisal see Stargardt
1995,204.
5. For the dating of the former see Stargardt 1995, for the latter e. g. Bauer
1991 and Skilling forthcoming (a). It should be stressed that the inscriptions do
not bear any dates, and that those assigned to them are tentative and approxi-
mate. A comprehensive comparative palreographical analysis of the
with the DvaravatI corpus remains a desideratum.
6. For details see Ray 1939,41-52; Luce 1974, 125-27; and Stargardt 1995.
Most of the texts are brought together in U Tha Myat 1963. Note that several
of the passages are known from more than one inscription.
SKILLING 95
-the formula of dependent arising (pa.tieea-samuppiida: cf. Vinaya
Mahiivagga, I 1.10-2.1);7
-stanzas sung by Sakka, Lord of the Gods, in praise of the Buddha enter-
ing Rajagaha (Vinaya Mahiivagga, I 38.15-23, 29-30);
-the maggiin' se.t!ho verse (Dhammapada 273);
-verses from three popular paritta-s: the Mangala-, Ratana-, and Mora-
sutta-s; 8
-the four confidences (vesiirajja) of a Buddha (MN I 71.32; AN II 8,
penult);
-the 37 factors conducive to awakening (bodhipakkhiya-dhammii);
-a list of miscellaneous numerically grouped items, in ascending order;
-a list of the 14 Buddha iiiilJa-s (cf. Pa.tisambhidamagga I 133.19-30);
-a fragment of a commentary onpa,tieeasamuppiida (cf. Vibhanga 144-
45);
-the opening of the miitika: kusala [dhammii aku]salii dhammii
abyaka[ta] dhammii (cf. DhammasangalJi 1.4);
-a fragment giving two of the 24 conditions: [adhi]patipaeeayo anan-
tarapaccayo ;
7. In addition to the pa.tieeasamuppiida inscribed on gold plates from
the Vinaya Mahiivagga version is known from a stone slab from
Kunzeik, Shwegyin township, Pegu: see Aung Thaw 1978, 111. As far as I
know this handsome and well-preserved inscription has not been published, but
fortunately most of it can be descried from the photograph at Aung Thaw p.
110. It opens (the readings here are preliminary) with the introductory [1]
t(e)na samayena buddho bhaga(vii) uruveliiyal1J. viharati na(j)j(ii) (neraiija-
riiya? unclear) [2] fire (a: tire?) bodhirukkhamiile pathamiibhisambuddho atha
kho bhagavii ... , followed by the full pa,ticeasamuppiida formula, both anu-
loma (lines 5-9) and pa,tiloma (lines 9-14). The latter opens with the phrase
avijjiiya tv eva asesaviriiganirodhii, characteristic of the Theravadin (Pali)
version only, and not known in versions of other schools, such as the
(MUla)Sarvastivadins or Lokottaravadins, or from the Prakrit inscriptions from
Devnlmon and Ratnagiri, all of which open with equivalents of avijjii-nirodhii.
The pa,tiloma is followed by the yadii have piitubhavanti dhammii verse (lines
15-18), known also from inscriptions from Siam. The last two lines continue
with the prose text of the Mahiivagga-atha kho (bhaga)vii r(attiyii) maj(jh)
imal1J. (yii)mal1J. palieea-suggesting that the slab is part of a longer inscription.
For the Devnlmon and Ratnagiri inscriptions see von Hintiber 1985; for a
suggestion that the former might be Vatslputrlya or SammatIya, see Skilling
forthcoming (c).
8. For these see Skilling forthcoming (b).
96 JIABS 20.1
-a list of seven of the eight vipassanii fiiifJa-s (cf. Visuddhimagga
XXI. 1).
(2) Inscriptions from the Chao Phraya basin: 9
-the ye dhammii hetuppabhavii verse;
-the formula of dependent arising (pa.ticca-samuppiida);
-an enumeration of the four truths of the noble (ariya-sacca), the twelve
links of dependent arising (pa.ticcasamuppiida), and the 37 factors
conducive to awakening (bodhipakkhiyacdhamma), inscribed together
on a rectangular stone bar from Nakhon Pathom; 10
-extracts from the prose Dhammacakkappavattana-sutta, the "first ser-
mon" spoken by the Buddha in the Deer Park at Sarnath, found on
stone dhammacakkas;l1
-the three yadii have piitubhavanti dhammii verses (Vinaya Mahiivagga,
12.3-26);
-the anekajiitisa11Jsiira11J verses (Dhammapada 153-54);
-the dukkha11J dukkhasamuppiida11J verse (Dhammapada 191); 12
-the abhififieyya11J abhififiiita11J verse (Suttanipiita 558);
-fragments of the 16 senses (attha) of the four truths (cf.
Pa.tisambhidiimagga 19.31-20.6); 13
-niibiidhaka11J yato dukkha11J ... , non-canonical verses on the four truths
(cited at Visuddhimagga XVI.25);
-sacca-kicca-kata-fiiifJa11J ... , a non-canonical verse on the twelve
aspects (dviidasiikiira) of the four truths (cited in the Pa.thama-
sambodhi and Siiratthasamuccaya);
-three verses from the Telaka.tiiha-giithii.14
The evidence of theinscriptions may be examined from two aspects: lan-
guage and contents. The language of both the and Dvm-avatI
palreographs is Pilii. Is the use of Pali sufficient to establish the presence
of the Theravada? Or could another Buddhist school have also transmitted
9. Most of the inscriptions may be found in Supaphan na Bangchang 2529
(1986), 15-40. As in the case of the inscriptions, several of the
passages are known from more than one inscription.
10. See Skilling 1992.
11. See Skilling forthcoming (a) for references.
12. See Skilling 1991 and 1992.
13. See Skilling forthcoming (a) for this and the two following passages.
14. See references below. The inscription is from Prachin Buri, and thus out-
side of the Chao Phraya valley proper.
SKILLING 97
its sacred writ in Pali, and have been responsible for the inscriptions?
From an early date, Buddhist tradition recognized dialect as one of the
key distinguishing features of the different schools (nikiiya). In the sec-
ond half of the first millenium of the Common Era, tradition spoke of four
main schools, each transmitting its canon in a different lndic dialect:
(Ml1la)Sarvastivadins, who used Sanskrit; MahasaIl).ghikas, whe used an
intermediate language; SammatIyas, who used and
Sthaviras (that is, Theras), who used Paisacl. 15 The tradition is confirmed
by the distinctive and consistent linguistic features of available texts of
the schools. On this evidence I conclude that it is unlikely that another
school would have used Pali, and that the use of that language in the
inscriptions is a strong indication of Theravadin activity in the region.
What about the contents ofthe inscriptions? It is true that the canonical
extracts-such as the various formulas, the Dhammacakkappavattana-
sutta, and the verses-belong to the common heritage of Buddhism: but
our epigraphs give them in their Theravadin recensions, and they agree
very closely indeed with the received transmission that we know today.l6
The "extracts" from the Abhidhamma and Pa.tisambhidiimagga are rather
more indicative. As far as is known, the seven books of the Theravadin
Abhidhamma are unique to that school, and employ a unique sys-
tem and technical vocabulary. The inscriptions preserve frag-
ments with counterparts in the Miitikii, the Vibhanga, and the list of 24
conditions (paccaya), all of which may be described as specifically
Theravadin. Inscriptions from both and Siam employ technical
categories known from the Pa.tisambhidiimagga (whether or not they are
actual extracts is not clear), an ancient commentary transmitted in the
Khuddaka-nikiiya of the Pali Canon, and unique to the Theravadin school.
The non-canonical inscriptions provide further convincing evidence for
a Theravadin presence. The list of seven vipassanii fiiilJ-a-s has a
parallel in the Visuddhimagga, and an inscribed octagonal pillar from U
Tapao gives a set of verses on the four truths that are cited in that work
and in other works of the school. 17 The Visuddhimagga is, of course, one
of the most representative and most authoritative texts of the Mahavihara
15. See Skilling forthcoming (c) for references. The Theravadins traditionally
describe the language of their texts as Magadhl, "the language of Magadha":
see von Hiniiber 1994.
16. There are a very few orthographic variants, for which see e. g. Skilling
1992, 84-with reference to the work of von Hiniiber-and forthcoming (a).
17. See Skilling forthcoming (a) for references.
98 JIABS 20.1
Theravada. An inscription found in association with a giant pair of Bud-
dhapada at Amphoe Si Maha Phot in Prachin Buri province gives three
Pali stanzas in homage to the Buddha, Dhamma, and The stan-
zas, in the vasantatilaka metre, are from the Telaka.taha-gatha, a work of
unknown authorship believed to have been composed in Ceylon. Accord-
ing to the opening Khmer portion, the epigraph was set up by one
Buddhasiri in CE 761.
18
The sacca-kicca-kata-fia1)an; verse is known
only from late Theravadin texts: it is noteworthy that the Siamese inscrip-
tions (the verse occurs several times) are much earlier than the known
texts that give the verse.l
9
From the point of view of both language and contents, I conclude that
the Pali inscriptions of Burma and Siam give firm evidence for a
Theravadin presence in the Irrawaddy and Chao Phraya basins, from
about the 5th century CE onwards.
2o
From the extent and richness of the
evidence it seems that the Theravada was the predominant school, and
that it enjoyed the patronage of ruling and economic elites.
21
But I do not
mean to suggest that religious society was monolithic: other schools may
well have been present, or have come and gone, and there is ample evi-
dence for the practice of Mahayana and Brahmanism in the region. 22
18. See Charuk nai prathet thai 2529, I: 179-86 and Rohanadeera 1988. The
Telaka.tilha-giltha was edited by Edmund R. Goonaratne (1884).
19. See Skilling forthcoming (a) for references.
20. We must wait for a comprehensive study of Indic loan-words in early Mon
inscriptions from Siam before we can determine the degree to which they use
Sanskrit or Pali. An example of the former is the word pU1)ya, ubiquitous in the
epigraphs. A possible example of the latter is the term upiljhily, derived more
probably from Pali upajjhilya (also upajjha and upajjhil) than Sanskrit
upiidhyilya, in an inscription from Lopburi:see Credes 1961, 8, II (1). Another
form, from two ca. 9th century "votive tablets" is pajhily: Charuk nai prathet
thai 2529, II: 85-89, 90-94 (note that the word occurs side-by-side with
ilcilryya ).
21. Stargardt (p. 200) remarks of the relic chamber of the "Khin Ba mound,"
the source of a 20-leaf golden Pali text: "although many other relic chambers
were discovered at SrI this was the only one to survive intact, and its
contents exceeded-in number, quality of workmanship, and concentration of
precious metals and stones--even the relic chamber of the Bhat):iprolu stilpa in
Andhra."
22. The practice of Mahayana is compatible with any of the Vinaya schools,
including the Theravada, and brahmans played (and continue to play) an active
role in South-east Asian "Buddhist" societies, both court and common. The
schools or religious groups should be regarded as interactive and complemen-
SKILLING 99
The Question of Origins
The Theravadin sa11Jgha of Ceylon was divided into two main rival
branches, the Mahaviharavasins and Abhayagirivasins. After more than a
thousand years of contention for legitimacy and patronage, the former
won out, and absorbed the monks and monasteries of the latter. Most
regrettably for our purposes, the literature of the Abhayagiri, which
included a chronicle of the school, was allowed (or perhaps encouraged)
to disappear, with the result that no undisputed PaIi text of the school sur-
vives.
23
The Theravada that we know today is the Mahavihara tradition,
as settled by the time of the prolific commentator Buddhaghosa in the 5th
century. The larer Pali literature of the sub-commentaries (fikils) and
manuals, although subject to further development and a variety of influ-
ences, also belongs to the Mahaviharavasin lineage.
Both schools maintained contacts with India: with Kaficipuram,
Andhradesa, and Magadha. Is there any evidence for the presence of
either school in early South-east Asia? The canonical inscriptions-
including the Abhidhamma "extracts"-could belong to either the
Abhayagirivasins or the Mahaviharavasins, since both are believed to
have transmitted a similar canon in PaIi, and both held broadly similar
tenets and used a similar technical vocabulary.24 It seems that the
Abhayagiri also transmitted the Pa.tisambhidiimagga, or at least a similar
text, since passages cited in the Vimuttimagga (for which see below) have
parallels in that work. The niibiidhaka11J yato dukkha11J verses, known at
present only from Mahavihara texts such as the Visuddhimagga, are given
in citation, and are not original to the works in question: that is, they
originate from an earlier text that may have been accepted by both
schools.
The Vimuttimagga, a treatise associated with the Abhayagiri, was well-
known outside of Ceylon (whether it was composed in that country or in
India remains under debate). A comprehensive manual of practice and
tary rather than mutually exclusive. For Avalokitesvara in South-east Asia see
Chutiwongs 1984 (especially ch. 3 on Burma and ch. 4 on Central Thailand)
and Chutiwongs and Leidy 1994; for brahmanism in the region see Dawee
1982.
23. See Skilling 1993a.
24. The canons of the two schools were not identical (and is it not historically
and humanly improbable, rather impossible, that two canons transmitted for
centuries from an early date-the Ahhayagiri was founded in the 1st century
BCE-at separate monastic centres should be so?): see the important refer-
ences in von Hiniiber 1995, 36-38.
100 nABS 20.1
theory, composed by Upatissa (Skt. perhaps by the 2nd century
CE, it was translated into Chinese in 515. Interestingly, the translator,
*San;tghabhara, was a of Funan (an early South-east Asian polity
known from Chinese sources, and located by the savants in' the deltaic
regions of Cambodia). 25 The manuscript of the Vimuttimagga, along with
the other texts translated by *San;tghabhara, was brought to China in 503
by another monk of Funan, *Mandrasena.
26
Since none of the other texts
brought from Funan are Theraviidin, and some belong to the Mahayana, 27
the; fact that the Vimuttimagga was among them attests only to the avail-
ability of that text in Funan: it cannot be interpreted as evidence for a
(non-Mahavihara) Theravadin presence. 28 Since *SaII).ghabhara did some
of his translation work in the "Funanese Pavilion,"29 and enjoyed the
patronage of the Emperor, it seems that Funanese Buddhism was
accorded some esteem. ;
(For insular South-east Asia, we have one clear piece of eVideuc:e: the
inscription from Ratu Baka in central Java, dated CE 792, which refers to
an "Abhayagiri-vihiira built for the Sinhalese sa17Jgha." On the mainland,
but outside of our period, there is mention of an Abhayagiri in the con-
cluding Khmer portion of a Vajrayanist Sanskrit palreograph, dated CE
1066, from the vicinity of Nakhon Ratchasima [Korat] in Central Siam. 30
The precise location of this Abhayagiri is unknown, and it is by no means
certain that the toponym should be related to the Abhayagiri school: the
inscription names only an "Abhaya Mountain" [giri: without the word
vihiira], where images of "Buddhalokesvara" and others were installed
and later renovated.)
25. For the school-affiliation (and name of the translator and date of transla-
tion, about which there has been some confusion) see Skilling 1994.
26. Li-tai.san-pao chi, T. no. 2034, 49.98c.6-7; Kai-yuan shih-chiao lu, T.
no. 2154,"'55.537c.18-l9. The Annals of the Liang Dynasty confIrm that Funan
was one of the countries that sent tribute in 503. I am grateful to
Vinita Tseng for checking the, Chinese sources.
27. The works are listed in Nanjio 1975, II 101, 102; Bagchi 1927,
18; Repertoire du canon bouddhique sino-japonais, Fascicule annexe du
HobOgirin (1978) 267 (s. v. "Mandarasen"), 281 (s. v. "Sogyabara").
28. The Vimuttimagga was also known in North India: the chapter on the
dhutanga-s was translated into Tibetan under the title DhutagulJ.anirdesa
around CE 800, and long sections were cited by DasabalasrImitra, a North
Indian scholar, probably in the 12th century, in a work preserved only in
Tibetan translation: see Skilling 1987, 1993b, and 1994 for references.
29. Bagchi 1927,416.
30. See Chirapat 1990, 12 (text line 32), 13 (tr.).
SKILLING 101
All told; there is no conclusive local evidence that the early Theravada
of South-east Asia was affiliated with either the Mahavihara or the
Abhayagiri. We may also note the absence of references to South-east
Asia of the period in the chronicles of Ceylon, 31 and reflect that in the
great period of reform that swept the region in the 14th and 15th centuries
the new ordination lineage was distinguished by the name Sfhala-siisana.
Might this not suggest that the old tradition did not associate itself with
Ceylon?
It is therefore probably futile to try to trace the Theravada of the period
to either of the Ceylon schools. It is likely that Buddhism arrived in the
area at an early date-perhaps even from the time of SOI}.a and Uttara's
mission to SuvaI}.I}.abhfimi during the reign of King Asoka, as traditionally
held. Whether this Buddhism belonged to the Theravadin lineage from
the start, or whether that lineage asserted itself later, cannot be said (and
what did the term Theravadin mean in the pre-Buddhaghosa period, and
outside of Ceylon?)-but there is no doubt that it evolved independently
of the Ceylon schools. Over the centuries it would have undergone mul-
tiple influences, as monks (and perhaps nuns) from different regions of
India criss-crossed the region, and as local monks travelled throughout the
region and to different parts of India. 32 There is evidence for connections
with Andhradesa and the South, for example in the layout of early Pyu
stiipas and vihiiras, such as those from Beikthano,33 There is also evi-
31. See here Ray 1939, 52. Sirisena (1978, 58) remarks that "Sri Lanka's close
religious contacts with Burma started only from the eleventh century." His
work offers a wealth of information-from chronicles, inscriptions-on the
relations between Ceylon and South-east Asia but, as the title indicates, all
from the later period.
32. IT anything is clear from the time of our earliest records-the
itself (e. g. the PUlJ1Joviida-sutta, MN 145)-up to the present, it is that monks
travelled, even in the face of adversity or danger. The subject is addressed by
Vasubandhu, who in his Vyakhyayukti gives in verse seven reasons why the
Buddha travelled (note the technical term, known from the canon, carikii11J
carati) and fifteen reasons why auditors (sriivaka) did so (Peking edition of the
Tibetan Tripitaka, vol. 113, cat. no. 5562, sems tsam si, 44b6 foIl.). The verses
are available in Sanskrit citation in Haribhadra 1960, 271.30 and 274.19.
33. See e. g. Stargardt 1995, 200, 205. It is intriguing that the dukkha11J
dukkhasamuppiida11J verse, inscribed at least twice in Siam, is also known (but
in a lightly Sanskritic form) from an inscription from Andhra: see Skilling
1991 and 1992 for details. The use of the Pa11ava script cannot in itself be cited
as evidence, since that script was employed from an early date throughout insu-
102 nABS 20.1
dence for contacts with North India: the influence of Gupta idioms on
DvaravatI Buddha images, and -the practice of enshrining the ye dhamma
verse or the pa,ticcasamuppada formula in stiipas, which was Widespread
throughout the North, but rare in the South
34
and Ceylon.
35
The
Telaka.taha verses suggest contacts with the latter country, as does, per-
haps, a short and enigmatic OldMon inscription from the Narai or Khao
Wong cave in Saraburi, dated to circa 12th century BE (CE 550-650),
which refers to an Anuradhapura.
36
Whether the reference is to the
ancient capital of Ceylon or to a local site cannot be said, although the
latter seems more likely: the important point is that the toponym is oth-
erwise known only from Ceylon.
37
lar, peninsular, and mainland South-east Asia, for secular and religious"(both
Brahmanical and Buddhistic) records. '
34. For some Southem examples in the Pallava script see Rea 1990, 149-80
and pIs. 51-64 (and also Mitra 1980, 218-20). The inscriptions that I am able
to decipher from the Stygian reproduction of the plates give the ye dharma
verse in Sanskrit. Rea describes the site as "one of the most remarkable groups
of Buddhist remains in the Presidency" (then in Madras, the site is now in
District Visakhapatnam of Andhra Pradesh). Further south, at Gummadidurru
(District Krishna) were found "127 clay tablets of the size of an eight-anna
piece and bearing the Buddhist creed in Nagari characters of the late medireval
period" (Archawlogical Survey of India, Annual Report, 1926-27. Rpr. Delhi:
1990, 155-56: see also Mitra 1980, 212).
35. That the practice was not unknown to the late Ceylon Theravada may be
seen from the Saratthaazpanf (a text some centuries younger than our examples
from the field), which defines a dhamma-cetiya as "[a cetiya] built after
depositing a book inscribed with conditioned arising, etc.": Mahlimakuta ed.,
vol. 1 (Bangkok), 2511 [1968], p. 263, ult pa,ticcasamuppadiidilikhitapottha-
kaTIJ. nida1}itva kataTIJ. pana dhammacetiyQ11J. nama. (I am grateful to the late U
Bo Kay of Pagan for the reference.) We may compare the definition with
Candragomin (6th-7th century CE?) as cited by Haribhadra (late 8th century)
in his Atoka (BST 4, 361.15) yatra hi nama pudgalanairatmya-dyotikaya ye
dharma hetuprabhava ityadigathaya adhi$.thito bhiibhiigaQ stiipo mataQ. For
some of the few ye dharma inscriptions known from Ceylon, see Mudiyanse
1967, 29-30 (in Nagan, on images that Mudiyanse, with good reason, deems
imported), 92-95 (in Sinhalese characters, possibly in Pili), and 97. Ceylon is
rich in deposited texts, but mostly in Sanskrit, and of mantra, dhara7Jf, or
Prajfiaparamita, rather than extracts from the Pili canon: see Mudiyanse 1967,
Schopen 1982, and von Hinuber 1984.
36. Charuk nai prathet thai 2529, ll:42-47.
37. That is, no other references are given in Monier-Williams 1976, 37c, or in
Malalasekera 1983, 83-85.
SKILLING 103
We shOlild not regard the establishment and development of Buddhism
in the region as a mere mechanical process: it was rather a human, and
hence unpredictable, progress in which decisions were made and acted
upon by individuals and communities. A single charismatic monk could
attract followers and sponsors of status to his school; a single ruler could,
whether for political, economic, or purely religious reasons, decide to
favour a particular san;gha.3
8
Changing trade routes or political alliances
could bring new patterns of patronage.
Perhaps because of the absence of indigenous information-of contem-
porary chronicles or histories-the Buddhism of early South-east Asia is
all too often portrayed as an inanimate cultural package that was passively
received from abroad. All the evidence, however, is against this. The
Buddhism of the Chao Phraya plain was not a simple copy from Ceylon
or India: from the time of the very first evidence, it already has a unique
face, implying an earlier evolution for which no records remain. The sur-
viving artifacts are expressions of a mature and refined culture, with spe-
cial features like the large and ornate stone dhammacakkas; the plan of
the stilpas or caityas, and the style of their stucco art; the style of the
Buddha images; the rich terracotta art (the so-called votive tablets); and
motifs that remain to be explained, such as the so-called Banaspati image.
From this evidence we can only deduce that the Buddhism of the Chao
Phraya valley is the flowering of a "local genius." The same may be said
of the Buddhism of the Pyu, which had its own architecture and terracotta
art, and local practices such as the urn-burial of people of status. The two
realms were flourishing centres of Buddhist culture in their own right, on
an equal footing with contemporary centres like Anuradhapura.3
9
To conclude, we may tum to Laos and Cambodia. Is there any evidence
of early Theravadin activity in these countries? Very little information is
available for Laos. In 1968 a standing stone Buddha in DvaravatI style,
38. That a single monastic could make enormous and enduring contributions
to a culture-in manifold aspects-may be seen from countries for which we
have records. Atisa and Bu ston spring to mind for Tibet, Kukai for Japan.
39. The situation was perhaps not much different from that of today, when the
Buddhisms of the Mon, Burmese, Central Thai, Shan, Lanna Tai, Lao, and
Khmer are each quite distinctive. We might also bear in mind that-from the
point of view of Madhyadda-Ceylon, Andhra, and South-east Asia were
equally foreign cultures, and that there is no valid reason to relegate the last-
named to a lower rank. In a sense "local" and "foreign" are modem constructs:
the South-east Asian cultures that adopted Indian cosmology did not hesitate to
place themselves within Jamburupa.
'104 JIABS 20.1
190 cm. in height, was found at Ban Thalat in Vientiane province. The
image and the accompanying Mon inscription have been dated to the 7th-
8th centuries.
40
The finds suggest that the Mon Buddhism of the right
bank of the Mekhong River (the Mun and Chi valleys) also spread to the
left bank, but much more research needs to be done into the nature of the
Buddhism of the middle Mekhong valley before anything more can be
said.
In Cambodia-which is rich in structural remains and lithographs-no
ancient Pali inscriptions have been found, and scriptural extracts of the
type discussed above are unknown, with one exception. This is an epi-
graph of two lines, engraved in small "pre-Angkorian" letters on the back
of a standing Buddha image (90 cm. in height) from Tuol Preah Theat in
Kompong Speu province (now in the Musee Guimet).41 The text reads: 42
ye dhamma hetuprabhava tesal1J hetul1J tathagato avaca
tesafi ca yo nirodho eval1Jvadf mahasamano.
The verse differs from the PaIi of the Mahiivagga (Vinaya 140) in giving
hetuprabhava for hetuppabhavii and avaca for aha, and cannot be cited as
evidence for a Theravadin presence. 43 Otherwise, the earliest PaIi
inscription dates from CE 1308-and thus belongs to the heyday of the
"Theravadin renaissance" in Ramafifiadesa, Burma, Central Siam, the
Lanna Kingdom, and other northern principalities. 44
40. Boun Souk 1971,14 (with photograph); Vothu Tinh 1983,42-43.
41. It is not without interest that the ye dhamma verse is also inscribed (in
Pilli) on the back of a standing DVilravatI-styIe Buddha image (196 cm. in
height) from Ratchaburi, dated to ca. 12th century BE (CE 550-650): see
Charuk nai prathet thai 2529,1:72-74. Another DvaravatI Buddha image with
a (fragmentary) Pilli ye dhamma inscription "en caracteres preangkoriens peu
soignes" is in the Korat Museum: "Inscription sur une statue de Buddha du
Musee de Karat," in Ccedes 1964, 162.
42. Ccedes 1964, 108. The image is illustrated in Dupont 1955, PIs. 45 B and
46C.
43. Note that there are many examples of the ye dharma verse in a mixed or
Sanskritic Pilli from India, and that they have yet to be subjected to sustained
linguistic and palreographic analysis.
44. Ccedes 1989, 282-89. The inscription is a royal record of a religious foun-
dation, and not a scriptural extract.
SKILLING 105
There is certainly evidence of the presence of Buddhism in the early
period: stone, metal, and wooden images of the Buddha,45 of Maitreya, 46
and of Avalokitesvara,47 and occasional mention in Sanskrit or Khmer
dedicatory inscriptions. Chinese sources record that monks travelled back
and forth between Funan and the Middle Kingdom, but say nothing about
their school-affiliation. The Vimuttimagga and other Buddhist texts,
including some of the Mahayana, were sent to China from Funan in the
early 6th century. The opening verses of the Telaka.tiiha-giithii are known
from an 8th century inscription from Prachin Buri, which may be said to
belong to the Khmer cultural sphere. Furthermore, some of the early
Buddha images of Cambodia are stylistically affiliated to those of
Dvaravafi. On the other hand, it is remarkable that in Cambodia there are
no ruins of monumental brick stiipas, so common in Pyu and Mon areas,
or even of smaller complexes of votive stiipas. Boisselier has noted that
none of the ancient epigraphs refer to stiipas, and that none of the known
stiipa remains are earlier than the 12th century.48 Nor is there any evi-
dence of a practice shared by Pyu and Mon Buddhists: the mass-produc-
tion from moulds of clay "votive tablets." Here too Boisselier remarks
that these pra!;, patima are not well-attested until the 12th century.49 In
sum, while Buddhists were certainly active in Cambodia during the early
period, it seems that the dominant ideology remained that of the brah-
mans, and that Buddhism or Buddhistic culture did not flourish among the
Khmer to the degree that it did among the Pyu and the Mon.
BIBLIOGRAPHY
References to Pali texts are to the editions of the Pali Text Society, by page and
line or by verse. BSR = Buddhist Studies Review (London); IIJ = Indo-Iranian
Journal; JPTS = Journal of the Pali Text Society (Oxford); JSS = Journal of the
Siam Society (Bangkok).
Aung Thaw. 1972. Historical Sites in Burma. [Rangoon].
Bagchi, Prabodh Chandra. 1927. Le canon bouddhique en Chine: Les traduc-
teurs et les traductions. Tome 1. Paris.
45. See Dupont 1955, 189-210.
46. See the examples in Chutiwongs and Leidy 1994, and Dupont 1955, pIs.
29 A and 30 A.
47. For examples see Chutiwongs 1984 (chap. 5), Chutiwongs and Leidy
1994, and Dupont 1955, pIs. 12 B, 22 AB, 28 A, 29 B, 30 B, and 31 A.
48. Boisse1ier 1966, 97.
49. Boisselier 1966, 300. For "Saintes Empreintes" in Cambodia, see
Boisselier's 219, 256-57, 303, and Fig. 70.
106 nABS 20.1
Bareau, Andr6. 1955. Les sectes bouddhiques du Petit Yehicule. Publications
de l'Ecole d'Extreme-Orient 37. Paris.
Bauer, Christian. 1991. "Notes on Mon Epigraphy." JSS 79.1: 31-83.
Boisselier, J[ean]. 1966. Le Cambodge. Manuel d'arch6ologie d'Extreme-
Orient, Premiere Partie: Asie du Sud-Est, Tome I. Paris.
Boun Soule, Thao. 1971. Eimage duBuddha dans rart lao. Vientiane ..
Charuk nai prathet thai. 2529. Bangkok
Chirapat Prapandvidya. 1990. "The Sab Bak Inscription: Evidence of an Early
Vajrayana Buddhist Presence in Thailand." JSS 78.2: 10-14.
Chutiwongs, Nandana. 1984. The Iconography of Avalokitesvara in Mainland
South East Asia. Diss. U. of Leiden.
Chutiwongs, Nandana, and Denise Patry Leidy. 1994. Buddha of the Future.
New York and Singapore.
Ccedes, George. 1961. Recueil des Inscriptions du Siam, Deuxieme Partie:
Inscriptions de Dvaravatf, de C;rlvijaya et de Lavo . Bangkok.
----.1964. Inscriptions du Cambodge. Vol. 7. Paris.
----. 1989. "La plus ancienne inscription en pali du Cambodge." Articles
sur Ie pays khmer. Paris: 282-89 (= Etudes cambodgiennes 32, originally
published in BEfEO 36).
Dawee Daweewarn. 1982. BrahmmJism in South-East Asia (From the earliest
time to 1445 A.D.). New Delhi .
. Dupont, Pierre. 1955. La statuaire preangkorienne, Ascona.
Haribhadra. 1960. Abhisamayiilal1Jkariiloka. Ed. P. L. Vaidya. Buddhist San-
skrit Texts 4. Darbhanga.
von Hinuber, Oskar. 1984. Sieben Goldbliitter einer Pancavil1Jsatisiihasrikii
Prajnapiiramitii aus Anuradhapura. G6ttingen.
----. 1985. "Epigraphical Varieties of Continental Pali from Devnimori
and Ratnagiri" Buddhism and its Relation to Other Religions: Essays in
Honour of Dr. Shozen Kumoi on his Seventieth Birthday. Kyoto: 185-200.
----. 1994. "On the History of the Name of the Pili Language." Selected
Papers on Pali Studies. Oxford: 76-90.
----.. 1995. "Buddhist Law According to the Theravada-Vinaya: A Sur-
vey of Theory and Practice." JIABS 18.1: 7-45.
Goonaratne, Edmund R., ed. 1884. "Telakataha-gatha." JPTS. Rpt. vol 1
(London: 1978) 49-68.
Luce, G. H. 1974. "The Advent of Buddhism to Burma." Buddhist Studies in
Honour of l. B. Homer. Eds. L. Cousins et al. Dordrecht and Boston: 119-
38.
Malalasekera, G. P. 1983. Dictionary of Piili Proper Names. VoL 1. New
Delhi. 1st ed. 1937.
Mitra, Debala. 1980. Buddhist Monuments. Calcutta. 1st pub. December 1971.
Monier-Williams, Monier. 1976. A Sanskrit-English Dictionary. Delhi. 1st ed.
Oxford: 1899.
Mudiyanse, Nandasena. 1967. Mahayana Monuments in Ceylon. Colombo.
SKILLING 107
Nanjio, Builyiu. 1975. A Catalogue of the Chinese Translation of the Buddhist
Tripi.taka, the Sacred Canon of the Buddhists in China and Japan. San
Francisco. 1st ed. Oxford: 1883.
Ray, Nihar-Ranjan. 1939. "Early Traces of Buddhism in Burma." Journal of
the Greater Indi.a Society 6.1 (Jan., 1939): 1-52.
Rea, A. 1990. "A Buddhist Monastery on the Sallkaram Hills, Vizagapatam
District." Delhi. 1st ed. Archfological Survey of India, Annual Report,
1907-8.
Rohanadeera, Mendis. 1988. "The Noen Sa Bua Inscription of Dong Si Maha
Bo, Prachinburi." JSS 76: 89-99.
Schopen, Gregory. 1982. "The Text on the 'DharaI)i Stones from Abhaya-
giriya': A Minor Contribution to the Study of Mahayana Literature in
Ceylon." JIABS 5.1: 100-08.
Sirisena, W. M. 1978. Sri Lanka and South-east Asia: Political, Religious and
Cultural RelationsfromA.D. c. 1000 to c. 1500. Leiden.
Skilling, Peter. 1987. "The SaIl).slqtasaIl).sl.qta-viniscaya of Dasabalasrimitra."
BSR 4.1 :3-23.
---.1991. "A Buddhist Verse Inscription from Andhra Pradesh." //J34:
239-46.
----. 1992. "Preliminary Report on a Recently Discovered PaIi Inscrip-
tion." Warasan chotmaikhao samnak-Iekhanukan Somdetphrasangharat.
Vol. 1, No.1, Oct.-Dec. 2535 [1992]:83-86; revised version under the title
"A Recently Discovered Pali Inscription from Nakhon Pathom," forthcoming
in JPTS.
----. 1993a. "A Citation from the * Buddhavan:tsa of the Abhayagiri
School." JPTS 18: 165-75.
----. 1993b. "Theravadin Literature in Tibetan Translation." JPTS 19:
69-201.
----. 1994. "Vimuttimagga and Abhayagiri: the form-aggregate according
to the San:tskrtasan:tkrtaviniscaya." JPTS 20: 171-210.
----. Forthcoming (a). "Pilii Inscriptions on a Stone Dhammacakka and
an Octagonal Pillar from Chai Nat." Forthcoming in JPTS.
----. Forthcoming (b). "A Paritta Inscription from SrIlq;etra in Burma."
Forthcoming in JPTS.
----. Forthcoming (c). "On the School-affiliation of the 'Patna Dhamma-
pada'." Forthcoming in JPTS.
Stargardt, Janice. 1995. "The Oldest Known Pali Texts, 5th-6th century:
Results of the Cambridge Symposium on the Pyu Golden Pali Text from Sri
K ~ e t r a , 18-19 April 1995." JPTS 21: 199-213.
Supaphan na Bangchang. 2529 (1986). Wiwathanakan ngan khian phasa bali
nai prathet thai: charuk tamnan phongsawadan san prakat. Bangkok.
U Tha Myat. 1963. Pyu Reader. Rangoon.
Vothu Tinh. 1983. Les origines du Laos. Paris.
ELI FRANCO
Distortion as a Price for Comprehensibility?
The rOyal tshab-Jackson Interpretation of Dharmakirti
Is Enlightenment Possible? Dharmaklrti and rGyal tshab rje on Knowl-
edge, Rebirth, No-Self and Liberation. Introduced, translated and armotated
by Roger R. Jackson. Snow Lion Publications. Ithaca, New York, 1993.
571 pages.
The PramiilJasiddhi chapter, the second chapter of the PramiilJaviirttika, is
unique in Dharmaldrti's writings. DharmakIrti (ca. 600-660) is the sort of
author who writes on the same issue several times, elaborating and refIning
his thoughts, sometimes modifying them radically. Of course, the major
and general subjects of classical Indian epistemology, namely, perception
and inference, are treated in one form or another in all of Dharmaldrti's
writings, but there are also some specific topics, such as the determination
of vyiipti, that run like a leitmotif through his work. 1 In stark contrast to
that, religious issues are dealt with nowhere else but in the PramiilJasiddhi
chapter. This chapter therefore stands apart as representing the only period,
early in his career,2 in which Dharmaldrti wrote on religious issues (albeit
in a philosophical marmer) such as karma and rebirth, modes of meditation,
the four noble truths, the Buddha's compassion and path to enlightenment,
etc.
Although two monographs and a number of important papers have been
written on the PramiilJasiddhi chapter,3 it has received far less attention
1. Cf. E. Steinkellner, "Remarks on niscayagrahalJa," Orientalia, Iosephi Tucci
Memoriae Dicata, eds. G. Gnoli and L. Lanciotti, Serie Orientale Roma 56.3
(Roma: 1988) 1427-1444.
2. Cf. E. Frauwallner, "Die Reihenfolge und Entstehung der Werke
DharmakIrti's," Asiatica, Festschrift F. Weller (Leipzig: 1954) 142-154 (= Kleine
Schriften, pp. 677-689).
3. Cf. T. Vetter, Der Buddha und seine Lehre in Dharmaklrtis Pramii1}a-
viirttika, Wiener Studien zur Tibetologie und Buddhismuskunde 12 (Wien:
1984); V. A. van Bijlert, Epistemology and Spiritual Authority, Wiener Studien
109
110 JIABS 20.1
than Dharmaldrti's work on inference, on which Steinkellner published his
pathbreaking editions, translations and studies in the sixties and seventies,4
influencing decisively the course of Dharmaldrtian studies for ,many years
to come. Therefore, Roger Jackson's voluminous book, which has been in
the making for more than ten years
5
and which contains a complete English
translation of the PramillJasiddhi chapter as well as the important commen-
tary ofrGyal tshab dar rna rin chen (l364-1432) thereon, the rNam 'grel
Thar lam gSal byed, should have been a major event in DharmakIrtian
studies.
Why this is not the case is due, I believe, above all to Jackson's'lack of
interest in Dharmaldrti. This may sound paradoxical in view of the book's
title, but it soon becomes obvious as one reads through the translation.
Besides, Jackson himself states this fact in no uncertain terms several times,
and he should be recommended for at least making his stand c1ear.
6
Jackson's interest centers on rGyal tshab's commentary; the translation of
the verses is not meant to be faithful to their Sanskrit original, not even to
their Tibetan translation, but only to their interpretation by rGyal tshab, as
understood by Jackson. The work as a whole is also symptomatic for a
current trend among scholars of Tibet who attempt to understand the
Tibetan philosophical tradition "as such" independently of the decisive
zur Tibetologie und Buddhismuskunde 20 (Wien: 1989); a number of papers
mostly dealing with the initial verses of the chapter appeared in E. Steinkellner
(ed.), Studies in the Buddhist Epistemological Tradition (Proceedings of the
Second International DharmakIrti Conference, Vienna, June 11-16, 1989),
Beitrage zur Kultur- und Geistesgeschichte Asiens 8 (Wien: 1991). One should
also mention the pioneering work of T, Vetter, Erkenntnisprobleme bei
Dharmak"frti (Wien: 1964),
4. Cf. E. Steinkellner, Dhannakfrtz"s Hetubindul;. Teil I. Tibetischer Text und
rekonstruierter Sanskrittext. Teil II. Ubersetzung und Anrnerkungen. VerOf-
fentlichungen der Kornission fur Sprachen und Kulturen Siid- und Ostasiens 4-5
(Wien: 1967); Dhannakfrtfs Prami'ilJaviniscaya. 2. Kapitel: Sv1irthanumanam.
Teil I. Tibetischer Text und Sanskrittexte. Teil II. Ubersetzung und
Anrnerkungen. Ver6ffentlichungen der Kornission fiir Sprachen und Kulturen
Siid- und Ostasiens 12, 15 (Wien: 1973, 1979). On the other hand, the
pratyak.$a chapter is relatively neglected at the present, and I know of no one
who currently works on it.
5. Cf. R. R. Jackson "Is Enlightenment Possible? An Analysis of Some Argu-
ments in the Buddhist Philosophical Tradition, with Special Attention to the
PramillJasiddhi Chapter of DharmakIrti's PramillJavi'irttika," diss., University of
Wisconsin-Madison, 1983.
6. Cf. pp. 11-12, 161, and with a different nuance p. 152.
\
FRANCO 111
background and long-lasting influence of the Indian tradition. This trend is
perhaps understandable as a reaction to the type of scholarship that was
dominant until recently, namely, to study only canonical Tibetan texts to
gain access to the Indian Buddhist philosophical tradition and as an ancil-
lary tool for its understanding. This tradition, which enjoys an illustrious
past, is fortunately not yet defunct, but has certainly been marginalized,
especially in North American Tibetology. Moreover, the tendency to view
and evaluate the Tibetan tradition within the context and on the basis of the
Indian tradition has been replaced by a penchant to present and interpret
Indian materials through the eyes of their Tibetan exegetes, past and pre-
sent. However, the results of this new, innovative but limited approach are
largely flawed, both factually and methodologically. This was clear already
some twenty years ago when scholars like Hopkins made their first contri-
butions in this field,7 and it should have become even clearer by now.
The fact that Jackson's dealings with Dharmakirti are at best perfunctory
is clearly illustrated by his treatment of previous scholarship on
Dharmakirti. This is, no doubt, the most shocking part of the book. Thus,
in a short and self-destructive chapter called "Scholarship on Dharmakirti"
(pp. 149-152) Jackson clearly shows that he has the habit of not only
referring to, but also commenting on, books about which he has only the
foggiest idea. Here are a few examples: Jackson recommends Potter's Bib-
liography of Indian Philosophies "not the least of whose virtues is its
inclusion of Japanese scholarship" (p. 149). Now, Potter's tremendous and
admirable work has many virtues, but perhaps its single most important
drawback is the exclusion of Japanese scholarship. Similarly, Frauwallner's
Geschichte def indischen Philosophie (1953) is said to "remain a treasury
of information on Indian and Buddhist logic and epistemology." However,
these topics are not even touched in that book. Stcherbatsky is credited
with translating only "portions of Dharmakirti's Nyiiyabindu." Similarly,
SteinkeUner translated the Hetubindu "in part." Both translations are com-
plete. On the next page (p. 151) even the edition of the Hetubindu by
Steinkellner seems to be incomplete ("the parts of the Hetubindu and Pra-
miif}lviniscaya edited by Steinkellner ... "). Does Jackson know of por-
tions of NyiJ.yabindu and Hetubindu that no one else is aware of? Most
readers may be surprised to find out that "Ganganatha Jha's translation of
Tattvasangraha, together with KamalasIla's PafijikiJ. (1937-
7. Cf. J. W. de Jong's review of J. Hopkins and Lati Rimpoche with A. Klein,
The Precious Garland and the Song of the Four Mindfulnesses. in Indo-Iranian
]oumal20 (1978) 136-139.
112 JIABS 20.1
39), remains one of the finest and ~ o s t useful sources for our understand-
ing of late Buddhist ontology and epistemology," or that "[t]he only general
text on Buddhist thought that contains reliable information on !he prama:(ta
tradition is that of A. K. Warder (1980)." One may ormay not agree with
the last two judgements, but other statements of Jackson are simply wrong,
no matter how generously one may wish to consider them. Thus, Gnoli
(1960) (i. e., his text edition with critical notes, entitled The Pra-
mttr;avarttikam of Dharmaklrti. The First Chapter with the Autocommen-
tary) is supposed to have translated "a significant portion of the Sviirthii-
numiina chapter." On the other hand, Satkari Mookerjee (1968) (The
PramiilJaviirttikam of Dharmakfrti [Sviirthiinumiina chapter. verses l-
SI]) is said to have edited the PramiilJaviirttika. However, the partial San-
skrit text appearing at the end of the book is a mere reprint, whereas
Mookerjee's main achievement, the translation, in collaboration with Hojun
Nagasaki, of the first fifty-one verses and extracts from KaTlJakagomin's
commentary, is never mentioned.
All these examples are taken from the above mentioned chapter, but there
is more in the same vein throughout the book. Let me point out just a few
more examples. Ravigupta's commentary is not on PrajiHikaragupta's
PramiilJaviirttikalmikiira (as claimed in pp. 114-115), but on DharmakIrti's
PramiilJaviirttika. This is worth mentioning because Jackson's misrepre-
sentation is based on Stcherbatsky's Buddhist Logic, and therefore a widely
repeated mistake. Incidentally, the name of Devendrabuddhi's commentator
is Sakyabuddhi (not Sakyabodhi, as on pp. 114-115). In the introduction
to his glossary (p. 503), Jackson refers the reader to Nagatomi (1957) part
D and Steinkellner (1967ref) for "a complete index of Dharmaldrti's San-
skrit and Tibetan terminology." However, Steinkellner's index does not
contain a word of Sanskrit, and can neither be used for Dharmaldrti's
"Tibetan terminology," as it is a piida-index to the Tibetan translation of
DharmakIrti's verses. As for the index in Nagatomi's dissertation, it covers
only the PramiilJasiddhi chapter. On the other hand, Miyasaka's complete
index to the PramiilJaviirttika (Sanskrit-Tibetan and Tibetan-Sanskrit) is
not mentioned (Acta Indologica 3, 1973-1975, and 4, 1976-1979). (After
the completion of this review there appeared M. Ono, J. Oda and J.
Jackson, KWIC Index to the Sanskrit Works of Dharmakzrti. Lexicological
Studies 8 [Tokyo: Tokyo University of Foreign Studies, 1996].) It would
be tedious to enumerate the more trivial mistakes, such as wrong years of
publication, etc. However, it seems symptomatic that the name of a promi-
nent Dharmaldrti scholar and translator, Tilmann Vetter, is rendered "F.
Tillmann Vetter" in the Preface (p. 11) and "Vetter, F. Tillman" in the Bib-
\
FRANCO 113
liography (p. 524); this is obviously a contamination with 'Tom J. F.
Tillemans." On p. 151 at least ''P.'' is dropped (''Tillman Vetter'), and on p.
533 Jackson finally hits the mark with ''Tilmann Vetter," only to regress to
"Vetter, F. TilImann" in the Index (p. 570)!
To return to DharmakIrti's "root-text." Jackson had at his disposition a
copy of Nagatomi's translation of the Pramii1J-asiddhi chapter, It would
have sufficed, at least in the majority of cases, simply to quote Nagatomi's
quite literal and accurate translation in order to provide a good framework
for rGyal tshab's commentary. However, Jackson decided to present the
verses only through rGyal tshab's eyes, and since rGyal tshab did not read
the original Sanskrit text, Jackson too decided to ignore it altogether and
sometimes even the Tibetan translation, in order to produce a translation
which is "geared toward rGyal tshab rje's interpretation, which does not
always accord with a straightforward reading of the Tibetan verses them-
selves" (p. 12). The results of this approach are quite disastrous and the
translation fails the reader on all three levels. It is not faithful to the San-
skrit original, nor to Sa skya and Sakyasrlbhadra's outstanding Ti-
betan translation-Jackson would admit that in general, but he does not
suspect how often this is the cases-nor even to rGyal tshab rje's
interpretation.
The initial verses of the Pramii1J-asiddhi chapter are notorious for their
difficulty. Yet they have been translated into English several times, and
three of the translations, by N agatomi, Katsura and van BijIert, even appear
in Jackson's bibliography. Had Jackson relied on any of these, he may not
have solved all problems of interpretation-something which cannot be
expected at the present stage of Dharmaldrtian studies-, but he could cer-
tainly have presented a reasonable translation. Instead we get an often
misleading concoction out of bits and pieces of rGyal tshab's commentary
on the verses. Worse, since usually there is no note to the contrary, we
have to assume that Jackson's translation also represents the literal meaning
of Sa translation. Let us take a look at the flISt verse:
pramii1J-am avisa71J.viidi jfiiinam, arthakriyiisthitilJ I
avisa71J.viidana71J., siibde 'py abhipriiyanivedaniit II
tshad ma bslu med can ses pa I
8. For he says (p. 12): "I have tried to indicate those places where rGyal tshab
rje significantly departs from the straightforward reading [of the Tibetan transla-
tion]."
'114 nABS 20.1
don byed nus par gnas pa ni
mi slu sgra las byun ba yan II
mnon par' dod pa ston phyir ro I
Jackson's translation:
AUTHORITY IS NON-DECEPTIVE COGNITION;
[NON-DECEPTIVENESS] ABIDES WITH REGARD TO [CAUSAL]
EFFICIENCY.
NON-DECEPTIVENESS ARISES FROM WORDS, TOO,
BECAUSE THEY SHOW A MANIFEST DESIRE [TO SPEAK].
My translation:
A means of knowledge is a cognition which does not belie [its promise].
Non-belying [means] standing fIrm in respect to effIcient action.
[Non-belying may occur] in verbal [cognition] too,
because it communicates the intention [of the speaker].
First, it is unlikely that sthitil gnas here means "to abide." Following
PrajiUikaragupta, who glosses it with avicalana,9 I take it in the sense of "to
stand firm," i. e., not to deviate. That is, when a cognition is correct, it
leads, or more precisely: has the capacity of leading, to effIcient action. In
any case, non-deceptiveness is a property of a cognition or of a means of
knowledge, it does not reside or abide "with regard to" [= in?] "[causal]
effIciency." More importantly, there is absolutely no evidence that rGyal
tshab (or Sa pal].) used gnas in the sense of "to abide." rGyal tshab (gSal
byed 229.18-20) merely states that ... btso bsreg la sogs pa'i don byed
nus par ran gis ji ltar gial ba Itar gnas pa ni las la mi slu ba yin la I de ran
gi no bo myon tsam gyis rtogs pa'i nes pa med pa'i phyir I. Jackson trans-
lates somewhat awkwardly and partly wrongly (pp. 176-177): " ...
because how could there ABIDE an auto-comprehension with REGARD TO
THE [CAUSAL] EFFICIENCY of cooking, burning, etc.? Non-deceptiveness
is in regard to [potential confrrmatory] action; it definitely is not cognized
merely by an experience of [a cognition's] own essence." One can clearly
see that nus parr) is not translated, that ran gis [ji ltar] gial ba ltar cannot
be interpreted as a compound "auto-comprehension," which serves as the
subject of gnas pa (as final predicate), and that nes pa is not an adverb
("definitely"). More misleading, however, is Jackson's explanation of the
above (n. 2 thereon):
9. Cf. PVA 4.4: tasyii!:t sthitir avicalanam avisal1Jviidanal1J vyavasthii vii.
FRANCO 115
A cognition of a patch of blue is immediately and self-evidently authoritative,
so the apperception of that cognition cognizes not only the cognition's con-
tents, but its non-deceptiveness, too. On the other hand, a cognition of fire on
a distant hill is subject to subsequent confirmation or disconfmnation, and thus
is not immediately and self-evidently authoritative.
This is not what Dharmaklrti and rGyal tshab are saying. The distinction
they make is not between the perception of a blue object etc., and the infer-
ence of fire. Both these types of cognition have to be confirmed by effi-
cient action. The distinction is rather between apprehension of an object
and self-apprehension of a cognition; and it is only the self-apprehension of
the cognition which need not be confirmed. rGyal tshab clearly refers to
vv. 4d-5a: svariipasya svato gati/:L II prarnlbJ,yan; vyavahare1J,a. "The cog-
nition's own form is apprehended by [the cognition] itself. [Its] validity [is
determined] by everyday practice." However, Jackson seems to have
missed the reference, perhaps because he translates vv. 4d-5a as follows
(p. 180): "COGNITION [EXPERIENCES] ITS OWN ESSENCE FROM ITS
OWN [SIDE]; AUTHORITATIVENESS [IS COGNIZED] THROUGH DESIG-
NATION." It goes without saying that validity or authoritativeness cannot be
cognized by a mere "designation," that vyavahara and its Tibetan translation
tha sfiad do not mean "designation" here, and that rGyal tshab does not
interpret tha sfiad in this sense. In fact, rGyal tshab says correctly (gSal
byed p. 232.8-10: ... tshad rna yin pa'i cha de dus phyis byuri gi don byed
snari can gyi tha sfiad pa'i tshad rna fiid las rtogs dgos par rnthori ba'i
phyir I. But Jackson mistranslates again (p. 182): " ... because we see that
the authoritativeness part must be cognized through authoritativeness that is
a designation that has an apparent object that arises at a later time." Clearly,
one should translate " ... through the means of knowledge that consists in
everyday practice in which efficient action appears (lit. which has the
image of efficient action) .... " The main problem in Jackson's translation
is due to a misunderstanding of the word snari, "appearance," which like the
Sanskrit word i'ibhi'isa (not abhasa, as in the glossary on p. 508, under
"fallacy") can be used in the sense of "false" (e. g., in hetvabhi'isa, ''false
reason," i. e., something which only has the appearance of a reason, but is
not really a reason)-hence Jackson's translation as "apparent (object)."
However, in the present context snari does not mean "appearance" in any
pejorative sense, but simply the appearance by which the cognition is
characterized, i. e., the image that appears in it, i. e., its inner object or its
content.
116 JIABS 20.1
rOyal tshab refers here explicitly to Devendrabuddhi's explanation, and
Jackson notes rather vaguely the reference as PVP 220/3-4. I can only as-
sume that he has in mind PVP 220dl-2, which, however, d o e ~ not agree
with his explanation: '0 naji ltar tshad ma fiid nes par bya ie na I tha sfiad
gyis ni tshad ma fiid I dus phyis 'byun ba can gyi don byed pa'i yul can gyi
ses pas so" "[Objection:] How can validity be determined? [Reply:] 'Valid-
ity [is determined] by everyday practice,' [that is] by a cognition which is
characterized by an object that consists in an efficient action which arises at
a later time." However, Jackson misunderstands both Devendrabuddhi and
the paraphrase by rOyal tshab. It is not the case that (p. 182, n. 15)
"Devendrabuddhi's point (PVP 220 / 3-4: rOyal tshab rje has paraphrased)
seems to be that the authoritativeness of any cognition (even an appercep-
tion) is dependent upon a subsequent designation for its ascertainment as
authoritative." Besides the fact that tha sfiad / vyavahiira could not possibly
mean "designation" here, neither Devendrabuddhi nor rOyal tshab nor, of
course, Dharmaklrti claims that "apperception" (or better, self-perception or
self-apprehension, as are the more usual, more transparent and more accu-
rate translations for svasa11Jvedana, ran rig and similar expressions)
depends on anything that arises subsequently. The point is simply that the
existence of a cognition is perceived immediately by self-apprehension and
needs no further collaboration or confirmation; the existence of an object
like something blue, on the other hand, has to be confirmed by a subse-
quent efficient action in everyday practice.
So much for the first half of the verse. Concerning its second part, let me
just note briefly that the translation "THEY (i. e., words) SHOW A MAN-
IFEST DESIRE [TO SPEAK]" is not only inaccurate, inasmuch as words
cannot a show a manifest desire to speak, but also mistaken, inasmuch as
abhipriiya means "intention," i. e., what a speaker wishes to express with
specific words, not "manifest desire [to speak]," i. e., the mere fact that a
speaker wishes to speak. Also, "manifest" is redundant even in Jackson's
interpretation. His quaint translation obviously results from a too literal
understanding of Sa par;t'S partially mechanical translation of this frequent
and unproblematic technical term with moon par (= abhi-) 'dod pa. I doubt
that the correct interpretation of the second half of the verse is that (pp. 177-
178, n. 4) "[a]n aural cognition is an authority simply because it non-
deceptively apprehends a word, or sound, that a speaker desires to
express." As words or sounds are not "expressed" (only their meanings are
expressed), I assume that Jackson means "a word, or sound, that a speaker
desires to pronounce or voice." However, it is not the word as mere appre-
hended sound that is the issue here-its apprehension would be a simple
. \
1
FRANCO 117
case of sense perception like the perception of smell, flavor,
etc. It is rather the word as meaningful sound that is involved here: the
means of knowledge that has its meaning as its object is called sabda, and it
is included in inference (anumiina). DharmakIrti explains his intention in
the next verse (2), where Jackson switches over to the correct understand-
ing of "word" as meaningful sound (cf. ''THE OBJECT THE SPEAKER [DE-
SIRES TO EXPRESS]" in his translation of the first half of the verse).
However, the translation of this verse is not only completely removed from
the Sanskrit text, but also from Sa paJ)'s clear and faithful Tibetan transla-
tion. Jackson presents us with a largely meaningless conglomeration, made
up of the individual words of the Tibetan translation without regard for
their syntactical connections. I do not think that it is necessary to comment
in detail on it.
yo 'rtho buddhau prakiisate I
priimii1Jyal1J tatra sabdasya niirthatattvanibandhanam II
smra ba po yi byed pa'i yulll
don gari blo la rab gsal ba I
de la sgra ni tshad ma yin II
don gyi de Hid rgyu can min II
My translation:
The validity of a word relates to the thing that forms the object of the
speaker's activity [and] appears in the cognition [of the hearer] (i. e., to
the meaning of the word); it does not depend on the reality of [that]
object.
Jackson's translation:
WORDS IN THE [SASTRAS] ARE NOT CAUSE-POSSESSING REAL
OBJECTS, AUTHORITIES; THEY BRING ABOUT
IN THE MIND A CLEAR [APPEARANCE OF] SOME OBJECT
THAT IS THE OBJECT THE SPEAKER [DESIRES TO EXPRESS].
Sometimes, I must admit, I have great difficulties understanding the
peculiar mode of expression chosen by Jackson. Verses 5d-6, relating to
the second characterization of a means of knowledge in 5c, are a good
example (p. 184):
118 nABS 20.1
[AUTHORITY] ALSO IS THE ELUCIDATION OF AN OBJECT NOT
COGNIZED [BEFORE].
AFTER ONE HAS COGNIZED THE OWN-NATURE [OF AN OBIpCT],
(6) ONE ATTAINS THE COGNITION OF THE GENERIC [IMAGE].
"S OME COGNITION OF AN OWN-MARK
NOT COGNIZED [BEFORE]"-THE QUOTATION'S INTENT IS THAT
INVESTIGATION [REVEALS] THE OWN-MARK.
Is this translation intelligible? I, for one, could only start to make sense of it
by working slowly and painfully from the Sanskrit and the Tibetan texts
back to the translation.
lO
But surely most readers would like to' use
Jackson's translation in order to understand Dharmaldrti's verses, not the
other way round.
Part of the problem with Jackson's translation is the rather eccentric way
in which he translates Sanskrit and Tibetan technical terms. Conveniently,
Jackson provides us with an English-Sanskrit-Tibetan glossary (pp. 503-
516) that is helpful to a certain extent. Without it the translation is some-
times almost unintelligible, for Jackson does not follow the mainstream of
Buddhist pramlilJa studies in his choice of vocabulary. Next to amusing
but innocuous neologisms such as *angin for yan lag can-whole
10. ajftatarthaprakaso va, svariipadhigateJ; par-am II
prapta11J samanyavijftanam, avijftate I
yaj jftlinam ity abhiprayat
rna ses don gyi gsal byed kyan II
ran gi no bo rtogs 'og tu II
spyi yi mam par ses pa thob II
rangi mtshan ftid mi ses pa II
ses pa gan yin ies dgons phyir II
ran gi mtshan ftid spyad phyir ro II
Or [a means of knowledge is] illumination of an unapprehended object. '\
[Objection:] The cognition of the universal that is subsequent to the apprehen-
sion of the own form [of the particular] would be [a means of knowledge].
[Reply: No] because [in the characterization of a means of knowledge just men-
tioned we] intend [only] the cognition in respect to an unapprehended particular,
for the particular is examined [here] (or more literally: ... because the intention
[of the characterization of a means of knowledge just mentioned] is "[That cogni-
tion] which is a cognition in respect to an unapprehended particular [is a means
of knowledge]," for ... ).
FRANCO 119
(avayavin),ll there are problematic, sometimes clearly mistaken, choices.
For instance, sarika (for sarika) is not "concern," not even "perplexity," but
"doubf'; svajati / rigs'dra (or rather: rari gi rigs 'dra, cf. under "homo-
gene") is not "continuum," but "of one's own kind," and I assume that
Jackson has realized this also, because e. g., on p. 226 he translates svajati
with "homogene." Still, I don't see in which context svajiiti / rari gi rigs
'dra could be translated as "continuum." Lak$man does not mean "exam-
ple," but "characteristic." Sthana / gnas pa does not mean "existence"; the
semantic field of this word, as an action noun, ranges from "act of standing"
over "staying, abiding" to "continuing," etc. bTags pa occurs as a transla-
tion for baddha ("bound") and bandhana ("binding, bondage"; ''bond''), but
upacara, which is also translated with (fie bar) btags pa, certainly does not
mean "fetter," but "metonymy" ("imputation" in Jackson's terminology).
The adjective "impossible" is not an equivalent for the noun abhava
("inexistence, absence"). Svabhavahetu is not a "reason based on syn-
onymity," but a "reason based on an essential property / the own nature [of
a thing]," and hetujs not a "syllogism," but a "reason." Jagat cannot be
rendered as "transmigrator"; although etymologically derived from the root
-vga ("to go"), it simply means "the (animate) world." Jvara is not
"sickness," but "fever," and atyantaparok$a is not a "very hidden phe-
nomenon," but something "completely beyond the realm of the senses," etc.,
etc., etc.
Another problem in the rendering of technical terms, which is apparent
already in some of the above examples, is Jackson's recurrent difficulty
with grammatical categories such as noun, adjective, etc. Thus, viruddha is
not "contradiction," but "contradicted" or "contradictory," sarvajfia is not
"omniscience," but "omniscient," and kalpanagoha (sic) is not "non-con-
ceptuality," but "free from conceptual construction." An intriguing case is
abhimukhl, rendered as "manifest phenomenon"; as far as I know this word
does not exist in Sanskrit, and I suspect that it is the first member of the cvi
compound abhimukhlbhuta on the loose. The Tibetan equivalent given by
Jackson, mnon gyur, correctly presents the complete adjectival form
("become manifesf'). Moreover, the glossary abounds in mistakes regard-
ing the diacritical marks and spelling in general of even commonplace San-
skrit words (compare this with Jackson's introductory remarks, on p. 13,
11. Another neologism, which is symptomatic for the cavalier attitude towards
Sanskrit terminology in the so-called "Hopkins school," is *cittamatrin. I
thought that it had become well-known by now that this term does not exist in
Sanskrit, but was surprised to fmd out that Jackson still uses it.
120 nABS 20.1
that he has "maintained precise transliterations rather than phonetic spellings
in [his] rendering of both Sanskrit and Tibetan terms"); kalpaniigO{jha (cf.
above), iiraddhii (for sraddhii, "faith"), dik$a (dfk$ii, "empowermenf') and
srotavijiiiina (for srotravijiiiina, "aural cognition") are just a few examples.
These kinds of mistakes cannot but raise serious concern about Jackson's
command over the Sanskrit language in general and the technical terms of
Indian Buddhism, as do some of his reconstructions of Sanskrit terms in
the glossary (cf. e. g., angin referred to above, aprativiparyaya for "irre-
versible," and pratibhiisavi$aya for "apparent object"). It would have been
advisable for him, as a Tibetologist, to consult a Sanskritist for the prepara-
tion of such an extensive glossary, to make it a useful and reliable tool
especially for non-specialist readers.
On the other hand, there are some pleasant surprises. Indriya / dban po
is translated as "sense faculty." It is indeed a most common mistake to
translate indriya as "sense organ." However, indriya never refers to the
physical and visible organ, but to the capacity. Similarly, it is a most com-
mon mistake to translate cak$us as eye. In the vast majority of cases, how-
ever, it means "sight"; when Indian philosophers want to talk about the
organ, they use the word golaka ("eye-ball"). For instance, when Bud-
dhists argue with Naiyayikas whether cak$us goes out to meet its object or
not, they do not argue about whether the eye-ball goes out of its socket or
not, to touch a distant object such as the moon. The same holds good for
srotra: in the vast majority of cases it does not mean "ear," but "hearing,"
and when some Indian philosophers claim that srotra consists in iikiisa
they do not ignore the fact that the ears (kanJa) are made of flesh and
blood. Similarly, when they say that the senses are invisible, they do not
mean that no one can see eyes, ears, tongues, noses, and skins. Unfortu-
nately, Jackson does not draw the consequences from his accurate transla-
tion of iizdriya / dban po as "sense faculty"; he still refers to the particular
senses as "eye, ear, body, nose, tongue," but at least he made a step in the
right direction.
As mentioned above, the initial verses of the PramiifJasiddhi chapter are
not easy, and it would have been a miracle if Jackson could have under-
stood them in the way he approached them, that is, by merely reading rGyal
tshab's commentary with a contemporary Tibetan scholar, no matter how
learned, without taking the trouble of reading not only DharmakIrti's verses
themselves (which fortunately are preserved in the original language), but
also the commentaries of Devendrabuddhi, Prajfiakaragupta and Ravigupta,
which were thoroughly studied by rGyal tshab before he undertook the
composition of his own commentary and on which he relied heavily.
FRANCO 121
However, the Pramii1)asiddhi chapter contains also some (though not too
many) relatively simple and straightforward verses, and there at least one
could expect Jackson's translation to be more reliable. The inaccuracies and
misunderstandings that result from Jackson's approach are perhaps easier to
observe when one looks at those relatively simple verses.
The programmatic verse 34 forms a clear, simple and perhaps somewhat
trivial example. It comprises the entire discussion on compassion and
rebirth (vv. 34-131ab) in a nutshell: the prooffor the Buddha's authority or
reliability is compassion, and this compassion arises from repeated practice.
The materialist opponent objects that repeated practice for more than one
life is impossible, because the body is the support (iisraya) of cognition,
and consequently when the body is destroyed, the cognition is destroyed
too. Dharmaldrti rejects this objection by denying the very notion of sup-
port, or at least that the relationship between support and supported obtains
between the body and the cognition:
siidhana11J karu1)iibhyiisiit sii buddher dehasa11Jsrayiit I
asiddho ' bhyiisa iti cen II
Sa skya pru;u;lita's translation:
sgrub byed thugs rjes goms las de I
blo ni Ius la brten pdi phyir"
goms pas 12 grub pa med ce na "
ma yin brten ni bkag phyir ro "
Compassion is the proof [of the Buddha's being a means of knowledge].
That [compassion arises] from repeated practice.
[Objection:] Since the cognition rests on the body, the repeated practice
[of compassion for more than one life] is not established.
[Reply:] No, because [we] deny [that the cognition] rests [on the body].
Theoretically, there are two ways to construe the word karu1)ii; either as a
nominative (siidhana11J karu1)ii; abhyiisiit sii), or as a first member of a
compound (siidhana11J karu1)iibhyiisiit, sii ... ). The first alternative is
obviously better. Leaving aside for the moment problems of meaning, the
12. Read pa, as in Sa paI)'s kilrikil-translation preserved in the Tibetan
translation of Prajfiakaragupta's commentary. Also the kilrikil-translations in the
commentaries by Devendrabuddhi and Raviguptahave this reading.
122 JIABS 20.1
second alternative is not quite correctfrom the grammatical point of view.
As a rule, a demonstrative pronoun such as sa should not refer to a word
inside a compound (in this case karu1Jii-).13 However, Sanskrit is more
)
flexible than English in this respect and usage does not always follow this
rule. More specifically, there is one famous case where DharrnakIrti him-
self breaks it. 14 Therefore, it is only highly improbable, but not absolutely
impossible, that DharrnakIrti meant his verse to be read as in the first alter-
native. Indeed, Devendrabuddhi, Ravigupta and Manorathanandin explain
this verse unanimously according to the first alternative. 1 5 Only
Prajfiakaragupta presents both alternatives. Jackson, however, opts e ~ c l u
sively for the second alternative: "ACCUSTOMATION WITH COMPASSION
IS THE ESTABLISHER."
In all likelihood, this interpretation is not faithful to DharrnakIrti's inten-
tion, but could it be faithful to the Tibetan translation of this verse? If the
Sanskrit wording is slightly ambiguous, the Tibetan translation is not at all,
and it does not agree with Jackson's interpretation: sgrub byed thugs rjes
goms las de 1/ "The proof is by compassion. That [compassion] is due to
repeated practice." Sa pal). used the instrumental case-ending precisely in
order to prevent that thugs rje would be misconstrued with goms, for the
object of goms can be construed either with the particle la or directly; thus,
if the instrumental had not been used, a literal translation such as sgrub
byed thugs rje goms las could have been misunderstood in the sense of the
second alternative. In order to prevent this possible error Sa pal). even
departs from the literal translation of the Sanskrit: he translates "The proof
is by compassion," rather than "The proof is compassion." But why does
Jackson misunderstand Sa pal).'S intention and commits precisely the error
that Sa pal). wanted his readers to avoid? I see no other reason except that
this is simply due to Jackson's English translation of abhyasa / goms pa as
"accustomation." First, it should be noted that "accustomation" is a rather
unhappy rendering of abhyasa / goms pa because this important term
13. The same rule applies in English as well. For instance, if one says: "Today is
your birthday. This is an occasion for celebration," "this" must refer to ''birthday,"
not to "birth-."
14. Cf. Nyiiyabindu 1.1, in ParJ4ita Durveka Misra's Dharmottarapradzpa, ed.
D. Malvania, Tibetan Sanskrit Works Series 2, 2nd ed. (Patna: 1972).
15. Cf. PVP(pe) 18alf. = PVPCDe) 16b3f.: sgrub byed thugs rje . .. de (i. e., thugs
rje) ni sgrub pa byed pa'o ... goms las thugs rje ies bya ba de yin no II.
Similarly PW 20. 10: sadhanal1J karUlJo., and 20.12: abhyiisiit so.. PW(R) 316a4
(= 631.4) and 316b2 (= 632.2) also construes in this manner: sgrub byed thug
rje ies bya ba ... goms pa la (las?) ies bya ba ... de fiid goms pa las.
FRANCO 123
means an intensive and repeated practice, resulting in a habit (sometimes
also called abhyasa / goms), and not something one gets (passively) accus-
tomed to. But more importantly, and unfortunately, the English word
"accustomation" can be construed with an instrumental "with," and this led
Jackson, perhaps unconsciously, to construe goms pa also with an
instrumental.
One more minor point in this connection: Jackson's misunderstanding of
the syntax goes not only against Sa paIJ.'S translation, but also against rGyal
tshab's understanding. In a different context we find the following quota-
tion (gSal byed p. 302.1-2): sgrub byed thugs rje I ie thugs rje la sgrub
byed kyi ston par mdzad do. Jackson himself translates (p. 338):
"[DharmakIrti] shows compassion as the proof by the words, 'compassion
is the establisher.'" Why Jackson chose to interpret the syntax of the verse
against the explicit words of rGyal tshab, remains unclear.
All this may seem very trivial, but one error leads to the next. If one
reads sadhanal1J karulJ-abhyasat, what does one do with the remainder of
the verse? Sa alone is not a sentence. PrajiUikaragupta ingeniously con-
strues sa buddher dehasan;.srayat, "that (i. e., compassion) is due to resting
on the body of cognition." The statement that compassion rests on "the
body" of cognition is not entirely clear. One probably has to connect this to
the discussion about the possibility of an unlimited increase of compassion
in vv. 120-131ab. ''The body of cognition" could then be interpreted as the
own nature of the cognition, i. e., compassion can increase to the highest
degree because it becomes the own nature of the Buddha's stream of con-
sciousness. This presupposes a rather rare, but not unknown usage of
words meaning "body"; cf. Apte, s. v. "tanu" (3): "Nature, the form or
character of anything." Cf. also P W, s. v. "kaya" (7): "Natur,"
''Eigenthi.imlichkeit.''
However, this unconventional usage of deha is apparently not recorded
for the Tibetan equivalent Ius. In any case, Jackson does not follow
Prajiiakaragupta's second alternative all the way, but suggests something
completely different (p. 223):
[THE LOKA.YATAS16] SAY: "[COMPASSION] IS NOT ACCOMPLISHED
THROUGH ACCUSTOMATION,
BECAUSE THE MIND IS BASED ON THE BODY.
16. Incidentally, Lokayata is usually a name for the materialist school; its follow-
ers are called Lokayatikas or Laukayatikas.
i24 JIABS 20.1
The Sanskrit text :;it the source of this translation, however, reads as fol-
lows: sii buddher dehasan;srayiid asiddho 'bhyiisa iti cet. We obviously
have a little problem of gender here. The feminine pronoun sii is c;onstrued
by Jackson with the masculine adjective asiddha/:l. Moreover, the n o n r i n ~
tive abhyiisal! is rendered "through accustomation," probably because of the
inferior variant goms pas for goms pa.
Yet we have to remember that Jackson's translation is not simply of the
Sanskrit verses, but of their Tibetan translation, and not even of the Tibetan
verses as they stand, but "geared towards rGyal tshab rje's interpretation."
Since Tibetan does not make a distinction of grammatical gender, one may
think that this is how rGyal tshab, and not Jackson, misunderstood the
verse in question. Moreover, since Jackson does not indicate here that
rGyal tshab departs from "the straightforward meaning," we should assume
that already Sa skya palJ.Qita erred in this point.
If Jackson's rendering of Dharmaldrti's verses were really representative
of Sa PalJ. and rGyal tshab's understanding, this would be a serious blow to
our appreciation of traditional Tibetan scholarship. rGyal tshab was one of
the two most important disciples of Tson kha pa. His gSal byed is consid-
ered to be one of the most important, perhaps the most important,
Pramii1).aviirttika commentaries in the dGe lugs pa tradition. If rGyal
tshab's (as well as Sa PalJ.'s) understanding of DharmakIrti's verses were as
poor as it would seem from Jackson's interpretation, our respect and admi-
ration for the great tradition of Pramii1).aviirttika scholarship in Tibet would
be indeed unjustified. Jackson for his part would suggest that rGyal tshab's
deviations from DharmakIrti's intention are to be accounted for by the time-
gap, the development of philosophical systems and altitude: "any commen-
tary, especially one separated from its root-text by seven centuries, several
philosophical systems and the highest mountains in the world, necessarily
distorts the original, but this distortion is a price that must be paid for com-
prehensibility" (p. 159). Does this imply that DharmakIrti's verses as such
are so incomprehensible that any attempt to make them comprehensible
necessarily means distorting them? It is hard to believe, but this seems to be
what Jackson is arguing for in general, for he says in his preface (p. 11)
that root-texts "are concise to the point of obscurity, and require commen-
tary to be truly comprehensible."17 Thus, the following reasoning is
implied: all root-texts require commentary to be comprehensible. All com-
17. Cf. also further below on the same page: "It is a moot question whether in
every case rGyal tshab rje reflects Dharmaldrti's own viewpoint (it is highly
doubtful that he does) .... "
FRANCO 125
mentaries distort their original root-texts. Therefore, root-texts have to be
distorted to be comprehensible. But if they are comprehensible only in a
distorted form, i. e., incomprehensible in their real form, how can we claim
that their real form is distorted in the first place? And how would Jackson
know about rGya1 tshab's distortion of DharmakIrti without taking the
trouble to read and try to understand the root-text in the original Sanskrit?
Should we really assume that rGyal tshab distorts DharmakIrti's verses to
the extent claimed (and presented) by Jackson? I fail to see clear evidence
for such a distortion. It seems to me that Jackson underestimates rGyal
tshab's competence and disregards his hermeneutical situation. Unlike
Jackson, he did not read Dharmaldrti's verses in a vacuum. He knew the
commentaries of Devendrabuddhi, Prajfilli<:aragupta and Ravigupta, albeit in
their Tibetan translation, and relied on them heavily-much more heavily
than is apparent from Jackson's notes. For these precursors of rGyal tshab
the basic text was obviously not incomprehensible; and on the whole their
commentaries provide coherent and precise explanations without giving the
impression of forced and artificial reading. Also, rGyal tshab's purpose
was not to compose an original treatise but to explain DharmakIrti's verses
as clearly and accurately as possible. And the high regard for his commen-
tary in the dGe lugs pa tradition is due precisely to this reason, and not
because he interpreted the verses in an innovative or distorting manner. In
this respect, Jackson's rendering of the verses in a radically different man-
ner-sometimes in such a distorted way that one can no longer recognize
them-goes against the self-understanding of the Tibetan exegetical tradi-
tion as well as against the historical context in which the commentary was
composed. Besides, having read rGyal tshab's comments as carefully as I
can, I claim that it is very often impossible to reconstruct the exact manner
in which he understood the wording of the root-text. Unlike the Indian
commentators rGyal tshab does not use the pratfka method of glossing; his
comments are more general and discursive, not so closely tied to the mula-
text as the comments of some of his Indian precursors. He writes about
and around the mula-text, but does not explain it word for word.
The shortcomings of Jackson's method do not only bear on his
"translation" of DharmakIrti's verses, but also on his understanding of
rGyal tshab's commentary. It should be obvious that in order to understand
this text one has to be familiar with rGyal tshab's sources. Without that
background, Jackson is often groping in the dark, especially when matters
become slightly problematic. For instance, the Lokayata position in the
above quoted programmatic verse (v. 34) is represented byrGyal tshab as
follows (Jackson's translation p. 223): "This follows, BECAUSE, since
126 JIABS 20.1
MIND IS BASEDON THE BODY, when the body is destroyed, [mind] is
also destroyed. For example, [mind] is a result of the body, as light [is the
result] of a lamp; or it is a quality of the body, as capacity to intoxicate [ ~ s a
qUality of] liquor; or it is naturally based [on the body, and the relation: is]
like that between a wall and a drawing that depends on it."
Let me start by emphasizing that Jackson's translation is literal and per-
fectly correct. However, a literal translation is not enough to understand the
examples and their implications, and indeed Jackson felt the need to explain
this sentence. However, his explanation is mOre likely to confuse and mis-
lead, than to enlighten the reader. It is likely to mislead the reader by sug-
gesting that rGyal tshab himself designed certain reasons, e. g., bemg a
result of the body,iS and provided their examples independently on the
basis of his acquaintance with Indian Lokayata sources, as Jackson claims
that "[t]he probative reasons and examples are supplied by rGyal tshab rje.
The examples, in particular, are stock Lokayata similes" (n. 6). He then
refers the reader to the section on Lokayata in the Sarvadarsanasaitgraha,
which is a convenient introduction to Indian materialism, but of little help
and no direct relevance for the present context. Finally, he coaches us in
modem philosophy and the "'category mistake' discussion initiated by
Ryle."
To understand rGyal tshab's examples one does not have to consult the
Sarvadarsanasa7igraha, a late doxographical work, where merely the
power to intoxicate is mentioned as an example, in a completely different
context than the one with which rGyal tshab is concerned here; nor is it
necessary to be familiar with the ideas of Gilbert Ryle. It would have been
more helpful to read the classical commentaries on this verse. Commenting
on the materialist statement (in v. 34 above), Devendrabuddhi, who is
followeqby Prajfilikaragupta, Ravigupta and Manorathanandin,i9 explains
the relationship of support and supported as a general term that can be
further analysed into three different relations:
i
ii 1) a relation between substance and quality,
I
ii 2) a relation between cause and effect,
,I 3) a relation between capacity (Saktz) and possessor of capacity.
I
'Ii
Ii
ii! IS. These reasons, however, do not appear as reasons in Jackson's translation,
'I' but rather as statements to be proved.
'Iii 19. Cf. PVP(Pe) ISa7 = PVP(De) 16bl, PYA 53.21, PW(R) 317al (= 633.1), PW
)i 20.1S-20.
:11,
:1'1'1
"i
FRANCO 127
The first tWo relations presuppose that the relata are ontologically different;
the third, that they are not: 20
If [the body and the cognition] are different, [the cognition] is supported by the
body inasmuch as the cognition is a quality (*gulJa) of the body, like the white
[color] of a cloth and the sweet [taste] of sugar. Or [the cognition] is sup-
ported by the body, because itis an effect (* phala) of the body, like the light
(*prabha) [is supported] by a lamp (*pradfpa). If, on the other hand, [the
cognition and the body] are not different, [i. e., the cognition] is a capacity
(* sakti), that has the nature of the body ...
Manorathanandin conveniently construes three inferences for the three
interpretations of "support": 21
1) The cognition is supported by the body,
because it is its effect,
like light is supported by a lamp.
2) The cognition is supported by the body,
because it is its capacity,
like the capacity of intoxication is
supported by the intoxicating substance.
3) The cognition is supported by the body,
because it is its quality,
like whiteness is supported by a cloth.
The upshot of these inferences is, of course, that in all three possible modes
of the relationship, when the support is destroyed the supported is
20. Cf. PVP(pe) 18a7f. = PVPCDe) l6blf.: don gian fiid yin na blo ni Ius kyi yon
tan yin pa de Itar na Ius brten pa yin te I dper na ras kyi dkar po dan bu ram gyi
mnar ba Ita bu'o II yan na Ius kyi 'bras bu ftid yin pa'i phyir Ius la brten pa yin te
I dper na sgron ma la 'ad Ita bu'o I don gian ma yin na yanlus kyi bdag ftid du
gyur pa'i nus pa yin pa ....
The translations from Devendrabuddhi and Prajiiakaragupta's commen-
taries are taken from chapter 4 in my forthcoming book Dharmakfrti on Com-
passion and Rebirth. (Published after the completion of this review; Vienna:
1997.)
21. Cf. PW20.18-2l: buddhir deham asrita karyatvat, pradfpam iva prabha,
saktirupatvad va, madyam iva madasaktil;, gUlJatvad va, pa,tam iva suklata.
tredhapy asrayavinase tasya nasa! kuto janmantaralJi, katha11J. va t e ~ v abhyiisal;
krpadel;? iti carvakal;.
128 JIABS 20.1
destroyed with Therefore, when the body is destroyed the series of cog-
nitions is interrupted. PrajiHikaragupta quotes (or paraphrases) some
Oirvaka examples to illustrate the point:
22
Thus, a mural does not last without a wall, nor step over to another wall, nor
has it come from another wall. Or [it is] similar to [the case of] the coll?ur
which arises from the ripening of the mango fruit etc., [which does not last
without the fruit, does not go to another fruit, nor has come from another
fruit]. Or [it is similar to the case of] smoke, which is the product [of a certain
fire], does not come from another fIre, nor sets out to another fIre. As for the
power of intoxication (madasakti), it rests on the intoxicating [substailce],
[and] it appears as something new because liquids like etc., are mixed.
When the [power of intoxication] is disappearing [in a certain intoxicating
substance], it does not take another intoxicating [substance] as its support. In
the same manner, specific senses, consciousness [etc., rest on a body and
when destroyed in that body, do not go to another body, etc.].
rGya! tshab knows the above inferences, though curiously enough in a
somewhat distorted form: the second and third inference are incoherently
mixed up, and the capacity is said to be a quality: "[o]r it [i. e., mind] is a
quality of the body, as capacity to intoxicate [is a quality of] liquor .... "
Was rGyal tshab the first exegete to have contaminated the two possible
relations and their examples? Or was he following Tson kba pa's under-
standing? Or, since the presentation of the issue in the earlier commentaries
is straightforward and could hardly be misunderstood, shouldn't we simply
assume that the gSal byed, as available to us nowadays, contains a short
22. Cf. PVA 53.29-54.2: tato na citraTIJ kurjyavirahitam kurjyiin-
taraTIJ vii saTikriimaty iigata11J vii kurjyiintariit. iimraphaliidipiikarilpavad vii.
kiiryaTIJ vii dhilmo na dhilmadhvajiintariid iigacchati. niipi dhilmadhvajiintaraTIJ
prayiiti. madaSaktis tu madyiisritii apilrvii priidur-
bhavati. vilfyami'inii na madyiintaram avalambate.
Note, however, that Prajfilikaragupta formulates these examples in the
context of a theory according to which the body of the parents is the support of
the cognition of the newborn. This theory must have been formulated some time
after Dharmakirti, and it is also unknown to Devendrabuddhi; cf. Tattvasan-
graha of with the commentary Panjiki'i. of Kamalasila, ed. D.
Shastri, Bauddha Bharati Series 2 (VariiI)asI: 1968) v. 1892, and E. Steinkellner,
Dhannottaras Paralokasiddhi (Wien: 1986) 11, where, however, the cognition
(not the body) of the parents is taken as the cause of the newborn's
consciousness.
FRANCO 129
lacuna here? 23 Unfortunately, only one edition of the text is available to
me, and I could not compare it with other editions.
Whatever the case may be, the question now arises: what is the use of
studying rGyal tshab's commentary? It seems that as a historical source for,
say, a better understanding of Dharmaldrti's opponents, the gSal byed
would be of little help. (This statement is not occasioned merely by the
above observation of an obvious confusion, but is based on the reading of
the entire commentary). If one is interested only in the literal interpretation,
neither the gSal byed nor any other Tibetan commentary could compete in
simplicity and clarity with Manorathanandin's PramalJflvarttikavrtti merely
because of the difference in the linguistic medium. As for philosophical
acumen, brilliance in argument, originality of thought, etc., the gSal byed is
certainly dwarfed by the towering achievements of PrajfUikaragupta. How-
ever, rGyal tshab's commentary does not only consist in a literal explana-
tion of the verses; it contains other features that prove useful, especially in
those areas of exegesis that are neglected in the Indian tradition. There is
one characteristic feature of Tibetan commentaries that is either completely
lacking, or present only in a rudimentary manner, in the Indian commenta-
torial tradition, namely, the structural analysis (sa bead). Here, as far as I
could see,24 Jackson has done an excellent job. Although the overall
structure of the Pramalfasiddhi chapter as a loose commentary on the five
epithets of the Buddha is clear, the inner structure within each section is
sometime most obscure, and the way in which the individual verses are
related to each other is often puzzling. Jackson not only translates rGyal
tshab's analysis, but also conveniently groups the items together in an
appendix (pp. 489-502), providing at the same time a most useful concor-
dance between the verses of the PV, the pages in the gSal byed and the
pages in his own translation. 25 Anyone who is interested in the structure of
the Pramalfasiddhi chapter will read these pages with great interest. Sev-
eral issues that have been subjects of debate among modem scholars are
reflected in rGyal tshab's analysis. For instance, the dividing line between
the anuloma and pratiloma sections (i. e., the presentation of the Buddha's
23. A simple aberratio oculi could explain the confusion.
24. I could not systematically check the entire commentary, but only a sample of
cases.
25. Some minor mistakes are inevitable in such a complex structure. E. g., p.
156, #211.211.222.231: "Disposing of the objection that [compassion, etc.]
increase only from [accustomation of] a homogeneous substantial cause"
(referring to vv. 127-128). This is precisely what DharmakIrti himself claims for
compassion. One has to read ''jumping'' instead of "compassion."
i30 nABS 20.1
characteristics in .e order of their arising and in the opposite sequence) is
placed after v. 146c.
26
It is also interesting to note that the section beginning with v. &3 (on the
parts and the whole, and on the atoms) is analysed as a second refutation of
materialism, even though the terminology used by DharmakIrti is typical of
rGyal tshab certalnIy senses that there is a problem here,
and says: "Even though this whole is a fundamental doctrine of the
V the opponent of these statements is the Lokayatika because [this
whole] is part of the refutation by way of analysis of the nature of the body
as [something] which functions as support, when the Lokayatika claims the
body to be a special support of the mental cognition."27 Of course, such
comments should not be accepted blindly, and our knowledge today of
Nyaya and Mimlin).sa texts is far superior to what rGyal tshab could have
known from the texts at his disposition, but the comment as such is
thought-provoking, and points at a genuine problem in DharmakJrti's text.
28
Another interesting characteristic of rGyal tshab's commentary are the
occasional short digressions into topics that are only implicit in
DharmakJrti's work and were more fully developed in the Tibetan tradition,
for instance the relationship between the Buddha's authority and the means
of knowledge, namely, perception and inference. Can one ascertain the
teachings propagated by the Buddha simply by relying on one's own facuI-
ties? And if this is the case, what is the point of proving the Buddha's
26. Cf. M. Inami and T. F. Tillemans, "Another Look at the Framework of the
PramiiI,lasiddhi Chapter of Pramat;laviirttika," Wiener Zeitschrift fUr die Kunde
Sadasiens 30 (1986) 123-142.
27. Cf. gSal byed 272.8-11: yan lag can 'di bye brag pdi rtsa bdi 'dod pa yin
kyan I rgyan, phan Ius yid blo'i rten khyad par can du 'dod pa na I rten byed pdi
Ius kyi no bo la dpyad nas' gog pa'i yan lag yin pas I giu1i ' di dag gi phyir rgol ni
rgyan phan yin no 8.
Jackson, however, translates as follows (p. 272): ''This 'whole' is a fundamental
assertion of the V but when the Lokayatas assert that the body is the .
special basis of mind, they are partly refuted by an analysis of the essential
[aspect of] body that acts as a basis, [so] the Lokayatas are the opponents in
this verse." The translation is typical and symptomatic: every word in
Tibetan has an equivalent in English, but the syntactical relationships of the
original are not respected in the translation. For instance, 'gog pdi yan lag
yin pas I is translated above by "partly refuted."
28. Cf. E. Franco, or Carvaka? The mysterious opponent in
PramiiI,laviirttika 2.63-72," Asiatische Studien / Etudes Asiatiques 48.2 (1994):
683-698:
FRANCO 131
authority? Or must one accept at least a certain pa..'1: of them on faith? rGyal
tshab says (p. 173) that without relying on the Buddhist scriptures one
would not even think about such topics as selflessness, momentariness,
e t c ~ , how much less infer them. But once taught by the Buddha, these top-
ics can be checked and confIrmed by perception and inference.
Finally, it is also instructive to observe that certain issues that have been
much debated by modern scholars do not seem to have been raised at all by
rGyal tshab. I looked in vain, for instance, for a statement concerning the
cirCUlarity, or alleged circularity, in DharmakIrti's writing concerning the
authority of the Buddha and the validity of the means of knowledge (i. e.,
perception and inference establish the authority of the Buddha, and the
Buddha establishes the validity of perception and inference
29
). According
to the introductory section (pp. 172-173) the authority of the Buddha has
to be proved by perception and inference, more precisely, by the authority
of his teachings which are proved by perception and inference.
To conclude, Jackson's work does not further our understanding of
DharmakIrti's verses; this was not Jackson's central concern anyway. But it
also does not contribute much to our understanding of rGyal tshab's inter-
pretation of Dharmaldrti's text, as long as this text remains in an indistinct
haze. The book illustrates once more, almost dramatically, that a genuine
understanding of the older indigenous Tibetan commentaries on Indian
Buddhist texts, or of independent works mainly based on these texts, is not
possible without a thorough, fIrst-hand understanding of these "root-texts"
and their Indian exegesis, of course in their original language if they are
preserved in it. Any evaluation of a commentary such as the gSal byed, as
to its faithfulness to the original, its expansions, additions and creative
modifications, cannot proceed without a clear understanding of the very
basis from which this evaluation has to start, namely, the "root-text." With-
out such an understanding it is neither possible to state that the gSal byed is
"a useful gloss on Dharmaldrti" nor that it is highly doubtful that rGyal
tshab reflects Dharmaldrti's own viewpoint (both statements on p. 11).
Nevertheless, Is Enlightenment Possible? is a very much needed intro-
duction to rGyal tshab's thought in general, and demonstrates the subtlety
and ingeniousness of the Tibetan scholastic tradition to a wider audience
than the narrow circle of specialists. It will certainly contribute to the
growing awareness outside this circle that Tibetan culture does not exhaust
itself in the various branches of its religious mysticism and spirituality that
29. Cf. T. F. Tillemans, Persons of Authority, Tibetan and Indo-Tibetan Studies
5 (Stuttgart: 1993) 18-24.
132 JIABS 20.1
have become known in the West, but,that it also offers a sophisticated
philosophical tradition. Jackson's painstaking tremendous effort of translat-
ing and annotating the gSal byed on the Prama1)asiddhi chapter ~ o n v i n c
ingly shows that the tradition of Tibetan Buddhism applied a rigorous
philosophical analysis to its central religious values, such as rebirth, self-
1essness and liberation, and was moreover deeply concerned with basic
epistemological and metaphysical questions, such as the mind-body prob-'
lem. This tradition is still alive today in those of its exponents with whom
Jackson had the good fortune to work, and we have to be grateful to him
for sharing their knowledge with us and the philosophically interested gen-
eral public.
ABBREVIATIONS
gSal byed: rGyal tshab dar rna rin chen, rNam 'grel thar lam gsal byed,
Vol. I (Sarnath: 1974).
PV: "Prama1)avarttika of DharmakIrti." Ed. Y. Miyasaka. Actalndologica
2 (Naritasan Shinshoji: 1971/72).
PV Tib: Tibetan translation of Prama1)aVarttika in pv.
PYA: Prama1)avarttikalarikara of Prajfiiikaragupta. Ed. R. Saitlqityayana
(patna: 1953).
PVP: Prama1)avarttikapafijika of Devendrabuddhi. Pe = Peking no. 5717;
De = Derge no. 4217.
PW: Prama1)avarttikavrtti of Manorathanandin. Ed. D. Shastri. Bauddha
Bharati Series 3 (VariiI).asi: 1968).
PW(R): Prama1)avarttikavrtti of Ravigupta. Derge no. 4224, references in
brackets refer to the Taipei reprint.
ROGER R. JACKSON
"The Whole Secret Lies in Arbitrariness":
A Reply to Eli Franco
I am pleased that Eli Franco has taken the time and effort to review my
book, Is Enlightenment Possible? Dharmakirti and rGyal tshab rje on
Knowledge, Rebirth, No-Self and Liberation (Ithaca: Snow Lion, 1993;
hereafter referred to as IEP).l Needless to say, fm sorry that he doesn't like
it better. I am not, however, entirely surprised: I knew from the outset that
of the several constituencies to which the book might be of interest, it was
the community of Dharrnaldrti scholars who were most likely to find fault
with it, on the grounds that it does not deal as directly as they might like
with DharmakIrti and his texts. What does surprise me is that Franco's cri-
tique is based on an analysis of so small and unrepresentative a sample of
the book's actual contents. I am reminded of an observation by
Kierkegaard:
The whole secret lies in arbitrariness .... You see the middle of a play, read
the third part of a book. In this way one derives a quite different enjoyment
from the one the author has been so kind as to intend for you. One enjoys
something entirely accidental .... 2
Substitute "misery" for "enjoyment" and "suffers" for "enjoys," and my
overall sense of Franco's treatment of Is Enlightenment Possible? will be
clear. This is not, of course, to say that the book is above criticism, or that
Franco has not occasionally found his mark; and I do believe that he has
raised some methodological questions that are worthy of serious discus-
sion. Nevertheless, I am quite troubled by how much that is basic to the
book he omits even to discuss, by the Indological and philological funda-
1. Eli Franco, "Distortion as a Price for Comprehensibility? The rOyal tshab-
Jackson Interpretation of DharmakIrti," The Journal of the International Asso-
ciation of Buddhist Studies 20.1 (1997) 109-132, hereafter referred to as DPC.
2. Spren Kierkegaard, Either / Or: A Fragment of Life, trans. Alastair Hannay
(New York: Penguin, 1992) 239.
133
134 JIABS 20.1
mentalism
3
that shapes his analysis of the parts that he does discuss, and by
his obvious ambivalence about rGyal tshab rje, the Tibetan commentarial
tradition, and the methods by which they ought to be studied. In the space
the editor has so kindly offered me, I will first address the scope of
Franco's treatment of Is Enlightenment Possible?, then respond to some of
his specific criticisms of those portions that he has selected for discussion,.
and finally consider some of his arguments about the proper method for
studying Tibetan commentaries on Indian Buddhist texts.
I am hardly the fITst author to feel that a reviewer has ignored what is most
important about his or her book, while focusing on passages and issues iliat
are of secondary consequence. Lest I utter too extended a Prufrockish
whine-''That is not it at all! That is not what I meant, at all" 4-I will point
out as briefly as possible that Franco's lengthy discussion of Is
Enlightenment Possible? addresses in detail only the following elements of
a 571-page book: one four-page chapter on "Scholarship on Dharmaklrti";
selected items in the appendix, glossary and bibliography; and two sam-
plings from my translation, which cover rGyal tshab rje's commentary on
roughly five of the 285 verses of the Pramiir:tasiddhi chapter of the
Pramiir:taviirttika.
Of course, no reviewer can discuss every element of a work he or she is
considering. Given the necessity for selectivity, however, it behooves the
reviewer to give at least a general indication of the actual range of the
book's contents, and to choose for analysis "core samples" that are repre-
sentative of the work as a whole. Franco has done neither. He never even
mentions that roughly the first third of Is Enlightenment Possible? consists
of an extended discussion of "Truth and Argument in Buddhism," which
attempts tovsituate both DharmakIrti and rGyal tshab rje within the history
of Buddhist thought, and to reflect, both intra-traditionally and compara-
tively, on the philosophical issues with which they grappled; and he barely
acknowledges that the translation is provided with an extensive apparatus
of explanatory footnotes, whose major purpose is to try, for the modem
reader, to make some philosophical sense ofrGyal tshab rje's commentary.
3. The phrase is a variation on "philological positivism," a category suggested
recently by Jose Cabezon ("Buddhist Studies as a Discipline and the Role of
Theory," The Journal of the International Association of Buddhist Studies, 18:2
[1995] 245).
4. T. S. Eliot, ' ' T h ~ Love Song of J. Alfred Prufrock," Selected Poems (New
York and London: Harcourt Brace Jovanovich, 1964) 15.
JACKSON 135
These sorts- of discussions may not interest France much, but he might at
least have mentioned how much, for better or for worse, they obviously
interested me. Furthermore, of the translation passages Franco selects for
discussion, all but one are from the opening verses of the chapter, which he
himself admits are "notorious for their difficulty" (DPC 113), and which,
moreover-interesting and important as they are-are neither typical of the
chapter in their content, nor central to DharmakIrti's major line of argument
(though they do, of course, help to set it up). Finally, in analyzing my
translation, Franco consistently launches his critique from an Indological
rather than a Tibetological perspective, lamenting my "lack of interest in
Dharmaklrti" (DPC 109), and insisting that I should be reading rGyal tshab
rje via DharmakIrti, when my explicit purpose is precisely the opposite,
namely, to cast some light backward on Dharmaldrti from a great and still
influential representative of the Tibetan commentarial tradition. Is Enlight-
enment Possible? is Tibetological and philosophical in its orientation; it is
obViously not the Indological and narrowly philological book that Franco
wishes I had written, but Tm not certain that ifs entirely fair-common as it
may be-to be criticized for the book that one did not write.
5
Although I wrote Is Enlightenement Possible? with issues of philosophical
import foremost in my mind (and am quite explicit about this in a number
of places), it would be quite disingenuous, in a work devoted in consider-
able part to the translation and elucidation of a Tibetan text, to claim that
philological issues-let alone correct translation--do not matter at all. They
do, so let me turn p.ow to some of Franco's major criticisms in this area.
Roughly, he identifies two types of failing: technical errors and miscon-
struals of Dharmakirti and / or rGyal tshab rje. I will comment on each of
these in turn.
Franco devotes two long paragraphs (DPC 111-113) to cataloguing a
variety of "shocking" technical errors he has located in my chapter on
Dharmakirti scholarship, as well as the appendix, glossary and bibliogra-
phy. He himself admits that some of the statements he considers mistaken
(e. g., on the value of contributions to Pramil.r;1a studies by Jha and Warder)
are matters of opinion. Nevertheless, I cannot deny that Franco has located
some mistakes and inconsistencies in the sections he has scrutinized. I am
5. For another recent example of this review style-which, in fact, raises many
of the same issues and entails many of the same problems as does DPC-see 1.
W. de Jong's review of Jose Cabezon's A Dose of Emptiness, Indo-Iranian
10umal38 (1995) 285-288.
136 JIABS 20.1
embarrassed by these, and only can note that the chapter on Dharmakirti
scholarship was a quite belated addition to the book, and was not checked
for accuracy with sufficient rigor; as for the appendix, glossary and,bibliog-
raphy, I did my best to weed out errors, but, obviously, a few remain.
Later (DPC 118-120), Franco devotes two further paragraphs to "the
rather eccentric way in which Jackson translates Sanskrit and Tibetan tech-:
nical terms" (DPC 118). Eccentricity apparently is measured against
something Franco describes as "the mainstream of Buddhist prama1}-a
studies" (DPC 118), into which he feels I have not entered with sufficient
enthusiasm. I do not wish to descend with Franco into terminological
trench warfare. Let me simply note that most of his criticisms of my terms
are based on their appearance in the glossary, rather than their instantiations
in the text, where, in their proper context, they often may seem considerably
less egregious. Furthermore, his criticisms are launched primarily from the
perspective of the Sanskrit side of the terminological puzzle, whereas I have
concentrated on the Tibetan terms, which, given Tibetan's grammatical and
terminological differences from Sanskrit may not-the best efforts of
translators and commentators notwithstanding-always match up exactly
with their Sanskrit prototypes. Even if we were to concede (and this is
dubious) that where the Sanskrit is unambiguous there can be little doubt
about a Tibetan term, I find-as in most philological discussions-that
sometimes the corrections offered are helpful (e. g., "reason based on an
essential property" rather than "reason based on synonymity" for rang
bzhin gyi rtags / svabhiivahetu 6), and sometimes trivial (e. g., "doubt"
rather than "concern" for dogs pa / sanka, "completely beyond the realm of
the senses" rather than "very hidden phenomenon" for shin tu lkog gyur /
p a r o k ~ a ) ..
Now, Fr1co clearly does not believe that the technical points he raises in
such detail are "trivial." Indeed, he sees them as calling into question the
very value of the book-for, he reasons, if the philology is weak, then the
real foundation of the work is flawed. Now I am far from a translational
(or any other kind ot) relativist. I certainly believe that there are better and .
worse translation choices to be made in any given circumstance. Nor do I
6. My choice here was based on a concern to avoid "essence-talk" in translating
Buddhist terms, but I probably should have simply stuck to the more literal ren-
dition, which Franco's clearly represents. It might be noted that, the first time the
term comes up (IEP 104), I do provide "own-nature" as a parenthetical alterna-
tive to "synonymy" (not, as franco has it, "synonymity") though I might have
explained my choice of a translation-equivalent that, in this case, I would agree is
"eccentric."
JACKSON 137
object to the correction of technical errors where they exist, or to the prof-
fering of alternative translation-tenns; such criticism is an important way in
which, individually and as a field, we progress. What I do find disturbing
is the spirit of philological fundamentalism that Franco brings to his cri-
tique' the apparent certainty that there is usually a single right translation for
any given term in any given context and that a translator's failure to provide
that single right term is evidence of his or her incompetence, hence of the
worthlessness of his or her entire project. As I have tried to argue gener-
ally above, and will attempt to demonstrate more specifically below, it is not
so evident that the rights and wrongs of translation-choices are utterly clear-
cut, for the choices that are made do depend in large part (but not com-
pletely) on the translator's own background and sense of the context of the
text under consideration-and Franco and I obviously have very different
understandings of the context of rOyal tshab rje's PramiilJaviirttika com-
mentary. Nor is it so evident that the presence of a few errors in a huge,
translated work-and I would challenge Franco to name a translation that
cannot be corrected-vitiates the worth of the work as a whole. This is
especially so when the translator has-as I believe I have-got it right the
vast majority of the time and, besides, is interested in, and deals extensively
with, issues that lie beyond, and are only partially dependent upon, the nar-
rowly philological. I do not claim that a philosophical account of rOyal
tshab rje's or DharmakIrti's arguments can be given in disregard of an accu-
rate reading of their texts; I do want to maintain that some at least partially
arbitrary notion of correctness cannot be invoked as the sole standard by
which to measure a scholar's efforts.
With these points in mind, let me turn briefly to a consideration of those
passages in my translation that Franco has selected to demonstrate my sup-
posed misconstrual of rOyal tshab rje or Dharmaklrti. I should remark
generally that I find much of what Franco had to say here quite interesting
and valuable. I cannot gainsay his understanding of Dharmaklrti's verses,
both in the Sanskrit original and in Sa skya Pal).Qita's Tibetan translation;
his command of Indian commentarial material relating to the PramiilJa-
viirttika.; or his broader Indological competence. I'm not, on the other
hand, convinced that his critiques of my reading of rOyal tshab rje's inter-
pretation of Dharmakirti-which was, let us recall, my basic angle of
approach in Is Enlightenment Possible?-are at all compelling.
The passages he analyzes fall into two major groups: (1) those relating to
the first six verses of the PramiilJasiddhi chapter of the PramiilJaviirttika.
i38 JIABS 20.1
[PV],7 in particular verses 1-2 (commented upon on pp. 229-230 of rGyal
tshab rje's commentary [GT],8 translated on pp. 176-178 of IEP and dis-
cussed on pp. 113-118 of DPC); 4d-5a (= GT 231-233, IEP 1 8 0 - ~ 8 4 , DPC
115-116); and (quite briefly) 5d-6 (= GT 233-236, IEP 184-188, DPC 117-
118); and (2) those relating to verses 34a-c (= GT 252-253, IEP 221-223,
DPC 121-128).
(1) The first six verses of the Pramiir;asiddhi chapter are notoriously diffi-
cult, and though I think that most of the time I have made sense of rGyal
tshab rje's interpretation of them, I am aware that my translation through
this section is often quite convoluted, and occasionally flawed. I do believe
that, in relation to PV lab, Franco is right to question my footnoted com-
ments (IEP 177, n. 2) on the relation between apperception (rang rig /
svasarr;vedanii) and subsequent confirmatory action. He is correct in
maintaining that Dharmaldrti and rGyal tshab rje are not distinguishing
between direct cognitions and cognitions that must receive subsequent con-
fmnation, but, rather, between apperceptions, none of which require subse-
quent confirmation, and cognitions of knowledge-objects, all of which
require subsequent confirmation. By the same token, his related criticism
of my footnoted remarks on Devendrabuddhi's alternative reading of PV
4d-5a (DPC 115-116, on IEP 182, n. 15), where similar issues are at stake,
seems also to be on target. I do not believe that my actual translation of the
rGyal tshab rje passages in question is as far off the mark as Franco claims,
but I would certainly now alter them somewhat, given the chance to do so.
Two brief examples will have to suffice. (1) I would alter my translation
of btso bsreg la sags pa'i don byed nus par rang gis ji ltar gzhal ba ltar
gnas pa ni / las la mi slu ba yin (GT 229, IEP 182-183) to something
along the lines of: "[a knowledge object's] ABIDING as comprehended as it is
Ui ltar] by oneself AS [CAUSALLY] EFFICIENT, such as for cooking,
burning, etc., is non-deceptive in [its susceptibility to subsequent confmna-
tory] action." The problem here is with the ji ltar, which I no longer would
read as interrogative; on the other hand, Franco, is wrong in claiming (DPC .
114) that "nus par r) is not translated" -it is incorporated into "EFFICIENT."
(2) I would alter my translation of tshad ma yin pa'i cha de dus phyis
7. Y. Miyasaka, ed., PramiilJaviirttika-kiirikii. Sanskrit and Tibetan. Acta Indo-
logica 2-4: 1977.
8. rGyal tshab rje, rNam 'grel thar lam gsal byed. 2 vols. Samath: Tibetan
Monastery, 1974. rGyal tshab rje's commentary on the PramiilJasiddhi chapter is
found in the first volume, and all references are to that volume.
JACKSON 139
byung gi don byed can gyi tha snyad pa'i tshad ma nyid las rtogs dgos par
mthong ba'i phyir (GT 232, IEP 182) to something like: "because we see
that the AUTHORlTATIVENESS part must be cognized THROUGH authori-
tativeness that is a CONVENTION which has an apparent efficiency that
arises at a later tUne." I agree with Franco that "convention" is preferable to
"designation" for tha snyad, and that don byed refers to "efficienc:;y" rather
than an "object"; on the other hand, my choice of "apparent" for snang can
need have nothing pejorative about it-the "apparent" is simply that which
appears.
In most of Franco's other comments on my translation of material relating
to the first six verses, I must confess that I see few significant semantic
distinctions between my own reading and his, whether of Dharmaklrti's
verses or rGyal tshab rje's commentary. Indeed, many of the differences
between our renditions are a matter of terminological preference, while still
others-especially where a rendition of the root-verse is in question-
reflect the fact that Franco is invariably reading a given verse of
Dharmaldrti "straightforwardly," while I am reconstructing it from rGyal
tshab Ije, distilling it from the commentarial text in which it is imbedded. In
fact, Franco seems not to have fully appreciated this aspect of rGyal tshab
rje's commentarial method. While he is correct in observing that the rNam
'grel thar lam gsal byed "does not explain [the PramiilJaviirttika] word for
word" (DPC 125), it is misleading to claim that rGyal tshab Ije's commen-
tary "is more general and discursive. . . . He writes about and around the
milla-text" (DPC 125), for rGyal tshab rje does indeed incorporate the
entire root-text of the PramiilJasiddhi chapter into his commentary, usually
in the syn tactical order of the Tibetan translation, and usually verbatim. It is
that incorporation of the root-text that is reflected in my capitalizations
within my translation of rGyal tshab Ije, and it is that, in turn that generates
my version of root-verses presented at the head of each section, and gives
the translation its occasionally "eccentric" character.
Thus, in the case of PV 2 (smra ba po yi byed pdi yul / don gang blo la
rab gsal ba / de las gra ni tshad ma yin / don gyi de nyid rgyu can min, GT
230, IEP 178, DPC 117) it is perfectly understandable how Franco might
arrive at the version he does (''The validity of a word relates to the thing that
forms the object ofthe speaker's activity [and] appears in the cognition [of
the hearer] (i. e., the meaning of the word); it does not depend on the reality
of [that] object") through a straightforward reading of the PV verse. Mine,
on the other hand ("WORDS IN THE [SASTRAS] ARE NOT CAUSE-POS-
SESSING REAL I OBJECTS, AUTHORITIES; THEY BRING ABOUT / IN THE
MIND A CLEAR [APPEARANCE OF] SOME OBJECT / THAT IS THE OBJECT
140 JIABS 20.1
THE SPEAKER [DESIRES TO EXPRESS]") is dependent upon rOyal tshab
Ije's presentation of the verse within his commentary. Here, as in a number
of instances, he has altered the sequence of the verse to make his point, so it
might appear as if a wildly off-base translation of the root-verse has been
offered. On the other hand, if one reads rOyal tshab Ije with the root-
verses close at hand, and! or notices in my translation of rOyal tshab Ije the
capitalized words that indicate his direct quotations from the root-verse, it
becomes evident that the translation of the root verse makes perfect sense in
the context of rOyal tshab rje's discussion. Our differences in approach
and translation notwithstanding, I think that here, as so often, Franco and I
actually have arrived at the same point, in this case, that words are authori-
tative to the extent that they demonstrate that a speaker intends a particular
object, and that their authoritativeness as words is unconnected to the real-
ity or unreality of the objects they designate.
In any case, whether or not Franco and I have differences that are
"significanf' (this obviously being a relative term), let alone whether or not
one of us is "righf' or "wrong" in a given instance, let us recall that verses
1-6 of the PramiilJasiddhi chapter are unusually dense and difficult, as well
as atypical of the chapter as a whole in their purely epistemological subject-
matter. They have bedeviled every translator that has attempted them, and if
I also have failed them on occasion, I am in good company. Let me reiter-
ate, though, that my "failure" in regard to them would be egregious only if I
were attempting a direct translation from the Sanskrit original, or perhaps
from the Tibetan translation. Neither of these, however, is the source for
my version, which is, rather, drawn from the context of rOyal tshab Ije's
incorporation of them into his commentary.
(2) Franco then turns (DPC 121ff.) to PV 34
9
to demonstrate that I
stumble even when confronted by "relatively simple" verses. In the process
of attempting to do so, he supplies a fascinating excursus on the commen-
tarial tradition surrounding this verse, noting the variety of readings sup-
plied by the likes of Devendrabuddhi, Ravigupta, Manorathanandin and
Prajiiakaragupta. His strongest criticism is reserved for my reading of the .
first line of the verse, sgrub byed thugs rjes goms las de, which I translate
"ACCUSTOMATION WITH COMPASSION IS THE ESTABLISHER" (IEP
221), and which he would render "Compassion is the proof [of the
Buddha's being a means of knowledge]. That [compassion arises] from
repeated practice" (DPC 121-122). Translation choices aside, the difference
9. sgrub byed thugs rjes goms las de! blo ni Ius la brten pa'i phyirJ gorns pas
grub pa rned ce nal rna yin brten ni bkag phyir ro (OT 175ff., IEP 221ff.).
JACKSON 141
between Franco's reading and mine is over whether the "establisher" (or
"proof') of the Buddha's authoritativeness is "accustomation" (with com-
passion) or "compassion" itself. Now, Franco himself admits that each of
these is acknowledged by Prajfiiikaragupta as a possible way of reading the
verse, and he concedes that the Sanskrit karur;a (thugs rjes) is ambiguous
enough that it might be read either as (a) a nominative in agreement with
siidhanii or (b) as part of a compound, karulJiibhyiisiit, in which it func-
tions as an object for abhyiisiit. He argues that Sanskrit grammatical rules
make the first reading preferable, and that Sa skya PaIJ.gi.ta's Tibetan trans-
lation removes the ambiguity entirely.
Matters are not, however, so unambiguous when we look at the way in
which rGyal tshab Ije incorporates the verse into his commentary. The rel-
evant Tibetan here is:
thugs rje chen po de tshad ma'i skyes bu de sgrub par byed pa la / dang
por sngon du song dgos te / dang por mtha' dag sdug bsngal grol bar
'dod pa'i snying rje bskyed nas / de'j rjes su sdug bsngal zhi ba'i thabs la
goms par byas pa las ston par 'gyur dgos pa'i phyir / thugs rje chen po
de chos can / rgyu med dang ma mthun pa'i rgyu las mi ' byung ste / rang
gi rigs' dra snga ma goms pa las grub pa'i phyir / snying rje chen po de
nyid theg pa chen po'i lam sgom pa'i thog ma'i sgrub byed yin pa dang .
. . . (GT252)
I translate:
When the Greatly Compassionate One accomplished [the state of] an
authoritative person, [his accomplishment] necessarily fIrst was preceded
[by great compassion], because it was necessary that, having first gener-
ated the compassion that desires to free [sentient beings] from all their
sufferings, he then accustomated a method of pacifying suffering, and
from that he became the teacher. Great compassion does not arise cause-
lesslyor from inappropriate causes, because it is accomplished through
ACCUSTOMATION WITH previous homogenes. Great COMPASSION IS
THE ESTABLISHER of the beginning of meditation on the Mahayana path .
. .. (lEP 222)
It is interesting to note that, though he usually reproduces both the order
and exact wording of the Tibetan root-verse in his commentary, rGyal tshab
rje here never adds the instrumental s to thugs rje, leaving the exact relation
between that central term and both sgrub byed and goms less than totally
obvious. His discussion is further complicated by the fact that, here as
142 nABS 20.1
elsewhere, he use$ the term sgrub byed in a double sense, as meaning both
logical proof and spiritual accomplishment In this particular instance, the
grammar and syntax seem to me to point toward the latter reading, 10 and
where spiritual accomplishment (i. e., becoming an authoritative person) is
at stake, it is not compassion itself, but its accustomation (or, as Franco
would have it, "repeated practice"ll) that is the "establisher' (sgrub byed) . . I
have made it clear in my notes and elsewhere that I understand perfectly
well that when we take sgrub byed in the sense of logical establishment,
that it is "compassion" (rather than "accustomation with compassion") that
serves as the logical reason; it is only when we take the term as referring to
spiritual establishment that "accustomation" may serve as the "esiab-
lisher"-but it is primarily in this latter sense that I see rGyal tshab rje
interpreting the verse, hence my retrojection of a translation of the root-
verse that would seem at odds with the grammar of the verse.
12
The type of argument I have made here about PV 34a may be applied,
mutatis mutandis, to my retrojected translation of PV 34bc,13 which, like
that of 34a, may not fit a straightforward reading of the Tibetan, let alone
the Sanskrit, but accurately reflects the way in which rGyal tshab rje incor-
porates the verse into his commentary (see GT 252, IEP 223). This much
Franco concedes (DPC 125), and yet he proceeds upon a lengthy disquisi-
tion on references to Lokayata similes in DharmakIrti's commentators (OPC
125-129). His comments are erudite and interesting, but the springboard
for them-his sense that my statement that "[t]he probative reasons and
examples are supplied by rGyal tshab Ije" (IEP 223, n. 6) implies that rGyal
10. It might be possible to construe thugs rje chen po de tshad mdi skyes bu de
sgrub byed pa la as "As for compassion, the proof of authoritativeness"; how-
. ever, the fact that "authoritativeness" (not "compassion") appears to be the subject
of the f o l l o ~ i n g clauses makes this reading a less promising one.
11. As noted already in the book (IEP 169, n. 11), fm not thrilled with
"accustomation" as a translation for goms/ abhyiisa. It may, perhaps, convey
greater passivity than "repeated practice"; on the other hand, unlike "repeated.
practice," "accustomation" does manage to convey the cumulative sense that is an
important aspect of the term.
12. My need to respond to Franco in such detail has made me think that I should
have commented on this issue in a footnote. I did note rGyal tshab rje's deliber-
ately ambiguous usage of sgrub byed upon its first appearance in the commen-
tary [lEP 170, n. 14], but a reiteration of the fact in relation to verse 34a might
have conserved both ink and spleen.
13. "[THE LOKAYATAS] SAY: '[COMPASSION] IS NOT ACCOMPLISHED
THROUGH ACCUSTOMATION,/ BECAUSE MIND IS BASED ON THE
BODY.'"
JACKSON 143
tshab rje "designed certain reasons. . . and provided their examples inde-
pendently on the basis of his acquaintance with Indian Lokayata sources"
(DPC 126)-requires a reading of my text that is, to put it gently, rather
imaginative. .
Furthermore, his condescending observation (DPC 126), that to under-
stand the Lokayata similes it is not necessary to be familiar with the ideas of
Gilbert Ryle, for "[a]ll one has to do is read the classical commentaries on
this verse," demonstrates how little he understands of the philosophical
purpose of my footnotes (or the book as a whole), which is to provide edu-
cated readers who do not have access to "the classical commentaries" with
some tools whereby they might understand Dharmaldrti's and / or rGyal
tshab rje's analysis in terms that are familiar to them. I am not suggesting
that there is a precise parallel between Lokayata arguments and those of
Ryle, or some other materialist; I am suggesting, here and throughout the
book, that there are strong analogies between the issues and arguments
developed by Indians and Tibetans and those that have arisen in the West.
(This, incidentally, strikes me as a potentially controversial claim that actu-
ally is worth arguing about, but Franco does not raise it at all; I begin to
suspect that he simply did not see the philosophical concerns that animate
the book.)
Thus, as with the more recondite verses at the Pramii1)asiddhi chapter's
outset, so with later and "simpler" [sic 1 verses, an examination of the way in
which rGyal tshab rje construes DharmaIdrti. often will produce a reading
of the root-verses that is different from that yielded by 'a "straightforward"
reading of either the Sanskrit or Tibetan of Dharmaldrti's text-but should
not, simply for that reason, be considered a "wrong" translation of
DharmaIdrti., any more than an accurate translation of rGyal tshab rje
should be considerd "wrong" if it seems to differ from what a
"straightforward" reading of the Sanskrit (or even, on occasion, Tibetan)
version of the verses would lead an interpreter to think rGyal tshab rje
ought to be saying. These last comments bring us squarely to the real nub
of Franco's critique of Is Enlightenment Possible?, which is methodolog-
ical, and it is to those issues-in my opinion the most interesting he
raises-that I will turn at last.
It is Franco's contention (DPC 131) that my translation leaves us with a
view not only of DharmaIdrti.'s original verses, but of rGyal tshab rje's
commentary upon them, that is spoiled by the "indistinct haze" emitted by a
faulty methodology. Franco's methodological criticisms are scattered rather
unsystematically throughout his review, but I would summarize the major
144 JIABS 20.1
points, in descending order of generality, as follows: (1) It is a mistake to
assume that commentary, by its very nature, distorts the meaning of an
original text; to make this assumption is to invite indifference to the original
text and an improper appreciation for the accomplishment of the commenta-
tor; (2) it is a mistake, all too typical of North American scholarship, to
attempt to present Tibetan commentaries on Indian texts without careful,
indeed, primary attention to the original Indian sources, both the "root-texf'
and its commentarial tradition; and (3) rGyal tshab rje's commentary on the
PramiilJaviirttika cannot stand comprehensibly on its own, for if it is
allowed to do so, we lose sight of the original it is supposed to be explain-
ing (and through which we should explain it), to the point where it might
even appear that rGyal tshab rje has failed to understand Dharmaldrti prop-
erly! Let me briefly consider each of these points in turn.
(1) Franco cites with considerable disapproval (DPC 124) my statements
to the effect that (a) original Indian philosophical texts often require com-
mentary to be comprehensible (IEP 11, emphasis added), but (b) commen-
tary necessarily involves distortion of the original, and the greater the tem-
poral and geographical distance between a commentary and its original text,
the greater the likelihood of distortion (IEP 159). This, for Franco, entails
the following absurd syllogism (DPC 124): "All root texts require com-
mentary to be comprehensible. All commentaries distort their original root-
texts. Therefore, root-texts have to be distorted to be comprehensible."
This is an amusing bit of sophistry, and provides Franco with the main
title for his review, but makes a caricature not only of my position, but, I
suspect, of his own as well. I do not, in the first place, claim that all root-
texts are obscurely concise. On the other hand, the existence of a large and
lively commentarial tradition surrounding a text like the PramiilJaviirttika
seems to testify in part at least to the fact that its meanings are not all and
instantly apparent, even to a highly educated reader 14; surely Franco would
concede that many important Indian texts can be and have been elucidated
by subsequent commentary. Nor, I suspect, would he want to deny that
commentary does, inevitably, involve some distortion: one does not have to
adopt a radical reader-response approach to authors and their texts 15 to
14. Recall the legend of DharmakIrti's repeated destruction of Devendrabuddhi's
attempts at a PramiilJaviirttika commentary, whose third draft he finally, grudg-
ingly, accepted as barely adequate (IEP 114).
15. For a recent Buddhological example, see C.W. Huntington, Jr., "A Way of
Reading," The Journal of the International Association of Buddhist Studies 18.2,
(1995) 279-308. If Franco is unhappy about the degree to which my book,
intended in part as a contribution to DharmakIrti studies, seems often to have lit -
JACKSON 145
appreciate the broader hermeneutical point, made by Heidegger, Gadamer
and so many others, that any reader of or commentator upon a text bears an
ambiguous relation to it: he or she seeks to understand or explicate what
seems to require clarification, and in the process inevitably brings to bear
his or her own culture and concerns, hence at the same time both illuminat-
ing and obscuring the text that is under consideration.
Now, the fact that texts are difficult to read and that the commentaries
they prompt will distort them to some degree, does not mean that there are
not better or worse commentaries, or that an author's original meaning is so
utterly lost to us (let alone unimportant or nonexistent) that we should not
bother attempting to retrieve it. I believe that it is precisely the task of his-
torically informed philological scholarship to attempt such retrievals. Let
me simply reiterate that the text and author I am concerned to ''retrieve" in Is
Enlightenment Possible? is not primarily Dharmakirti, but rGyal tshab rje,
and that the philosophical standpoint I am most intent on exposing is that of
the Tibetan, rather than the Indian, tradition.
(2) My focus on rGyal tshab rje at the apparent expense of Dharmakirti
Franco finds "symptomatic for [sic] a currrent trend among [North Ameri-
can] scholars of Tibet who attempt to understand the Tibetan philosophical
tradition 'as such' independently of the decisive background and long-last-
ing influence of the Indian tradition" (DPC 109-110). But, Franco main-
tains, "a genuine understanding of the older indigeneous Tibetan commen-
taries on Indian Buddhist texts, or of independent works mainly based on
these texts, is not possible without a thorough, first-hand understanding of
these 'root-texts' and their Indian exegesis, of course, in their original lan-
guage if they are preserved in it" (DPC 131).
Now, I would certainly agree that the more one knows about the back-
ground of any text, the more nuanced one's appreciation and exposition of it
is likely to be. Thus, any Tibetan commentary or treatise, especially one
that is linked explicitly with an Indian text, is likely to be most fully under-
stood against the background of its Indian forerunners. The question is not
whether such background is interesting or even important, but, rather,
whether, as Franco maintains, it is essential. I would maintain that such
background-especially a detailed exploration of it-is only essential in a
primarily Indological context, where one's real concern is with the original
Indian text, and the Tibetan commentary or treatise is valued only for the
tle to do with Dharmakirti, I can only imagine his apoplexy in the face of
Huntington's discussion of Niigiirjuna, which seems to relate to practically
everyone but the great Miidhyamika!
146 nABS 20.1
light it may shed on the original. . If, on the other hand, one's concern is
primarily with the Tibetan textual or philosophical tradition, a detailed
exploration of the Indian background sources is of considerably less impor-
tance than an attempt to understand the meaning that the Tibetan text must
have had for its author and audience--even if that meaning seems at times
to "deviate" from that which one would expect from a careful consideration
of the Indian background texts.
What is more, the attempt to read a Tibetan text through an Indological
prism--even a very well-constructed one-runs as great a risk of distorting
the way a text was received in Tibet as does its converse, reading a Sanskrit
text through its Tibetan translation or commentary and then claiming that
the original has thereby been captured. I doubt seriously that Franco's
characterization of the methods of North American Tibetology, cited above,
is very accurate. To the degree, though, that there have been and are schol-
ars (e. g., Hopkins, Klein, Lopez, Cabez6n, myself) who in some of our ,
works focus quite deliberately on the Tibetan side of the Indo-Tibetan Bud-
dhist equation, our methodology is irresponsible only if we make excessive
claims for it, i. e., naively insist (or, perhaps more insidiously, imply by
silence) that (a) Tibetan translations are faithful to their Sanskrit originals,
(b) Tibetan commentators perfectly preserved the interpretive traditions of
Indian Buddhism, and therefore (c) Tibetan translations and commentaries
give us access to the unalloyed meaning of the original. If, on the other
hand, we make it clear that we are concerned above all to represent a pri-
marily Tibetan, rather than Indian, textual and philosophical tradition, and
make no claims to the effect that the Tibetan interpretation is a philologically
accurate equivalent of the Indian original, then I think we have discharged
our methodological obligations in good faith.1
6
(3) What" finally, of the method I have applied to the presentation of
DharmakIiti and rGyal tshab rje in Is Enlightenment Possible? While I can-
not deny that my translation and exposition of rGyal tshab rje's
Pramii1J.(lviirttika commentary might have been enriched by a more detailed
consideration of the Indian commentarial tradition than I have undertaken, .
or that I might have clarified more often than I did the ways in which rGyal
16. This point has been made not only by many in the "younger" generation of
North American Buddhist scholars, but also by such very different, yet
respected, representatives of the previous generation as Herbert V. Guenther (in
many works) and David Seyfort Ruegg (especially in his Leiden inaugurallec-
ture, The Study of Indian and Buddhist Thought: Some Problems and Perspec-
tives [Leiden: E. J. Brill, 1967]).
JACKSON 147
tshab Ije's commentarial glosses on Dharmaldrti entail a less-than-straight-
forward reading of his root-verses, I am not convinced that my omissions
on this score reflect an essential weakness in either the approach or the
content of the book. My central purpose, recall, was "to make available ...
to readers of English an important source of past and present Buddhist
philosophizing" (IEP 13), namely, rGyal tshab rje's commentary on the
PramiilJasiddhi chapter of the Pramiir;,aviirttika, which has been a pro-
foundly influential source of dGe lugs pa Tibetan philosophy of religion
from the time of its composition down to the present-where it still is
actively studied and expounded. I believe that it is entirely possible to read
rGyal tshab rje's text from a Tibetan-oriented perspective that employs
Indian sources sparingly, and have it make sense as a philosophical docu-
ment. That, after all, is precisely the way that most of rGyal tshab rje's
contemporaries read his text, and certainly the way it has been read by dGe
lugs pa intellectuals since. If I have presented rGyal tshab rje's version of
Dharmaldrti's arguments accurately most of the time, and cast some light on
them in my footnotes, then I will have succeeded in the central purpose for
which I undertook the translation.
Franco, of course, believes that I have not presented rGyal tshab rje
clearly most of the time. I hope, however, to have shown above that,
almost invariably, his sense that I have misconstrued the rNam 'grel thar
lam gsal byed is a result of his inability to accept that the way in which
rGyal tshab rje utilizes the Pramiir;,aviirttika may sometimes take him quite
some distance from a "straightforward" reading of the Tibetan translation,
let alone the Sanskrit original, of the root-verses. It is interesting to observe
that, as unambiguously critical as Franco is of my approach to rGyal tshab
rje, he seems very much of two minds about rGyal tshab rje himself. Ear-
lier in his review, he maintains that rGyal tshab rje's "differences" from
Dharmakirti cannot be so great as they sometimes seem in my translation,
for this would imply not only that rGyal tshab rje had misunderstood
Dharmaldrti, but that the entire Tibetan tradition of PrarrillI,la commentary
had been considerably over-valued. Later in the review, however, he seems
increasingly to acknowledge that there are significant ways in which rGyal
tshab rje departs from Dharmaklrti; as a result, he ends up quite ambivalent
about the value of rGyal tshab rje's commentary-ambivalent enough, it
appears, that the distorted version of Dharmaklrti he sees himself as criticiz-
ing in the review is, in his subtitle, not mine alone, but "the rGyal-tshab-
Jackson interpretation." The problem with insisting either that rGyal tshab
rje must have read Dharmaldrti "correctly," hence I have misread rGyal
tshab rje; or that rGyal tshab rje has strayed sufficiently from Dharmakirti
148 JIABS 20.1
that his commentary can have little value, is that both positions are based on
a thoroughly Indocentric perspective, wherein a Tibetan commentary's only
importance is in its faithful (or unfaithful) representation of an,original
Indian text. Once one frees rGyal tshab rje from the requirement that he be
read through DharmakIrti and the Indian tradition or not all, it becomes
possible to regard him as a Tibetan intellectual who is intrinsically i n t e r e s t ~
ing in his own right, as well as for the ways in which (like all commenta-
tors) he appropriated his source text in ways that sometimes were "faithful
to the original" and sometimes quite "creative" (even if distorting) in their
utilization of it.
As for my presentation of DharmakIrti: I made it clear in the book, and
have reiterated here, that the verses that head each section of the translation
are intended lWt as straightforward renditions of the Tibetan-let alone the
Sanskrit-of Dharmaldrti's verses, but as indicators of the way in which
rGyal tshab rje has used them in his commentary. Sometimes, therefore,
my translation may happen to coincide with a straightforward reading of the
Tibetan (or even Sanskrit) verses, but often it will not. As long as the
reader does not think that I am trying to pawn off those verses, or the
commentary, as something they are not, and is content with exploring an
argument that is Dharmaklrtian without always being precisely
DharmakIrti's (do we cease to study Plotinus because his doctrine is Pla-
tonic without being Plato's?), he or she may find my translation of some
value. If, on the other hand, like Franco, the reader is an Indological and
philological fundamentalist, or an uncompromising DharmakIrti-centrist, he
or she may be incapable of seeing any value in a book that is frankly
Tibetological and philosophical, and only secondarily concerned with
issues of commentarial fidelity. In that case, I can only regret the disap-
pointrnent I have caused (and, of course, any errors that I have committed),
but not that! wrote the book that I did, or as I did. The ocean of Buddho-
logical meanings and methods is vast, and I am confident that there is room
in it both for the sorts of fish favored by Franco and for the types that I pre-
fer. And, as Kierkegaard might remind us, the secret (perhaps not the
whole secret, but a very real part of the secret) to our readings and delecta-
tions-and even to the words ofleamed commentators we seek to under-
stand-lies, at last, in an arbitrariness that never entirely can be eliminated,
but can, perhaps, be attenuated to the degree that we are willing to acknowl-
edge it.
ELI FRANCO
A Short Response to Roger Jackson's Reply
I am honored that Roger Jackson considered my review worthy of
response. Unfortunately, I find his response as flawed as his book. The
passages I discussed were not chosen arbitrarily; they simply form the
beginning of the first two sections of his translation (on pramiilJabhatatva
Naturally they constitute only a sample, but I won-
der why Jackson is so confident that they are unrepresentative for the
whole work ("a few errors in a huge, translated work"). Let him rest
assured: almost every single verse of his translation is faulty. The crux of
the matter, however, is whether Jackson's mistaken translation represents
rGyal tshab's understanding. I will argue against it not by adducing fur-
ther "arbitrary" examples, but by explaining what is wrong with his
methodology. Jackson assumes that if a word, which appears in the verse,
appears in the commentary in certain manner, then rGyal tshab interprets
the word in the verse in that manner. However, this assumption is unjus-
tified. Furthermore, what happens if a word appears in two or more
statements? Lefs look again at GT252 / IEP222 quoted by Jackson in the
response. Apart from obvious mistakes (stan par 'gyur dgas pa ... is not
"he became the teacher" but "the purpose / motivation of becoming a
teacher," ma mthun pa is not "inappropriate" but "dissimilar," etc.), we see
that thugs rje appears twice: the first time it is mistranslated
("compassionate',),but even the second time it is not capitalized (i. e., not
taken as "incorporation"). On the other hand, the translation of sfiin rje,
which also appears twice, is capitalized once. Yet Jackson argues that
thugs rje (not sfiin rje) is not an instrumental. Even if this were of any
relevance to his mistranslation "ACCUSTOMATION WITH ," Jackson is
inconsistent in the application of his own method because he should have
capitalized "ACCUSTOMATION WITH previous HOMOGENES" and
claimed that rGyal tshab construes "accustomation" with "homo genes"
(svajiiti) in the following verse. Even Jackson shrinks from such an inter-
pretation, although philosophically it represents Dharmaldrti's opinion
correctly. The fact that it crosses the boundaries of a verse is not unusual
149
iso nABS 20.1
for DharmakIrti's statements. In the final analysis Jackson will have to
admit that he did not capitalize "homo genes," because it is grammatically
impossible to read the verse in this way. In numerous cases, h o ~ e v e r , he
attributes grammatically impossible or highly improbable readings to
rGyal tshab. Furthermore, when rGyal tshab "has altered the sequence of
the verse" Jackson translates accordingly; however, it is obviously absurd
to assume that rGyal tshab understood verses in reversed order.
Consequently, Jackson's translation of the verses does not represent
rGyal tshab's understanding. Jackson erroneously maintains that "the
capitalized words ... indicate direct quotations from the root-verse." He
simply confuses quotations and more or less close references to what'is
expressed in a verse. When rGyal tshab quotes, he adds ces, etc. Let me
repeat myself: often it is impossible to "distiIY' rGyal tshab's interpretation
of the verses word for word, certainly not by Jackson's method, which is
mechanical, arbitrary, and furthermore disregards rGyal tshab's
hermeneutical situation. Would rGyal tshab really suggest repeatedly and
nonchalantly readings of the verses which blatantly go against
DharmakIrti, Devendrabuddhi, PrajiHikaragupta, Ravigupta and Sa skya
PaI)qita taken together? For Jackson this is not even problematic. How-
ever, the "incorporation" of words from the verses in the commentary
does not imply that rGyal tshab was the ignorant, misunderstanding, mis-
interpreting, distorting, "innovative" fool that Jackson would have us
believe. Obviously, commentaries may distort, but the kind of literal,
ungrammatical distortion assumed and presented by Jackson is, to my
knowledge, unprecedented in any Indian or Tibetan commentary. To ren-
der sa ... asiddho 'bhyasaJ; ncompassion is not accomplished by accus-
tomation" is absurd, and to attribute this interpretation to rGyal tshab is an
insult to tragitional Tibetan scholarship.
Naturally Jackson attempts some damage-control, but his approach is,
again, not quite scholarly. He takes some of my statements out of con-
text, distorts others. (I did not "insist" that Jackson "should be reading
rGyal tshab rje via DharmakIrti"; 1 simply claimed that Jackson mistrans-
lated and misinterpreted both DharmakIrti and rGyal tshab, and 1 substan-
tiated this claim with a considerable number of examples.) He minimizes
and obscures the differences between us as well as the relevance of the
general problems and mistakes pointed out by me (our differences are not
even "significant"). His main strategy, however, is to limit the scope of
my criticism: 1 focus "on passages and issues that are of secondary conse-
quence." 1 address only "selected items" in the glossary and "two sam-
plings" of the translation. Even concerning these insignificant parts, what
FRANCO 151
I have to say is hardly relevant, because Jackson conveniently compart-
mentalizes the whole issue: he is the soaring philosopher-Tibetologist, I
the lowly, "fundamentalist" philologist-Indologist who is not interested in
philosophical discussions, or worse, cannot understand them. Jackson's
approach is unacceptable: it condones ignorance, promotes shallowness,
and, in this case, led to distortion that is not rGyal tshab's but.1ackson's
own. Jackson certainly deserves a more response than was possible here
within the limited space granted to me.
TOM J. F. TILLEMANS
On a Recent Translation of the San:tdhinirrnocanasiltra
Wisdom of Buddha. The Sarr.tdhinirmocana Siitra. Translated by John
Powers. Tibetan Translation Series 16. Berkeley CA: Dharma Publishing,
1995. Xxii and 397 pages, 14 illustrations and line drawings.
Following the traditional explanation by Mahayanist schools, Buddhist sutras
are classified according to three cycles (dharmacakrapravartana). The
first consists in the texts of the so-called "Small Vehicle" (h'fnayiina), which
supposedly taught the non-existence of a real personal identity, reducing it
to real elements (dharma). The second was the voluminous collection of
the Prajiilipiiramitiisiitras, which were said to have taught the idea of the
unreality of persons as well as that of the elements, i. e. their voidness
(siinyatii) and absence of any own-nature (ni/:tsvabhiivatii). The third cycle
consists, inter alia, of reconciliations and reinterpretations of the first two,
elaborating a hermeneutic whereby apparent conflicts and unacceptable philo-
sophical consequences of certain statements in the ftrst and second cycles
could be taken as being of only provisional meaning (neyiirtha) and not in
actual contradiction with the deftnitive meaning (n'ftiirtha) of the Buddha's
teaching, because these statements were supposedly only designed to lead
disciples, therapeutically and with non-literal language, to an ever closer
approximation of the deftnitive thought of the Buddha. The Sarr.tdhinir- .
mocanasiitra (henceforth "SNS") is traditionally classifted as belonging to
this "third cycle" of Buddhist teachings. It has as one of its main goals the
interpretation of very provocative statements in the Prajiiiipiiramitiisiitras
(henceforth "PP"), namely the passages where these sutras clearly and re-
peatedly say that all dharmas, indeed everything there might be, is without
own-nature (ni/:tsvabhiiva), unproduced I unborn (anutpanna), undestroyed
(aniruddha) , primordially calm (iidisiinta) and essentially in nirva.Q.a
(prakrtinirvrta ).
SNS effects its hermeneutical tour de force by introducing a famous
schema of three natures (svabhiiva), so that when PP speaks of all dharmas
being ni/:tsvabhiiva, anutpanna etc., these words are to be taken as having
provisional meaning, needing interpretation. The SNS, which not surprisingly
claims to give the deftnitive meaning of the Prajiilipiiramitiisiitras, reinter-
153
154 JIABS 20.1
prets the controversial passages about all dharmas as being amphibolic:
each of the three natures is niJ:zsvabhava, anutpanna, etc. in its own very
specific way. Only the "imagined nature" (parikalpitasvabhava) of dhannas,
which is as thoroughly unreal as a "flower in the sky," can be literally said
to imply a lack of own-nature, etc. DhanDas thus have a type of niJ:zsvabhavata
in the sense that they have, in themselves, no real characters such
as material properties, like shape or color, or metaphysical properties like
oneness, manyness, etc.; such "characters" are exclusively mind-created and
language-dependent. The other two natures of dharmas, viz. conditionality
or "other-dependency" (paratantrasvabhava) and the fmal, or "perfect" (pari"
mode of being, however, imply types of niJ:zsvabhavata whkl:J._
are quite different from the implied by the imaginary.
The conditional nature of dharmas, for example, implies a lack of own-nature
only in the sense of its type of production (utpattiniJ:zsvabhavata), i. e., the
dharma arises completely in dependence on other things and has no nature
which it itself would cause and which would thus be self-produced. The
"perfect" nature implies a lack of own-nature concerning the ultimate or
absolute, a paramarthaniJ:zsvabhavata, although as we shall see below the
Sanskrit term is somewhat difficult to translate adequately, given that it is
explained in two quite different ways in the SNS.
The SNS is one of the main sources for this "three-nature" doctrine as
well as for the fundamental ideas of Vijfianavada Buddhism, all of which it
skilfully uses to defuse the apparent nihilism of the PP, laying the groundwork
for an idealism where "mind-alone" exists and where the yogic transformation
and reorientation of mind becomes the main goal of Buddhist practice. It
is a main source for the doctrine of the alayavijfiiina ("storehouse-
consciousness") (chap. 5) and the meditational method of samatha ("qui-
etude") and vipaSyana ("insight") (chap. 8); it develops themes such as the
seven types of tathata ("thusness") and the eighteen sorts of siinyata ("void-
ness"), the bodies of the Buddha, the stages (bhiimi) of bodhisattva practice,
and even has a section on logic in its tenth chapter developing four types of
yukti. This sutra thus is of key importance for the Y ogacara-Vijfianavada
schools in India, in Tibet and in China: it has several different Chinese
translations (some only partial) and one Tibetan translation (found in various
editions of the Tibetan canon); it has four canonical commentaries; it was
extensively used by indigenous Tibetan writers like Tsong kha pa; it also
figures prominently in the so-called "other-voidness" (gzhan stong) schools
which synthesized Y ogacara and Madhyamaka thought. It is without any
exaggeration one of the most important texts of the Mahayana.
The SNS was translated into French in 1935 by Etienne Lamotte, who
TILLEMANS 155
preceded his translation with a detailed discussion of the history of the text,
its importance, the different translations, the contents of the chapters, the
dateof the sutra's composition and the philosophical ideas which are devel-
oped there. 1 He then gave us an edition of the Tibetan text and a philologically
sound, annotated translation of the SNS based on the Tibetan and Chinese
(the Sanskrit was and still is lost). One can contest points-sometimes the
translation is perhaps too free, sometimes it is unclear whether Lamotte
based himself on the Chinese or the Tibetan. There can no doubt be major
revisions to Lamotte, but the fact is that Lamotte's work on SNS has been
and remains a classic in the field of Buddhist Studies, and justifiably so.
Equally one should mention the German translation of extracts from Chapters
VI and VII of SNS to be found in Erich Frauwallner's 1956 work Die
Philosophie des Buddhismus? Frauwallner states that he translated from
the Tibetan. And although the translation has almost no notes, it remains
valid and reliable. It and Lamotte's translation should be read, and read very
closely, bj anyone who wishes to translate the SNS. The SNS deserves
that much.
Recently, John Powers published an English translation of the SNS.
Powers's audience is in part a lay North American public-as the editor of
the Tibetan Translations Series, Tarthang Tulku, stated, the goal was. to
present the sutra in readable English to convey the deep and subtle meanings
of the text. Powers, thus perhaps understandably, did not develop the
detailed philological and historical aspects discussed in Lamotte, but instead
concentrated on translation. He based himself upon the edition found in the
sDe dge Tibetan canon, relied heavily on the commentaries in Tibetan by
indigenous writers, as well as on those by Indian writers such as Asariga
and Vasubandhu, and mentioned a debt to oral explanations from contempo-
rary Tibetan scholars. The unfortunate fact is that, despite the plethora of
sources which Powers claims to have used,4 the translation is often quite
unreliable, having errors which seriously obscure the basic sense of the
1. San;dhinirmocana Sutra. L'Explication des mysteres. Texte tib6tain edite
et traduit par Etienne Lamotte. Paris et Louvain, 1935.
2. Recently reprinted for the fourth time by Akademie Verlag (Berlin: 1994).
3. Cf. Lamotte's conclusion on the importance of the SNS (op cit. p. 24): "En
somme, dans son etat actue1, Ie San;dhinirmocanasutra est un resume complet
du Grand Vehicule bouddhique aux premiers siecies de notre ere."
4. His bibliography mentions Lamotte's translation (but not that of Frauwallner)
and the various Chinese texts. In his introduction (p. xx) we find: "In my
studies, I have consulted ten different Tibetan editions, as well as three Chinese
editions, and have noted their variant readings."
156 JIABS 20.1
slUra's words, let alone its deep and subtle meanings. In the case of
difficult passages, Powers would certainly have profited if he had used
Lamotte and Frauwallner. But there is little evidence in this work that he
used them at all. Moreover, the Chinese of Xuanzang would also have
helped, as we shall try to show, to clarify certain points.
Let us examine a few representative passages from different chapters in.
some detail, citing the Tibetan, and, where profitable, Lamotte, Frauwallner
or the Chinese of Xuanzang.
In Chapter I (Powers p. 12f.) we have an interesting argument whic!J.
comes back later and which is muddled both times by Powers. Let us look
at the details. On Powers's p. 12 we have brjod pa ni dngos po med pa
can yang ma yin te dngos po de yang gang zhe na ... Now, Lamotte had
it right when he translated (p. 170): "Mais, dira-t-on, une expression ne va
pas sans un objet designe. Que1 est donc ici l'objet?" The point is that this
whole passage, as Lamotte clearly shows, is an objection which can be
summarized as: "But words must have objects, so what is the object here?"
Powers, however, translated the Tibetan as: "An expression is also not
without thingness. What then is a thing?" Granted "thingness" for dngos
po is gauche, but this is perhaps not the real problem. The real problem is
that it is not clear that Powers saw the passage as an objection at all. His
rendering of ... zhe na turns the passage into an oddly general question
"What is a thing?", and the point is lost. Partly responsible for this transfor-
mation of the objection into a general question is Powers's rendering of the
first occurrence of yang as "also"-yang should be contrastive here (i. e.
"but"), as is clear from Xuanzang's Chinese (T. no. 676. f. 689a5) ~ ~ ~ ~
*W1ffflIDl ("But there is no expression lacking a thing") which uses ~ .
In short, wehould translate: "But an expression does not lack a thing
[which is its object]. What then is that thing?"
Now, let us look at the SNS passage which answers the objection. Here
is the Tibetan: 'phags pa mams kyi 'phags pa'i shes pa dang I 'phags pa'i
mthong bas brjod du med par mngon par rdzogs par sangs rgyas pa
gang yin pa ste I brjod du med pa'i chos nyid de nyid mngon par rdzogs
par rtogs par bya ba'i phyin 'dus byas zhes ming du btags so I. Powers's
translation on p. 13: "It is that to which the Aryas completely and perfectly
awaken without explanation, through their exalted wisdom and exalted vision.
Because they have completely and perfectly realized that very reality which
is inexpressible, they designate the name 'compounded.'"
What could it mean to say, as Powers would have it, that the Aryas use
the word "compounded" because they understand the inexpressible reality?
TILLEMANS 157
Compare Lamotte (p. 170): "C'est la [realite] ineffable sur laquelle les Saints,
par Ie saint savoir et la sainte vue, sont parfaitement eclaires; mais pour
eclairer les autres sur l'ineffable Nature des choses, ils ont forge l'appellation
conditionne." Lamotte clearly is close to Xuanzang's Chinese here: ...
(f. 689a7-8) "Since they wished to
make others understand ... ". Powers, in effect, should have read a dative
for the Tibetan ... de nyid mngon par rdzogs par rtogs par bya ba'i phyir,
i. e., "in order that [others] might perfectly understand reality," rather than
an ablative ''because they [i. e. the A.ryas] understand .... " Surely it makes
more sense to say that the A.ryas understand ineffable reality, but that they
speak of compounded entities to illuminate others who have not yet under-
stood this reality. A mistranslation is all the worse here, because the whole
argument repeats itself, mutatis mutandis, later on Powers's pp. 15.1-6: we
find the exact same passage, but this time concerning the asan:zskrta ("un-
compounded").
We now go on to look at a passage in Chapter Five, "The Questions of
Visalamati." Powers mistranslates an important passage on the iilayavijiiiina
("storehouse consciousness") which gives an etymological explanation of
the term iilayavijfiiina, turning on the meaning ii If, iilfyate "settle down
upon," "melt," etc., and hence iilaya "abode," "receptacle," etc.; various such
explanations also figure in Chapter I of the Mahiiyiinasan:zgraha of Asanga
as well as in Won ch'uk's commentary on SNS. Here is the Tibetan and
Powers's translation of the etymological explanation which interests us
(Powers 70.7-9): kun gzhi mam par shes pa zhes kyang bya ste / 'di ltar
de Ius 'di la grub pa dang bde ba gcig pa'i don gyis kun tu sbyor ba dang
rab tu sbyor bar byed pa'i phyir ro /; ( Powers p. 71) "It is called the
'basis-consciousness' because there is the same establishment and abiding
within those bodies. Thus they are wholly connected and thoroughly con-
nected."
Besides having echoes in the Mahiiyiinasan:zgraha, this passage has been
discussed by Lambert Schmithausen in his Alayavijiiiina: On the Origin
and Early Development of a Central Concept of Yogiiciira Philosophy
(Tokyo, 1987), who on his p. 22 translates as follows: "[The mind-containing-
all-seeds] is also called 'iilayavijfiiina,' because it sticks to and dissolves
into or hides in the bod!, in the sense of sharing its destiny (i. e. becoming
closely united with it)." Powers regrettably missed all this.
5. Cf. Schmithausen's recontruction op cit. n. 181 iilayavijiiiinam ity apy
ucyate, yaduta tasyiismin kiiya iilayanapralayanatiim upiidiiya ekayogakJemiirth-
ena. Xuanzang T. no. 676, f. 692b. 16-17:
158 nABS 20.1
If we look at Powers's understanding of the syntax of the passage, it
should be clear that breaking the sentence at grub pa dang bde ba gcig pa'i
don gyis, not translating the don gyis = arthena and then starting 'a new
sentence with "Thus ... " is quite unacceptable: making two arguments
. here, where there is only one etymological explanation, deforms the passage
badly. In fact, SNS passage asserts that the consciousness in question can
be called iiZayavijfiiina because it clings to and hides in the body in the
sense of sharing the body's same ("fate," "welfare,"
Schmithausen devotes a considerable part of his monograph to the question
of just what was this early or even initial conception of the iiZayavijfiiina
"sticking to" and" being concealed in" the body. .
Why did Powers translate grub pa dang bde ba gcig pa =
as "same establishment and abiding"? Right away, the Chinese of Xuanzang,
which has tong anwei fiij*:fa here (literally: "same security and danger"),
would alert us that Powers's translation might well be odd, and would
suggest something more like "same destiny," the iiZayavijfiiina having the
same kinds of favorable and unfavorable things happening to it as the
body. The term figures in such texts as DharmakIrti.'s Pra-
mii1J.aviirttikasvavrtti ad k. 43 in the sense of an identity between x and y
due to an identity of destiny / vicissitudes.
6
Thus MookeIjee and Nagasaki
glossed: This is the commonplace cliche in philosophical
parlance. Things supposed to be identical must have identical yoga and
Yoga means accrual of a new advantage and means the
continuity of the status quo. That which has the same incidents, gain or loss
with another, is identical with the other.,,7 .
Interestingly enough, some indigenous Tibetan texts have interpreted
as a type of strict temporal identity, placing the key idea in
the context oJ;the typically Abhidharmic schema of the three characters
Bodbiruci T. no. 675, f. 669a 24-25:
(Paramartha did not translate ch. 5.) Lamotte p. 185:
"EIle est aussi appe16e 'Connaissance-receptac1e,' parce qU'elle se joint et s'unit a
ce corps dans une commune securite et dans un risque commun." Lamotte
largely follows Xuanzang, but with an insufficient rendering of
as use joint et s'unit a ce corps." Schmithausen in his n. 183 points out that .5'1:
has the sense of "appropriates and lies hidden"; cf. Bodbiruci's flf "dwells
in and sticks to."
6. R. Gnoli, ed., The Pramii1J.aviirttikam of Dharmakfrti, Serie Orientale
Roma 23 (Rome: 1960) 26.7-8: na hi yo na bhavati sa
tatsvabhiivo yuktal,! I. My thanks to T. Much for reminding me of this passage.
7. S. Mookerjee and H. Nagasaki, The Pramii1Javiirttikam of DharmakJrti
(patna: 1964) 99, n. 1.
TILLEMANS 159
(lak$wta) of composite things, viz. production (utpiida), abiding / duration
(sthiti) and perishing (vyaya). This temporal explanation of ekayogak$ema
as involving simultaneity in all three characteristics is what we find in
Tibetan bsDus grwa texts and, hence, in much of the dGe lugs pa philosophical
literature, which was so permeated by bsDus grwa / bsDus pa -style concepts.
Indeed, the terms grub bde gcig and grub bde rdzas gcig and grub bde
dbyer med kyi rdzas gcig are very frequently used with their basic bsDus
grwa-style meanings in texts by rGyal tshab rje (e. g. rNam 'grel thar lam
gsal byed) and mKhas grub rje (sDe bdun yid kyi mun set, etc.), dGe'dun
grub pa (e. g. Tshad ma rigs rgyan), etc. etc. Moreover, this bsDus grwa
interpretation of the term has even been reproduced in modern dictionaries,
such as the Tibetan-Tibetan-Chinese dictionary bl Zhang Yisun et al. (Bod
rgya tshig mdzod chen mo; Zanghan da cidian). In short, Powers, in his
8. Bod rgya tshig mdzod chen mo vol. 1, p. 404 grub bde rdzas gcig: gzhi
gcig gi steng gi ldog pa rigs mi mtshungs pa mams mnyam skye mnyam gnas
kyi skye gnas 'jig gsum dus mnyam zhing rdzas gcig pa yin pa. As for grub bde
gcig and grub bde (dbyer med) rdzas gcig, they are not quite the same notions,
as we see by the discussion in Yongs 'dzin bsDus grwa chung f. l6b. (See T.
Kelsang and S. Onoda, Textbooks of Se-ra Monastery [Kyoto: 1985] 8-9): kha
gcig na re / grub bde gcig yin na / grub bde rdzas gcig yin pas khyab zer na /
tsandan gyi kha dog dang / tsandan gyi dri gnyis chos can . .. The point is that
the color of sandalwood and the smell of sandalwood are grub bde gcig because
they are established simultaneously, abide simultaneously, and perish simulta-
neously (khyodgnyis grub pa dus mnyam / gnas pa dus mnyam / 'jig pa dus
mnyam pa'i phyir /). However, they are not grub bde rdzas gcig , simply
because they are not "of one substance" (rdzas gcig = ekadravya): they have
different "substances" (dravya) and are perceived as being separate things by
direct perceptions. Note that the Tibetan logical manuals speak of grub pa dus
mnyam ("simultaneously established") or skye ba dus mynam ("simultaneously
arising") more or less indifferently. Cf. Bod rgya tshig mdzod chen mo s.v.
grub bde dbyer med kyi rdzas gcig: chos gnyis po mnyam skye mnyam gnas kyi
skye gnas 'jig gsum dus mnyam zhing ...
Finally it should be remarked that the terms grub pa dang bde ba gcig pa
and (or something just like them, such as
figure in some Indian pramii1:ta texts without implying strict temporal identity
but rather just "same gain / utility." Dharmottarapradfpa (18,20-23 ed. Malvania)
cites the Sanskrit of a passage from Dharmottara's PramalJaviniScaya.tfka: yada
tu ( ... ) prathamenaiva ca vastu-
santana/; pravrttivi$ayfkartun;t niscayat sakyate, tadottare$iin;t tatsantanabhilvi-
niim pramii1:tyam apiisyata iti "Only the first moment
of a perception or inference can, because of niscaya ("ascertainment"), make the
continuum of the entity (vastusantana), which is causally efficacious
(arthakriyasamartha), an object of practical application. So, we reject the validity
160 JIABS 20.1
translation, followed that Tibetan interpretation where grub
bde gcig means grub pa (or skye ba) dus mnyam gnas pa dus mnyam 'jig
pa dus mnyam "established / arising simultaneously, abiding simultaneously
and perishing simultaneously." This interpretation of grub bde gcig =
became so widespread as to become almost good Tibetan
common sense. But it is, curiously enough, an un-Indian understanding, as
far I can see.
9
Let us move on to the key chapter on the three nature theory and the
application of this theory to the controversial statements of the Prajfiiipiirami-
(priimii1J.ya) of the subsequeqt [moments] which occur in the continuum of this
[first moment] without there being any difference in gain / utility I profit (abhin-
(The Skt. text is given on p. 35 of E. Steinkellner and H.
Krasser, Dharmottaras Exkurs zur Definition guZtiger Erkenntnis im Pramli1J.avi-
niscaya [Vienna: 1989]; see their translation p. 79; see also H. Krassers Dhar-
mottaras kurze Untersuchungder GuZtigkeit einer Erkenntnis, Laghuprlimli1J.ya-
[Vienna: 1991] 46, n. 69.) Steinkellner and Krasser translate the term
in this context as ceder Nutzen." This is possible for and
makes good sense in the philosophical argument which Dharmottara is developing,
which is grosso modo the problem of grhrtagraha1J.a ("an understanding of
something already understood") not being a pramii1J.a. Such a grhrtagraha1J.a is
not valid (pramli1J.a) because it brings nothing new: there is nothing different
which it "gains," from what is "gained" by the first moment of the cognition. For
our purposes, what is noteworthy is that it would seem that two things can be
I without being rigorously simultaneous, for
a pramli1J.a and l!- subsequent cognition are certainly not simultaneous.
9. Let us speculate a little bit on what happened to this term, which seems to
have come from Sanskrit and then taken a different meaning amongst bsDus
pa-inspired writers. It is clear that grub bde equals yogak$ema and that the
Tibetans took inis a conjunctive compound (grub pa dang bde ba). If we take
the individual members of the dvandva compound, it could be understood as
something along the lines of yoga = "advantage"; "acquisition" and k$ema =
"continuity of the status quo"; "abiding at ease." Tibetans however seem to have
taken grub pa as being like its more usual equivalent siddha "established," which
is not far from the idea of skye ba ("production"; "arisal"), but is certainly
different from yoga in the sense of "advantage" I "acquisition." They then took
bde ba as meaning simply "abiding" (sthiti = gnas pa), which is however not a
natural Tibet;m understanding at all, given that bde ba means "pleasure," "happiness"
to 99.9% of Tibetans. Finally, it should be stressed that Tibetans, in speaking of
grub pa I skye ba, gnas pa and 'jig pa are not providing just a paraphrase: this is
what they took the parts of the compound to mean. The strained character of this
interpretation is brought out by the fact that although the compound grub bde
(yogak$ema) has only two members, they were obliged to somehow add 'jig pa
(vyaya) to come up with the Abhidharmic trio of characters. .
TILLEMANS 161
tiisiitras about ni/:tsvabhiivatii, etc., in other words, the seventh chapter,
entitled "Questions of Paramfuthasamudgata" (henceforth "P"). We will
first focus on a passage which gives similes for
utpattini/:tsvabhiivatii and then goes on to describe the two sorts of para-
miirthani/:tsvabhiivatii, the lack of own-nature which pertains to the "perfect
nature" of dharmas. The passage is difficult in
Tibetan, especially because of a rather odd double use of gcig ("one"),
employing gcig . . . gcig in the sense of "one aspect / part . . . another
aspect / part." In fact, it is really only with the aid of the Chinese, which
uses -5} ... -5} ("one part ... one part"), that this use of gcig ... gcig
becomes clear, as Lamotte had pointed out in notes 7 and 8 (pp. 194-5) to
his translation. The Tibetan otherwise remains obscure. Frauwallner (p.
292) also understood the text in this way translating by "ein Teil ... ein
Teil." Powers, however, translated the ftrst gcig in an impossible way and
the second by taking a kind of face value reading. Neither of the two work
out right. Although I think that the last sentence of the passage is slightly
different in Tibetan and in Chinese, we can use Lamotte's and Frauwallner's
leads and Xuanzang to do a signiftcantly better translation than what Powers
did.
Tibetan (Powers p. 100.12-102.2; Lamotte p. 69 ):
don dam yang dag 'phags de la 'di ita ste dper na I nam mkha'i me tog ji ita
ba de Ita bur ni mtshan nyid ngo bo nyid med pa nyid Ita bar bya'o II don
dam yang dag 'phags de la'di lta ste dper na I sgyu ma byas paji Ita ba de
Ita bur ni skye ba ngo bo nyid med pa nyid kyang blta bar bya I don dam pa
ngo bo nyid med pa nyid de las gcig kyang blta bar bya'o II don dam yang
dag 'phags de la ' di Ita ste dper na I nam mkha'i gzugs kyi ngo bo nyid med
pa nyid tsam gyis rab tu phye ba dang I thams cad du song ba ji Ita ba de Ita
bur ni don dam pa ngo bo nyid med pa nyid las chos bda
S
med pas rab tu
phye ba dang thams cad du song ba gcig blta bar bya ste I. 1
Powers's translation (pp. 101-103):
P, for example, you should view lack of own-being in tenus of character as
being like a sky-flower. For example, P, you should also view the lack of
own-being in tenus of production as being like a magical apparition. [New
paragraph in Powers:] The ultimate lack of own-being should be viewed as
being something other than those {first two characters]. For example, P, just
10. Xuanzang's Chinese T. no. 676, f.694bl-6:

...
162 JIABS 20.1
as [space] is distinguished by being just the lack of ownbeing of forms in
space and as pervading everywhere, in the same way the ultimate lack of
own-being is distinguished by being the selflessness of phenomena and should
be viewed as all pervasive and unitary. (The italics are mine.)
Lamotte (pp. 194-195):
C'est a une fleur de I'air qu'il faut comparer I'Irrealite de caractere; a une
magie, I'Irrealite de naissance; de meme aussi l'Irrealite absolue sous un de ses
aspects. C'est a I'espace manifeste seulement par I'absence de matiere et omni-
present, qu'il faut comparer l'Irrealite absolue qui, sous un autre aspect, est
manifestee par la Non-substantialite des choses et omnipresente. (Our italics.)'
The first part of Lamotte's translation describing
and utpattini/:zsvabhiivatii involves some ellipsis, but it is accurate; Powers's
is too, more or less. Afterwards, Powers badly mistranslates the passage
concerning paramiirthanif:zvabhiivatii, because he did not understand don
dam ngo bo nyid med pa nyid de las gcig kyang blta bar bya (i. e. -:$}Jm
translating it wrongly as "The ultimate lack of
own-being should be viewed as being something other than those [first
two characters]." The real problem was that he translated de las gcig by
"being something other than those [fIrst two characters]." This is impossible:
no-one who understands Tibetan could understand de las gcig in this way,
nor for that matter could anyone understand the Chinese construction -:$}
... in that way either. The Tibetan literally says: "one also from that
paramiirthani/:zsvabhiivatii has to be regarded .... " And that means "one
also from [among the two types of] paramiirthani/:zsvabhiivatii has to be
regarded as being like [a magical apparition]," or less literally, "para-
miirthanif:zsvabhaviitii, in one of its aspects, is also to be likened [to a
magical apparition]." For the moment, suffice it to say that it is when the
compound paramiirthani/:zsvabhiiva(ta) is to be understood as referring to
paratantra that paramarthani/:zsvabhiiva is comparable to a magical appari-
tion. The two interpretations of the compound will be explained in more
detail below.
Now, let us take up the second aspect of paramarthani/:zsvabhavata
described in this passage. Again we have the construction don dam ngo bo
nyid med pa nyid las . .. gcig ita bar bya (i. e.
"paramarthani/:zsvabhiivatii, in one of its aspects, should be regarded .
. . ," the point being that here the sutra is talking about the other aspect of the
two-aspected paramiitthanif:zsvabhiivata, the sort which pertains to pari-
Powers translated this second occurrence of gcig (= -:$}) by
the word "unitary," which destroys any parallel with the earlier gcig. His
TILLEMANS 163
translation here is plainly no more than a guess.
Let us try to give what is at least a syntactically more accurate translation
of the whole passage: ''P, one should liken to flowers
in the sky; P, oIle should liken utpattinibsvabhavata to magical apparitions;
paramarthanibsvabhavata, in one of its aspects, is also to be likened [to a
magical apparition]; P, paramarthanibsvabhavata, in its other aspect, is to
be likened to space,11 which stands out by its mere lack of material nature
and which is present everywhere, because it [i. e., paramarthanibsvabhava-
ta] stands out as being the selflessness of dharmas (dharmanairatmya)
and because it [too] is present everywhere.,,12
Our translation of the last sentence using two ''because-clauses" is obviously
inspired by the Chinese 1llt:flG'ItiZ.FJTmii'i:)(ji--IJ]i'i:)(. Frauwallner also adopts
this solution. Remaining closer to the Tibetan ( ... ) chos bdag med pas rab
tu phye ba dang thams cad du song ba gcig blta bar bya, however, would
yield something like: "P, paramarthanibsvabhavata, in its other aspect,
which stands out by being the selflessness of dharmas and which is present
everywhere, is to be likened to space, which stands out by its mere lack of
material nature and which is present everywhere."
Later on in Chapter VII (Powers p. 131ff.), the SNS again takes up the
theme of two sorts of paramarthanibsvabhavata, once again using the
expressions de las gcig two times (Powers Tib. pp. 130.16 and 132.5) to
designate the respective sorts. The section was translated competently by
Lamotte on pp. 203-204, who had in his note 7 on p. 194 already referred
us to this passage as a development of the earlier theme of the two ways of
interpreting paramarthanibsvabhavata. Powers, on the other hand, missed
this fact. (He translated the two de las gcig here in a completely different
way from what we saw in the earlier passage discussed above. This time it
becomes "additionally.") Ironically, not only had Lamotte spoken of these
two sorts, but Louis de la Vallee Poussin, on p. 556 of La Siddhi de
Hiuan-Tsang (Paris, 1929), had translated a passage from SNS which tells
us about one type of paramarthanibsvabhavata, namely the type which
11. The Tibetan has the genitive nam mkha'i, which is somewhat odd. The
Chinese just has followed by the appositions
"which stands out (= Tib. rab tu phye ba, Skt. prabhavita) by the mere lack
of material nature and which is present everywhere (= Tib. thams cad du song
ba)."
12. Cf. Frauwallner op. cit. p. 292: "Und wie der Raum, def aus der blossen
Wesenlosigkeit (= dem Nichtvorhandsein) der Materie hervorgeht und sich iibe-
rallhin erstreckt, so ist ein Teil der Wesenlosigkeit der hOchsten Wahrheit nach
anzusehen, insofern sie aus der Ichlosigkeit der Gegebenheiten hervorgeht und
sich iiberallhin erstreckt."
1M JIABS 20.1
pertains to conditioned phenomena (paratantra)-his translation is substan-
tially in agreement with that of La.l1otte. 13 The passage is also translated in
the first paragraph on p. 101 of Powers. But, alas, Powers does not seem
to have adequately consulted his predecessors' work.
Briefly, the point, in fact, seems to be that there is one sort of para-
marthanil:tsvabhavata: which pertains to paratantra and another which
pertains to p a r i n i ~ p a n n a , depending upon how we understand the compound
paramarthanil:tsvabhavata. To take the first sort, the SNS tells us that it is
the Ultimate (paramartha) which is the "pure object" (mam par dag pa'i
dmigs pa. Cf. De la Vallee Poussin: vyavadanaZambana "l'objet de con-
naissance qui comporte purification"; Lamotte "l'Objet pur"), but that condi-'
tioned things (paratantra) are not themselves this pure object and thus lack
this pure nature. Hence conditioned things are paramarthaniT:tsvabhava in
that they lack the own-nature which is ultimate: on this interpretation we
thus have a "lack of ultimate own-nature." The other interpretation of the
compound is to say that p a r i n i ~ p a n n a is the lack of conceptual and linguisti-
cally imagined (parikalpita) natures and that p a r i n i ~ p a n n a is ultimate: it is
thus an "ultimate lack of own-nature." Powers's presentation of this double
aspected paramarthaniT:tsvabhavata does not come clear. In fact, there is
no evidence in the translation or in the notes that Powers was aware that the
SNS spoke of two sorts of paramarthaniT:tsvabhavata at all.
It is time to conclude. I have up to now, as is obvious, focused exclusively
on problems in Powers's translation. We should stress that much of the
text of the SNS is more or less correctly rendered into readable English by
Powers and that the Anglophone reader will thus have access to the SNS
(although he should exercise caution and healthy skepticism.) On the
whole, however, Powers's work is not a step forward from that of Lamotte.
It lacks sufficient accuracy, rigour and philological analysis. Simply trans-
lating commentarial passages, as Powers often does in his notes, does not
replace penetrating analysis of philological or philosophical problems. In
sum, someone should do a better job on this important text.
13. Siddhi vol. II, p. 556: "Ce qui, dans les Dharmas, est Ie vyavadanalambana
(l'objet de connaissance qui comporte purification), je declare que c' est paramartha.
Or Ie paratantraZak$alJa n'est pas cet aZambana; done il est paramarthanil:tsva-
bhavata." Cf. Frauwallner op. cit. p. 291.
JOHN R. MCRAE
Zenbase CD 1. Computer software produced under the
direction of Drs App. Kyoto: International Research
Institute for Zen Buddhism, 1995.
There have been a number of projects undertaken in the past few years for
the computer input of Buddhist texts, but the most commendable of these is
that undertaken by the International Research Institute for Zen Buddhism
(hereafter IRIZ), under the leadership of Associate Director Urs App, in
Kyoto, Japan. Following the intellectual guidance of Professor Yanagida
Seizan, the founder of IRIZ, App has published a very informative journal
in English and Japanese, whimsically entitled The Electronic Bodhidhanna,
explaining various issues of computer text input in Zen studies. Now he
has published ZenBase CD1, which includes the best of The Electronic
Bodhidhanna, a large number of Zen texts in electronic form, plus a hand-
full of extremely useful tools for electronic text manipulation. The contents
of ZenBase CD1, plus several updated fIles, are available at the IRIZ World
Wide Web site (http://www.iijnet.or.jp/iriz/irizhtml/irizhome.htm); in
addition, this review and certain additional tools and information are avail-
able from my web site at Indiana University. (The online version of this
review includes all the Internet references as active hyperlinks. See
http://www.easc.indiana.eduljmcrae/ZB/ZBreview.htrnl). Since the IRIZ
web site has been updated somewhat since the publication of ZenBase CD1
(but most recently only in late May 1996, ten months prior to this writing),
users are encouraged to explore it and the other sites listed here for more
up-to-date information and tools.
Although there are now a number of online sites that provide access to
electronic Buddhist texts (see in particular the web site of the Center for
Buddhist Studies at National Taiwan University and directed by Prof. Ven.
Shih Heng-ching, http://ccbs.ntu.edu.tw/, and the other sites cross-listed
there), Zenbase CD] is a remarkable and even unprecedented contribution.
This is because of the comprehensive and systematic approach taken by
App and his colleagues in including not only a major collection of Chan /
Zen texts, but also a very significant collection of tools for the manipulation
165
166 JIABS 20.1
of those texts in ordinary users' environments, as well as research utilities
to assist those engaged in Zen studies. For example, there are tools that
allow the reformatting and conversion of text files from Japanese t9 Chi-
nese encodings; a set of alphanumeric character fonts that allow both Win-
dows and Macintosh users to display and print all the diacritically marked
characters used in Buddhist studies; and Christian Wittern's KanjiBase,
which provides a systematic method for the specification of characters not
found in the most widely used codesets. (The alphanumeric fonts, known
as Appeal, should be adopted by Asianists as a convention for the
representation of diacritic ally marked text. KanjiBase is the best interim
solution until a modification of Unicode now under consideration is'
adopted.) The research utilities include a set of bibliographies of English
and other writings on Zen studies; a set of Macintosh dictionaries for Zen-
related character input and information lookup (presented in a variety of
formats for different input methods); and a number of Macintosh
dictionaries and Zen lineage diagrams prepared for use with a specific
commercial dictionary utility, TSM Passport, for which a demo version is
also included. Not all of these tools work flawlessly (even with the updates
available on the IRIZ web site), and the documentation provided is
frequently insufficient, but their overall value is remarkable.
In addition to these various resources, on which I will comment in spe-
cific below, a major reason why ZenBase CD1 is so important is its under-
lying design philosophy. Three major principles are involved in this design
philosophy: (1) the use of a master / distribution tape metaphor for the
preparation of electronic texts; (2) the adoption of conventions for tagging
texts deVeloped by and in conjunction with the Text Encoding Initiative
(TEl) and the Electronic Buddhist Text Initiative (EBTI); and (3) the use of
HyperText Markup Language (HTML) for the simultaneous navigation of
both ZenBase'CDl and the IRIZ web site.
First, let us look at the master / distribution tape metaphor. In articles
previously presented in The Electronic Bodhidharma, App has explained
the metaphor that governs his approach to text archival, which is that of
"master" and "distribution" tapes used in the music recording industry. That
is, where a commercially available tape of, say, Carl Orffs "Carmina
Burana" or the Jerry Garcia Band would contain two sound tracks for
reproduction on ordinary home stereo equipment, such a distribution tape
would be mixed from a master tape having 17 or 20 tracks or more. Simi-
1arly' whereas individual computer users may only have the capacity to read
characters found in either the BigS code (used for traditional characters in
Taiwan and elsewhere) or the JIS code (used in Japan), this will probably
MCRAE 167
not be sufficient to represent all of the characters used in a given Zen or
other Chinese Buddhist text from centuries past. Therefore, the producer of
the master text should take pains to include as much information as possible
about the original printed text, including the precise specification of charac-
ter variants, so that as users' technical capabilities improve (as they
inevitably will, in part through the efforts of IRIZ and other such institu-
tions) they can use more and more accurate distribution texts. In addition,
the master version of any electronic text, rather than the less accurate but
certainly usable distribution versions, should be that used for the creation of
more complex and sophisticated database tools, such as are certainly going
to appear in the near future.
Second, ZenBase CD] is the first Buddhist input project to adopt the
conventions of tagging developed in conjunction with the Text Encoding
Initiative (TEl) and the Electronic Buddhist Text Initiative (EBTI). Simply
put, these conventions involve the inclusion in each electronic text of (a) a
header explaining the source of the text, the people responsible for its cre-
ation, the standards of accuracy achieved, the specification of the codeset
and additional characters used, and defInition of tags used, etc. (part of this
header is known in TEl terminology as a Document Type Definition
[DTD]); and (b) tagging of specific items within the text using Standard
Generalized Markup Language (SGML) in ways conventionalized by TEl
and EBTI. Since there are obvious advantages to having information about
the provenance of any electronic text attached to the text itself, the header
may be included even if there is no SGML tagging applied. In keeping with
this rule, every electronic text provided in ZenBase CD] includes such a
header, but due to the fact that the conventions of TEl / EBTI tagging for
Buddhist texts are still in a very early developmental phase, only one text
on the CD has such tagging. In fact, there are three categories of texts
included on the CD, one that includes a first attempt at TEl / EBTI tagging
(represented by only one very substantial text), one that includes texts
which have been thoroughly proofread and which are believed to be
extremely accurate (represented by some 16 texts), and a third that includes
texts which have been input and proofread only to a level of moderate
reliability (represented by some 58 texts).
Third, the explanatory and instructional files on both ZenBase CD] and
the lRIZ Web site are all written using HTML, so that they may be navi-
gated using any of the many Web browsers available today, such as
Netscape Navigator, Mosaic, or (for those without graphical interfaces)
Lynx. In fact, all these files are available in two versions, one in English
and the other in Japanese; it would certainly be appreciated if IRIZ would
168' nABS 20.1
make available Chinese-language versions.ofthese same files at some point
in the future. (I have'prepareda Chinese translation of App's article on
preparing electronic editions; see http://www .easc.indiana.edu! -jmcrae/ZB
/ App_msinput.Big5.html.) Using recent versions of Netscape Navigator
or Mosaic, the user needs simply to have Japanese capabilities available on
his computer and to select the appropriate Japanese font. (In Netscape 3.0
for the Mac, for example, this is done by pulling down the "Options" menu,
selecting "General Preferences" and then ''Fonts,'' and choosing a suitable
Japanese font. Other Web browsers will have procedures that are concep-
tually similar.)
ZenBase CD1 even includes both Netscape LIN for the Macintosh and'
Mosaic for Windows. These versions are now long since out of date, and
one suggestion is in order for users wanting to install Mosaic for Windows
from the CD: my attempts to follow the installation procedure on three sep-
arate PC clones (two no-names and a Compaq) were unsuccessful until I
realized that I had to copy the files from ZenBase CD 1 to a temporary file
on the PC and run the installation procedure from there. In any case, those
with Internet access should download a more recent version of Netscape
Navigator via ftp from ftp.netscape.com, and those without Internet access
should be able to find a version of either Netscape or Mosaic (for either
PC, Macintosh, or Unix machines) at a computer store or bookstore, or
through a mail order house. In any case, the problems just mentioned are
indicative of the one shortcoming of ZenBase CD1, which is a lack of accu-
rate, meaningful, and/or sufficient instructions for many of the tools
provided.
There are three categories of tools found on ZenBase CD1: (1) font utili-
ties, especially the Appeal font, for diacritically marked characters, and
KanjiBase, which provides a means by which to handle characters not in
the Big5 character set; (2) tools specifically relevant to electronic text cre-
ation, most of which utilize PERL or MacPERL; and (3) other useful tools
not necessarily directly related to electronic texts per se.
1. I am placing Appeal and KanjiBase together because, although quite dif-
ferent from each other, they both resolve difficulties involving fonts.
Appeal is a font (represented in both Mac and Windows versions) that
should be used as a standard convention for alphanumeric text by everyone
in Buddhist studies. It is based originally on the fonts created by K. R
Norman for use by Indologists, but it has been substantially improved by
people at the Hob6girin (where it has been used for the preparation of
recent editions of the Cahiers d Extrfme-Orient) and IRIZ. Those who ac-
MCRAE 169
cess my Web site will find that I follow my own advice: users wishing to
see diacritically marked text displayed correctly are instructed to download
and install Appeal on their systems. The differences between Appeal and
the Norman fonts are too technical to be discussed here; see http://www.
easc.indiana.edu! -jmcraelfonts/Norman+Appeal.html for a discussion and
references. Although font loyalties tend to approach the strength of reli-
gious conviction, for Buddhist studies Appeal has the following advan-
tages: it works better with a greater number of Asian and European lan-
guages, it displays better in both roman and italic styles, and it is technically
better suited to work on a variety of platforms. Fortunately, for the great
majority of users who have to convert files from one font to the other, only
one annoyance occurs: the need to do a search-and-replace for 11 (long u,
lower case).
The KanjiBase system, developed by Christian Wittem, is currently
available only for Windows users, although I have used it quite success-
fully on the Mac under SoftWindows. Although it will probably be super-
seded in a year or so by a modification to the Unicode standard recently
proposed by Prof. Hsieh ChingChun of the Institute of Information Science
at the Academia Sinica in Taiwan, this is an important innovation that
should be adopted as at least an interim solution by every project for the
input of Chinese texts, Buddhist or non-Buddhist. What KanjiBase does is
provide a sensible and reliable manner for the treatment of characters not
found in the Big5 character set. Now, Big5 includes about 13,000 charac-
ters, which is generally adequate for ordinary written Chinese communica-
tion but not sufficient for the representation of classical Chinese texts.
Although the number of characters in any given text that do not occur in the
Big5 code is small (probably no more than 3-5% of the total character rep-
resentation of any text, and often less than 1 %, according to statistics
compiled at IRIZ), the requirements of accuracy demand that these charac-
ters be represented in some fashion.
Most Chinese software systems allow the user to create and specify a
small number of extra characters, referred to as waizi in Chinese or gaiji in
Japanese. However, since each text input project has defined its own set of
waizi / gaiji, it is extremely difficult if not impossible to exchange data
between different computers. In theory, conversion tables could be con-
structed, but they would certainly be "lossy" (since sets of extra characters
created by different users would never correspond to each other exactly),
and in practice this is hardly ever done. Also in theory, it might be possible
to use an entirely different code from Big5 as a master code, such as the
CCCIl code used by libraries in Taiwan, the closely related REACC code
170 JIABS 20.1
used by the U.S. Library of Congress and comprehensive library databases
in the U.S., or the recently developed but still underused Unicode. For dif-
ferent technical reasons explained by Wittem in an article printed in The
Electronic Bodhidharma and included in ZenBase CD1 (see The Eledtronic
Bodhidharma, no. 4 (1995), or the file CEFINTRO.HTM on the CD or
IRIZ web site), none of these codes is acceptable.
KanjiBase resolves this problem by the combined use of the Big5 code
(actually, the most commonly used variant of Big5, that adopted by the
ETen operating system) and the CNS (for Chinese National Standard)
code. Although CNS is officially promulgated by the government of the
Republic of China, it has never been adopted by any Chinese software or
system creator. However, it has the advantage of specifying a codeset of
some 48,000 characters in a fairly rational fashion, with relatively little
overlap and omission. Wittem has designated ETen Big5 as the fIrst "plane"
of his system, with five CNS planes and three additional planes (to allow
for future growth) residing on top of this. (As a computer standard, CNS
involves features other than codepoint definitions which ae irrelevant.) In
order to find a character variant, one enters either the radical number and
stroke count or a four-comer code, then uses arrows to select the specific
character desired. When a character is selected, the program returns a larger
bitmapped image and other information, which includes some but not nec-
essarily all of the following: Chinese pronunciation in Pinyin, Japanese
pronunciation, English meaning, and numbers in Morohashi's Dai Kan- Wa
jiten, BigS, Unicode, and CCCIL Macros to allow the bitrnapped image to
be introduced into a Microsoft Word 6.0 file were promised on Zenbase
CD1, but are actually only available online from Wittem's own web site.
Also available there are updated Big5<->SJis conversion utilities and an
improved Windows version of the Appeal font; for all of these see
http://www.gwdg.de/%7Ecwitter/bugfIx.htm. Most promising, however,
is that this site features an online implementation of KanjiBase, although
unfortunately at this point it is still of limited availability. (I have not yet
been able to access it.)
It is unfortunate that KanjiBase is not available yet for the Macintosh
platform, but it is possible to use it on a Mac under SoftWindows. I have
successfully run KanjiBase under SoftWindows on a PowerBook 280c,
but there are a couple of provisos to be made. First, the program should be
copied from the CD while operating under SoftWindows; I fIrst copied it to
a directory accessible by SoftWindows using the Macintosh Finder, and it
just didn't see some of the most important files. (It seems that they are
experienced by the MacOS as hidden files, a technical oddity of ZenBase
MCRAE 171
CD]\) Second, make sure to copy the program to a hard disk file named D:,
since it just won't run from disk drive C: This last piece of advice should
apply to those running under native Windows environments as well.
As this review goes to press I have just learned of the proposal for a
modification to Unicode by Prof. Hsieh, already mentioned above. The
recommendation is that characters not in the Uni-Han area of Unicode (that
reserved for CJK characters) should be represented by a combination of
codes representing the individual elements used to write the characters and
combined in the order the character is written. In this fashion Unicode
would be spared the necessity of designating specific codes for the tens of
thousands of characters (estimated in the 160,000-200,000 range) that are
used only very rarely (less than 0.1 % of the time, on average), but at the
same time a unique code could be generated for each such character shape.
This proposal is now under consideration by the Unicode Consortium, but
until it is adopted KanjiBase can provide a very workable (and eventually
easily upgradeable) solution. And, if KanjiBase is in fact superseded, one-
for-one correspondences between the characters it covers will allow for that
Valhalla of computer technology, an easy upgrade path.
Also included in ZenBase CD] is an implementation of Werner
Lemberg's CJK TeX, a platform-independent implementation that allows
the typesetting of documents that can include Chinese, Korean and
Japanese text. This implementation of CJK TeX includes support for CNS
via the KanjiBase code references. This may be wonderful news to some
people, but fm a FrameMaker kind of person: I have never taken the time
to learn how to use TeX, or even LaTeX, which are dearly beloved of true
computerphiles but just too far from WYSIWYG (what-you-see-is-what-
you-get) for me. (Just as, prior to moving to Cambridge I had a nightmare
in which I actually liked the Celtics, I dread the day someone convinces me
to take the plunge into TeX). Hence, I cannot comment on this system.
2. The tools most closely related to electronic text creation found on Zen-
Base CD] include a number that utilize PERL, and there are instructions for
installing PERL (or MacPERL) on either a PC or Macintosh computer.
(Mac users should note that the "pyramid" shaped icon referred to in the
instructions does not occur in the CD versions of the Mac environment,
where it appears-for reasons only the original program authors under-
stand, no doubt-as a camel on wheels. This has been changed in the
updated tools to include both camel and pyramid.) These PERL-based tools
include the following:
172 1IABS 20.1
A. A concordance maker, which "creates a complete concordance from a
text file (1IS or Big-5)" which in conjunction with related Word 6 macros
can be used to print "a perfectly formatted and paged concordance."
There were substantial problems in this tool as published on the CD, and
a newer version has been installed on the IRlZ web site. I have not tested
this utility.
B:Kanji (Chinese character) code conversion tools, which can be used to
convert text files from Big5 to 1IS or vice-versa. Code conversion is a
very inexact process due to the differences between codes (i. e., one code
may have nothing to correspond to a certain character in another code, or
it may have more than one character). As a result, ZenBase CD1 provides'
different tools, with three degrees of strictness of conversion.
Curiously, though, ZenBase CD1 provides tools only for conversion
between Big5 and 1IS codes. Conversion utilities that handle additional
codes (most significantly, the Guo Biao or GB code used in the People's
Republic of China) are easily available and could easily have been
included. The interested PC or Mac user can use ftp or Fetch, etc., to
access ftp.ifcss.org:/software/ or its mirror site ftp.cuhk.hk:/pub/
chineselifcss/software (or use a web browser to get to http://www.ifcss.
org/ software!) for Sinocode.zip (for the PC) or Hanzi Converter (HC
1.5 or HC 3.0, for the Mac; I actually find HC 1.5 to be more con-
venient). Also check Carlos McEvilly's Chinese Language Homepage
(http://www.netcom.comlbamboo/ chinese) for his Bamboo Helper and
a great deal more. I have worked with both Sinocode and Hanzi Con-
verter to convert files written using the WPS program so widely used in
the PRC to Big5 files; both programs need the source file to be cleaned of
extraneous format codes. Recently, rve found that for my purposes the
import capabilities of the shareware text editor Unicorn (available through
ifcss.org) are more convenient than any of these stand-alone GB <->Big5
conversion utilities, and Unicorn is convenient in other ways as well.
C. Kanji normalization tool (Big5 texts only), which "normalizes a text
file for more convenient use in ordinary text search or in concordances;
the program also generates a record of all conversions that were effected."
I have not used this tool, but the problem it addresses is a real one. Since
the Big5 code contains variant glyphs of some characters at different
codepoints, this program should convert the variants into the codepoints
most commonly used.
D. Text format conversion tool, which "converts the format of text files
among the RAW, APP, and TAB formats." I have not tested this tool; the
MCRAE 173
three formats are those used for different purposes within the IRIZ text
archives.
E. IRIZ text statistics tools, which can be used "to count characters and
compounds and to generate gaiji information lists."
F. The search utility IRIZGREP, for performing text searches within a
file or set of files, for which a new version is offered on tht:; IRIZ web
site. This works very nicely indeed, and is perhaps the tool users will
find most useful.
3. There are a number of tools included on ZenBase CD] which, although
not directly related to electronic text creation, are nevertheless very useful.
These include:
A. Versions of Netscape LIN and Mosaic 2.0 for the Mac. See the
discussion above regarding installation of these or more up-to-date
versions.
B. Three Mac system utilities, FontPatchin', SearchFiles 1.3, and
UltraFind 1.0.3. There is no explanation of how the people at IRIZ find
FontPatchin' useful (documentation, or at least a brief statement, would
have been helpful), but Charles Wivell, of the University of Rochester
(retired), has told me that it is useful for including Chinese and Japanese
characters in electronic mail. However, Wivell warns that the
FontPatchin' control panel causes some problems with ordinary computer
operations, so that he disables it by putting it into a "Control Panels
(disabled)" folder when it is not needed. SearchFiles (this is the most
recent version) allows one to search all files in a given directory for the
presence of specified text strings. UltraFind is described as allowing one
to find specific files or groups of related files anywhere on one's com-
puter or network and move or perform other operations on them; version
2.2 is now available (www.ultradesign.com/ultrafind). Both programs
are shareware. In my own experience, SearchFiles has worked nicely
(even finding Chinese text, although without appropriate character dis-
play), whereas UltraFind has not. That is, given the same search task,
SearchFiles returned results and UltraFind did not.
e. BBEdit Lite 2.3.1, a Mac editor intended for use in writing HTML text
(version 3.5.1 is now available on the Internet, and ver. 4.x is available
by purchase), and SedMac 1.0. I have not used these.
D. A demo version of TSM Passport, a Japanese dictionary utility for the
Mac. After toying with the demo (which is almost thoroughly crippled), I
purchased the complete program at a store in the Akihabara electronics
174 JIABS 20.1
mart section of Tokyo. ZenBase CD1 provides dictionaries for use with
TSM Passport that allow the user to access an extensive bodies of
information about Chinese and Japanese Zen. Although 1 have had some
difficulties with this program (I have almost got it running correctly on a
PowerBook 280c, but it crashes when used on the PowerMac at home),
but if you can take the time to get it to work it is highly recommended. If
you don't have someone in Japan to purchase a copy for you at an
Akihabara discount price, you can order the program from the developer:
Mercury Software Japan, Hassei Bldg., 2F; 20 Rengezo-ch5, Sh5goin;
Sakyo-ku, Kyoto; Japan 606; (075) 751-0205; fax: 751-0206. The
program is listed on the company's web site, but with no information'
about the current version number, and no order or demo download
options. (See http://www.iijnet.or.jp/mercury /WORPRO.html.)
JOE BRANSFORD WILSON
The International Association of Buddhist Studies
and the World Wide Web
There is some risk in writing about the World Wide Web while off-line-
that is, while not "on" the Internet. The web presents us with such a con-
stantly changing terrain that words written in the winter for publication in
the summer may well point the reader to information that either no longer
exists, has moved to a new location (with no forwarding address), or is
uselessly outdated. With that caveat, there are a few programmatic notes
that I feel ought to be made about the World Wide Web and the direction in
which it careens down the "information superhighway."
It is hardly possible to open a magazine, listen to public radio or watch
television in the United States without learning the World Wide Web
address of the magazine (or television network), its SUbscription or public
relations department, the author of the article one is reading or hearing, or -
especially - the World Wide Web address of the advertisers (e. g.,
www.toyota.com). If this is not already the case for all of our members in
Asia and Europe, rest assured that it will be soon.
As is well known, presence on the Internet had its beginning in the aca-
demic world-at first mainly among those who deal with numerical data,
then among those who work with texts (especially collections oftexts), but
in more recent years virtually all academic disciplines have found some
aspect of networked information valuable. (Jamie Hubbard has provided
us with a well researched overview of the development of Buddhist Studies
on computers and the Internet in JIABS 18.2, "Upping the Ante: budstud@
millenium.end.edu." In light of his title, I suppose the present essay could
be renamed ''http://www.millenium-end.edu/budstud/ .")
As of this writing there are a vast number of web sites that present
information on Buddhism-most of them created either by interested ama-
teurs or technologically literate Dharma Centers outside of traditionally
Buddhist countries. The latter are a gold mine for those who track western
Buddhism, the former a potential trap for the unwary web surfer.
175
176 nABS 20.1
The International Association of Buddhist Studies has had a fledgling
presence on the world wide web (now at http://www.uncwil.eduliabs/)
for several years, but at the end of 1996 began a more extensive involve-
ment when it took on responsibility for the Buddhist Studies Web Sites
directory of one of the more venerable web institutions, the World Wide
Web Virtual Library-created and up to then maintained by T. Matthew
Ciolek of the Australian National University.
While the lABS web site is currently administered by its Treasurer, we
hope to include input (including linked web sites) from members in their
own areas of specialization. Indeed, a number of our members are already
deeply involved in the Internet and the World Wide Web. In addition to
Matthew Ciolek's huge cluster of pages with pointers to web sites on Asia
(the Asian Studies World Wide Web Virtual Library at http://coombs.anu.
edu.aulWWWVL-AsianStudies.html) and the on-line Journal of Buddhist
Ethics (http://www.cac.psu.edu/jbe/jbe.html [see below]), a number of
individual scholars have developed extensive web sites. Some that come to
mind are Charles Muller's work on Korean Buddhism (http://www2.gol.
comlusers/acmuller/Buddhism-Korean.html) and on graduate schools of
Buddhist and Asian Studies (http://www2.gol.comlusers/ acmuller/
GradStudies.htm), Johannes Tuemmer's Buddhistische Studien-Buddhist
Studies site (http://www.uni-koeln.de/phil-faklindologie/buddha.htm).
John McRae's peripatetic Chinese Buddhism site (see the IABS page, http:
//www.uncwil.edu/iabs/vll, for its address), and my own Mind-Only-
Cafe (http://www .uncwi1.eduipeople/wilsonjiMind-Only-Cafe.html). I
apologize for any omissions here; the list is necessarily not up to date and,
at any rate, is intended merely as a set of examples. If its brevity serves as
an incitement to readers to make us aware of their own web sites, please
send e-mail to with descriptive information.
It seems, however, that whatever else a web site sets out to do,it always
includes an index of other related web sites. Each of the web sites men-
tioned above has some content-in words or in images-but all seem to
descend from the original concept of having a home page that gives some-
one a web presence and then lists his favorite web sites. From the begin-
ning. of widespread use of the Internet, a central index or registry has been a
desideratum (the World Wide Web Virtual Library was one such attempt),
but this has been a hope never successfully fulfilled.
I mentioned earlier that much of the information on Buddhism that may
be found on the web is created either by Buddhists (or by those interested
in Buddhism) or by Dharma Centers located outside of traditionally Bud-
dhist countries. This raises a number of problems for the academic Bud-
WILSON 177
dhologist. First, is there anything useful out there useful for our own
research-apart from access to library catalogs and on-line texts?
(Regarding on-line texts, see, for example, the site of the "Group in Bud-
dhist Studies at the University of California at Berkeley and the Berkeley
Buddhist Research Center" at http://garnet.berkeley.edulyaoming/ .) Sec-
ond, assuming, naturally, that we can ourselves distinguish objective
description from proselytization, what can we tell our undergraduate stu-
dents before we send them into the wilds of the Internet to do research?
(The issue of copyright and the ownership of information is also a problem,
but one that is outside the scope of this essay. The interested reader will
fmd, however, that a cottage industry of position papers, studies, editorials
and on-line debates now flourishes, some seeking to clarify the issue and
set up guidelines, others arguing forcefully that no solution is to be found
and copyright is dead.)
The answer to the first question-is there anyt:h.ip.g useful out there-will
be found in our own willingness to introduce content to the web. One of
the earliest and most successful examples of this sort of initiative is the
Journal of Buddhist Ethics (http://www.cac.psu.edu/jbe/jbe.html), a peer
reviewed journal that is published only in digital format: it can be read on a
computer screen or downloaded in preformatted form for the individual to
print out as she desires. We are also beginning to see on-line syllabi for
courses on Buddhism, although there is no true index of these.
On-line prepublication of papers for comment and criticism by peers
might be one direction in which to go, and the lABS could perhaps provide
a forum-distinct from the JIABS-wherein that could take place. (The
Journal of Buddhist Ethics has done something like this in its on-line
forums.) This requires a review process, however-and that is what the
World Wide Web seems to lack.
We need not only links to pages on Buddhism and Buddhist Studies
but-annotating those links-we need review, by qualified reviewers, of
their content. This would be an aid not only to all scholars, whether Bud-
dhologists or not, but especially to our students. I invite lABS members
who do spend time retrieving networked digitized information (i. e., surfmg
the net) to consider reviewing the sites you find useful (or time-wasting)
and helping us update our indices. (Send e-mail to wilsonj@uncwil.edu).
The International Assocation of Buddhist Studies
David Seyfort Ruegg
President
Oskar v. Hiniiber
General Secretary
Joe. B. Wilson
Treasurer
Regional Secretaries: Janet Gyatso (Americas),
S. Katsura (Asia), Oskar v. Hiniiber (Europe) (temporary)
Members of the Board: Robert Buswell, H. Durt, R. Gupta, K.
Kimura, E. Steinkeller, T. Tillemans, Akira Yuyama
The International Association of Buddhist Studies, founded in 1976, is
devoted to promoting and supporting scholarship in Buddhist Studies
in all its aspects, past and present, around the world. Membership is
open to scholars of all academic disciplines.
Membership dues are: $40 for full members, $20 for student members,
$1000 for life members. Dues may be paid by personal check (US
only), Visa, or MasterCard. Prospective members from developing
countries may contact the Treasurer concerning subsidized memberc
ship rates. Dues are payable per calendar year by December 31 of the
previous year. Payments may be sent in US dollars to Professor Joe B.
Wilson, Department of Philosophy and Religion, University of North
Carolina at Wilmington, Wilmington NC 28403 USA.
Email: wilsonj @uncwil.edu
Fax: 910-962-7107
The lABS world wide web site is located at
http://www2.uncwil.edu/iabs