Journal of the International Association of

Buddhist Studies
Volume 23 Number 2 2000
CARMEN DRAGONETTI
Marginal Note on the Idealistic Conception
of citta-miitra
JOHN KIESCHNICK
Blood Writing in Chinese Buddhism
KLAUS-DIETER MATHES
Taranatha's Presentation of trisvabhiiva
in the gZan ston sfiin po
SARA McCLINTOCK
Knowing All through Knowing One:
Mystical Communion or Logical Trick
165
177
195
in the Tattvasarrzgraha and Tattvasarrzgrahapafijikii 225
LINDA PENKOWER
In the Beginning ... Guanding 11m (561-632)
and the Creation of Early Tiantai 245
PETER SKILLING
Vasubandhu and the Vyiikhyiiyukti Literature
The XIllth Conference of the International
Association of Buddhist Studies,
Bangkok 8-13 December 2002
First circular
297
351
e watermark
The Journal of the International
Association of Buddhist Studies
(ISSN 0193-600XX) 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 Lausanne,
BFSH 2, CH-1015 Lausanne,
Switzerland.
Address subscription orders and
dues, changes of address, and
business correspondence
(including advertising orders) to:
Professor Cristina Scherrer-Schaub,
Treasurer lABS,
Section de langues et civilisations
orientales, Faculte des lettres
Universite de Lausanne, BFSH 2
1015 Lausanne-Dorigny
Switzerland
email:
iabs.treasurer@orient.unil.ch
Fax: +41 21 6923045
Subscriptions to JIABS are $40
per year for individuals and $70
per year for libraries and other
institutions. For information on
membership in lABS, see back
cover.
Copyright 2000 by the International
Association of Buddhist Studies, Inc.
EDITORIAL BOARD
Cristina A. Scherrer-Schaub
Tom J.F. Tillemans
Editors-in-Chief
Robert Buswell
Steven Collins
Collett Cox
Luis O. Gomez
Paul Harrison
Oskar von Hinuber
Roger Jackson
Padmanabh S. Jaini
Shoryu Katsura
Donald S. Lopez, Jr.
Alexander Macdonald
D. Seyfort Ruegg
Robert Sharf
Ernst Steinkellner
Erik Zurcher
Editorial Assistant:
Yves Ramseier
JIABS 23.2
Contributors to this issue:
Carmen DRAGONETTI is "Investigador Superior" (Superior Investi-
gator) in Indian Philosophy and Buddhism at The National Council of
Scientific Research (CONICET), Argentina, since 1966, and Professor
of Buddhism in The Institute of Buddhist Studies Foundatipn, Buenos
Aires since 1989. Her last research focuses on many aspects of the Lotus
Satra. Together with Fernando Tola she has translated from Sanskrit
into Spanish the Lotus Satra (Mexico: El Colegio de Mexico 1999). Her
most recent publication has been the translation from Chinese into
Spanish of the Satra of the Infinite Meanings.
John KIESCHNICK is an Assistant Research Fellow at the Institute of
History and Philology, Academia Sinica in Taipei. Author of The
Eminent Monk: Buddhist Ideals in Medieval Chinese Hagiography
(University of Hawaii Press 1997), he is currently writing a book on the
impact of Buddhism on Chinese material culture.
Klaus-Dieter MATHES has been in charge of the Nepal Research Centre
and the Nepal-German Manuscript Preservation Project in Kathmandu
for the last seven years. After his return to Germany in spring 2001, he
plans to complete his "Habilitation" at Hamburg University with a study
of the tathagatagarbha theory based on newly found Tibetan material,
including the Ratnagotravibhaga commentary by , Gos Lo tsa ba gZon
nu dpal. Main fields of interest: Yogacara, Madhyamaka and tathagata-
garbha in the light ofIndo-Tibetan hermeneutics.
Sara McCLINTOCK is a doctoral candidate in Religion at Harvard
University. Her dissertation is on the topic of omniscience in the Tattva-
saf!!graha and Tattvasaf!!grahapafijikii.
LINDA PENKOWER is Associate Professor of Religious Studies at the
University of Pittsburgh. Her Tientai Buddhism and the Construction of
Lineage during the Tang will be published by the Kuroda Institute
(UHP).
PETER SKILLING is in charge of the Fragile Palm Leaves for the
preservation of Buddhist Literature (Bangkok). Among his various
contributions to Buddhist studies, one may mention Mahiisatras: Great
Discourses of the Buddha. Vol. I and II, Oxford: Pali Text Society 1994
and 1997.
CARMEN DRAGONETTI
Marginal Note on the Idealistic Conception
of citta-matra
The Yogacara system is composed of a great number of theories. In
order to establish the way it was formed, it would be necessary to study
when, where and how each one of these theories originated, and also to
study when, where and how these diverse theories where assembled
giving rise to a new structure of philosophical thought. (The same thing
would have to be done mutatis mutandis in regard to the origin of the
Mahayana). The creation of the Yogacara theories (as of those of the
Mahayana) has been a dynamic process, covering a long period of time,
and in which many factors have participated. To these circumstances,
that make difficult a study of the origin of the Buddhist idealistic
philosophy, are to be added other facts that render that study more
difficult still: the texts that must be used for this study are in most cases
anonymous; their relative chronology is difficult or impossible to
establish; great part of Buddhist literature is lost; many important texts
are known to us only in their Chinese and Tibetan translations, which
often present problems of interpretation.
We shall limit ourselves to point out some factors of a doctrinary
nature, that seem to have contributed in a great extent to the formation
of the fundamental Yogacara doctrine, cittamiitra. Similary, factors, that
participated in the formation of the other theories of the Yogacara
school, could also be traced.
The importance of citta (mind)
Since its beginnings Buddhism has given to mind (citta, cetas, manas,
vififialJa or vijfiana) a great importance, attributing to it fundamental
functions. Mind is the determining condition for the arising of the indi-
vidual existence (namariipa) in the twelve members of the Dependent
Origination (pariccasamuppada or pratftyasamutpada). Human conduct
depends on mind. Mind is the cause of purity or impurity. Man is
directed by mind. Karman gets its moral qualification according to the
mental state or disposition with which it has been carried out. Individual
destiny and world destiny depend on karman and therefore indirectly on
Journal of the International Association of Buddhist Studies
Volume 23. Number 2.2000
JIABS 23.2 166
mind. Many of the moral qualities propounded by Buddhist ethics
belong to the realm of mind (sati or sm!ti, appamiida or apramiida,
etc.). The two pillars of the Buddhist Path are knowledg,e (fiiifw or
jfiiina, pafifiii or prajfiii) and compassion (karulJii), and knowledge is
gained through the activity and development of mind. In the way to
Liberation meditation (jhiina or dhyiina) and concentration of mind
(samiidhi) play an important role. Through a well-trained and purified
mind the Supreme Enlightenment (bodhi), the ultimate goal of Buddhist
effort, is reached.
In many texts this special importance of mind is extolled as fO,r in-
stance: Sarrzyuttanikiiya I, p. 39 PTS edition; Atiguttaranikiiya IT, p. 177
PTS edition; Dhammapiida I, 1- 2; Aryaratnamegha quoted in Santi-
deva's pp.121-122 ed. Bendall. See E. LAMOTTE's
translation of Vimalakfrtinirdesa III, par. 34, and his Introduction,
pp. 51-53, and J. MAY: "La philosophie bouddhique idealiste", pp.268-
270, for other references.
The preeminent position that citta has in the Y ogacara is thus coherent
with the importance it always has had in Buddhism. The idealistic
system of Buddhist philosophy carried one fundamental trend of
Buddhist thought to its extreme point, making mind the only existing
entity and deriving everything from it.
"Nominalism"
Buddhist philosophy distinguished between things existing dravyasat and
things existing prajfiaptisat. Dravyasat points to something that exists as
a substance, as a real entity; prajfiaptisat on the contrary points to
something that has only a nominal existence, the existence of a mere
concept, that is conventionally assumed to exist but has no objective
reality. Dravyasat exists in re, prajfiaptisat exists in mente.
In Milindapafiho IT, pp.25-28 PTS edition, in the celebrated dialogue
between the King Milinda and the Venerable Nagasena, it is declared (in
the context of the negation of an iitman) that 'Nagasena' exists only "as
a denotation (sa1ikhii), appellation (samafifiii), designation (pafifiatti), as
a current usage (vohiiro), merely as a name (niimamattam)" (I.
HORNER's translation). And a stanza of Sarrzyuttanikiiya (1, p. 135 PTS
edition) is quoted where it is said that "just as when the parts are rightly
set the word 'chariot' is uttered, so when there are the kandhas (=
skandhas), it is the convention (sammuti) to say that there is 'an indi-
vidual' (satto)".
DRAGONETTI 167
The attribution of a prajiiaptisat existence, nominal existence, to
diverse kinds of things is frequently met with in the llinayana literature.
We give some examples, which show how spread this opinion was.
The Vatslputrlya (Thesis 1, Vasumitra, Samayabhedoparacanacakra)
maintained that the "pudgala" (person, individual) is a mere denomina-
tion (prajiiapti) established in relation to the skandhas, the iiyatanas and
dhiitus. Cf. Mahiiyiinasutriilmikara XVIII, 92: prajiiaptyastitayii viicyaJ;
pudgalo dravyato na tu and commentary ad locum: prajiiaptito 'astfti
vaktavyo dravyato niistfti vaktavyaJ;.
The Prajfiaptivada school (Thesis 3, Vasumitra ibidem) taught that all
the sa1J1skiiras or sa1J1skrtas (composed or conditioned things) are
prajiiaptisat.
According to a text of Paramartha's commentary on Vasumitra's
treatise on the sects (Samayabhedoparacanacakra, quoted by Chugan =
ChOzen in Japanese, in his San louen hiuan yi, TaishO 2300, p.459 b29-
c2), the sect of the Ekavyavaharikas held that all the mundane (laukika)
and supra-mundane (lokottara) dharmas have only a nominal existence.
The Bahusrutlyas also declared that the Four Great Elements that
constitute matter are only nominally existent (prajiiaptisat), according to
the Satyasiddhisiistra of Harivarman (middle of the third century C.B.)
who expresses the point of view of that school (Taish6 1646, p.261a,
Section 37 and b-c, Section 38; Sanskrit "reconstruction" and English
translation by N. AIYASWAMI Sastri), and that the dharmas are not real
and consequently are only name (niimamiitra), conventional denomina-
tion (TaishO 1646, p. 327a, beginning of Section 141).
The Sautrantikas or Sari.krantivadins or fully adhered to
the nominalist conception of reality. They considered prajiiaptisat many
entities that for the Sarvastivadins, in their realistic inspiration, were
really existent. According to them form (sa1J1sthiina) does not exist as a
dravya (substance, thing) (Vasubandhu, Abhidharmakosa IV 3b, p. 573),
and according to the context it exists only prajiiaptitaJ; (nominally, con-
ventionally). In the Sarvastivadin classification of dharmas, priipti was
an important cittaviprayuktadharma (a dharma non-associated with
mind), whose function was to connect any acquired object with the
individual who possessed it, specially to connect the accomplished act
with the series of consciousnesses of the individual who had accom-
plished it. Priipti allowed the Sarvastivadins to explain the mechanism
of the causal retribution of actions. Sarvastivadins considered priipti as
having a real existence (dravyatafL). The Sautrantikas attributed to priipti
JIABS 23.2 168
a nominal existence (prajfiaptitaM (Vasubandhu, Abhidharmakosa II ad
3 6 c ~ d at the end, pp.217-218). Another dharma, to which the Sarvasti-
vadins attributed a real existence, was the avijfiapti. Any volition
(cetana), which is of a mere mental nature, may externally manifest it-
self through a corporal or vocal act. The gesture or words are vijfiapti,
"information", because they make known the will of the person. But, at
the same time, the volition gives rise to an invisible act, which continues
to exist and is the receptacle of the moral responsability derived from
that act. This invisible act is the avijfiapti, "non-information", because,
as it does not appear, it does not give any information. For the
Sautrantikas avijfiapti existed also prajfiaptitaJ:t, as a nominal entity.
Even the Sarvastivadins, who represented an extreme realistic position,
maintained that all beings had a nominal existence grounded on the
series (santati) that constitute them (Thesis 33, Vasumitra, op. cit.).
In the treatise Bhavasalikrantiparikatha attributed to Nagarjuna, karika
lla expresses: 'di dag thams cad min tsam ste / 'du ses tsam la rab tu
gnas / rjod par byed las tha dad pa'i / brjod par bya ba yod ma yin (all
things are only name (namamatra), they dwell only in thought; sepa-
rated from the word, what it designates does not exist).
Many other examples of the attribution of a nominal existence to
diverse entities can be found in the theses maintained by the HInayanist
sects. In fact, the nominalist conception agrees with the non-substantial-
ist position adopted by Buddhism since its very beginning. This
"nominalism", which pervades the thought of the Abhidharmic period,
does not mean the negation of the existence of beings and things; it
affects the kind of existence that beings and things possess. But anyhow
it undermined the consistency of existence, paving the way for the future
conception of Voidness and Only-Mind, in the Mahayana period.
Perception without external objects
Dreams (svapna), magical creations (nirmalJa), illusions (maya),
mirages (mariCi), eye disorders (timira), the whirling firebrand (alata-
cakra), the moon reflected in water (udakacandra), and other similar
phenomena interested Buddhist thinkers. They saw in them cases of
cognitive experiences in which non existing objects appeared to the mind
as if they were really existing. Thus these perceptions were used as
comparisons (upamana) or examples (dNtanta) for the unreality of the
empirical world, as for instance by Nagarjuna in Mulamadhyamaka-
karika VII. 34: yatha maya yatha svapno gandharvanagara1J1 yatha I
DRAGONETTI 169
tathotpadas tatM sthana1J1 tatM bhaliga udahrtam. Cf. Ta tche tou
louen (Mahiiprajfiiiparamitopadea or Mahiiprajfiaparamitilsastra),
TaishO 1509,pp.101c and ff., for a detailed enumeration and explana-
tion of these upamanas, and LAMOTTE's translation, pp:357 ff., for
more references.
In several Sutras magical creations are employed as upaya, means to
obtain some beneficial effects, as for instance Bhadramayakiiravya-
kara1}a, Vimalakfrtinirdesasutra (See E. HAMLIN, "Magical Upaya in
the Vimalakrrtinirdesasutra"), Saddharmapu1}t}arrkasutra, Chapter VII,
pp.187-188 and pp.195-197, KERN-NANJlO edition.
Moreover, these peculiar cases of perception showed the possibility of
the existence of acts of perceptual cognition which do not comply with
the conditions required by the common notion of normal perception: a
sense organ and a real external object corresponding to that sense organ.
The Sautrantikas accepted the existence of cognitions without an exter-
nal object against the opinion of the Sarvastivadins who argued that all
cognition necessarily has a real entity as its object. Cf. Collect COX, "On
the Possibility of a Nonexistent Object of Consciousness: Sarvastivadin
and Theories". In a similar way Ramanuja affirms the thesis
that any cognition has a corresponding real object (sarva1J1 vijfianajata1J1
yathartham) and he studies all the cases where it seems that there is a
cognition without a real object to prove that even in these cases a real
object is present. Cf. of the Brahmasatra, para-
graphs 66-67. In Vi7J1satika, kilrikas 1-4 and commentary ad locum, the
cases of taimirikas, persons who have their visual sense organ afflicted
by ophthalmic disorders, of dreams and mirages are mentioned as
examples of representations without object, and in Trisvabhiiva, Section
K, karikas 27-30, the magical creation of an elephant by the power of
the mantras is presented as a case of representations without object.
The acceptance of representations without a real external object is the
conditio sine qua non for the arising, development and establishment of
an idealistic explanation of reality. If the possibility of cognitions with-
out an object did not exist (as it was maintained by the Sarvastivadins),
an idealistic conception has no place.
Meditation
The Pratyutpanna-buddha-sa1J1mukhavasthita-samadhi-sutra or Bhadra-
palasutra mentions as instances of cognitions without really existing
external object, dreams (3H, HARRISON's edition), asubhabMvana or
JIABS 23.2 170
meditation on the repulsive practised by the (3J, ibidem), images
reflected in a mirror (3K, ibidem), and compares to them the visualiza-
tions of Tathagatas that occur in the meditative concentration of the
Bodhisattvas. The SaY[ldhinirmocanasiltra VIII, paragraph 7
(LAMOTTE's edition and translation = pp.152-155 POWERS' edition and
translation) expresses the same idea in relation to images seen in medita-
tion: gzugs briiam de rnam par rig pa tsam du zad pa'i phyir teo In
Trisvabhdvakdrikd, Section M, kdrikds 35-36, reference is made to the
"three knowledges" thanks to which Bodhisattvas, dhydyins and wise
people have also the experience of cognitions without external object.
An important difference between the cases of cognitions without
object given in the previous section (dreams, etc.) and those occuring in
meditation is that these latter take place as a result of the practitioner's
voluntary resolve and the application of a yogic technique.
The experience of meditation could contribute in another way to the
constitution of the cittamdtra theory. Meditation, as a yogic process, has
as its effect to allow the meditator to get diverse attainments and also to
void his mind, to liberate it from its psychological and intellectual con-'
tents, passing through the diverse stages of the meditative process, in
which the experience becomes gradually deeper. At the end of the
process the external world and the internal world (sensations, notions)
have disappeared for the meditator, who "enters in a state of calm and
cessation similar to nirvdl}a" (Ta tche tou louen, Taisho 1509, p.216a,
lines 2-3).
A complete description of the meditative process is found in the just
quoted Ta tche tou louen, pp.206a-217a = LAMOTTE's translation III,
pp.1216-1309. LAMOTTE gives in the notes to his translation a detailed
account of references concerning meditation. In Mahdvyutpatti Nos.
1477-1540 there is a complete enumeration of dhydnas, samdpattis,
apramdl}as, abhibhvdyatanas and krtsndyatanas, which consti-
tute the elements of the meditation path, in Sanskrit, Tibetan and
Chinese.
The meditative process shows that t.\1e mind, citta, can remain alone
and isolated in itself, that it can subsist without the presence of an
object, freeing itself from the subject-object duality, getting rid of the
empirical reality, and manifesting itself as the transcendent supreme
reality.
DRAGONETTI 171
The instantaneity of the dharmas
Buddhism has a dynamic conception of reality. This manifests itself in
the peculiar doctrine of the dharmas. The dharmas are the elements, the
constituent faCtors of all that exists. All that is 'material', as human
body, is constituted by material dharmas. The mental phenomena as per-
ceptions, sensations, volitions, acts of consciousness are nothing but
dharmas. And man is only a psycho-physical aggregate of material
dharmas and of mental dharmas. Reality, in its integrity, is likewise
nothing else than dharmas, isolated or accumulated. Dharmas are unsub-
stantial (aniitman), because (using the Western terminology) they do not
exist in se et per se or (using the Buddhist terminology) they do not exist
svabhiivena, i.e. they do not possess an own being; they are dependent,
produced by causes and conditions. And, besides that, since the first
period of Buddhist thought, dharmas were conceived as impermanent
(anitya). For Early Buddhism and for the Hlnayanist schools dharmas,
although unsubstantial and impermanent, were real. But in the Hlnayana
several sects added to the transitory dharmas the attribute of instantane-
ity: dharmas not only are impermanent, but also they disappear as soon
as they arise, and are replaced by other dharmas of the same species as
long as the causes that provoked the appearance of the replaced dharma
continue to exist. Thus reality is an accumulation of series of dharmas,
in a process of vertiginous constant replacement. The result is that, as
D.N. SHASTRI, The Philosophy of Nyiiya, p.189, says: "the reality,
according to the Buddhist, is not static; it is dynamic. It is not being; it
is becoming."
Among the Hlnayanist sects that maintained the instantaneity of the
dharmas were the Sarvastivadins, the Vatslputrlyas, the MahIsasakas,
and the Kasyaplyas, and the sects derived from them, according to Vasu-
mitra's I pu tsung [un [un, Taisho 2031, pp.16c, line 2; 16c, lines 15-
16; 17a, lines 13-14; and 17b, line 1 (= A. BAREAU: "Trois Traites sur
les Sectes Bouddhiques attribues a Vasumitra, Bhavya et VinItadeva":
255, 257, 262 and 265, and J. MASUDA: "Origin and Doctrines of Early
Indian Buddhist Schools": 50, 54, 62 and 65). The Pubbaseliyas and the
Aparaseliyas, both derived from the Mahasarp.ghikas, affirmed also the
instantaneity of the dharmas, according to Buddhaghosa's commentary
to the Kathiivatthu included in PancappakaraIJa-Atthakathii (XXII, 8,
p.198 PTS edition). Vasubandhu in Abhidharmakosa IV, 2d, pp.568-
569, emphatically declares that "what is conditioned (- and all is -) is
momentary" (sarrzskrtarrz and ad locum: ko yarrz
nABS 23.2 172
nama? atmalabho 'nantaravinasr, so ' syastfti Yasomitra's
commentary ad Abhidharmakosa II, 46b, p.262, line 26, refers to the
V with the term On the contrary the Thera-
vadins, according to the quoted text of the Kathavatthu, xxn, 8, did not
accept the momentariness of the dharmas, and this explains why they
remained attached to the realistic conception of the world.
The new attribute of instantaneity produced an enormous effect in the
Buddhist theory of knowledge: if dharmas are not only impermanent but
also instantaneous - and dharmas constitute the whole reality - and we
do not perceive that momentariness of the dharmas but only compact
things that seem to be there as the objects of cognition, then we do not
see reality as it truly is.
Nagarjuna's conception of reality
One of the principal tasks of Nagarjuna is to establish the logical impos-
sibility of the existence of elements, manifestations, categories of the
empirical reality, as for instance: birth and destruction, causality, move-
ment, time, sensorial activity, the elements that constitute man, passion
and its subject, action and its agent, suffering, karman, saf!lsara, etc.
This impossibility derives from the fact that all is conditioned, related,
dependent, contingent, and as such lacks an own being, a svabhava, an
existence in se et per se. Everything is sanya, "void", svabhavasanya,
"void of an own being". The abolishing analysis, to which Nagarjuna
submits the whole reality, leaves a great void, Sanyata, Voidness, in
which nothing belonging to the empirical reality which appears before
us remains. But normal knowledge does not reach the true reality of
Sanyata, which is covered, concealed by an apparitional reality, the
empirical,reality, beyond which normal knowledge cannot go. We do
not perceive what really exists (paramarthasatya = Supreme Truth or
Reality); we only perceive something that is inexistent, false, illusory
(saf!lvrtisatya, concealing truth or reality = Relative Truth or Reality), as
the dreams, mirages, magical creations etc. to which Nagarjuna's school
so frequently compares the world in which we live. The situation in the
case of Nagarjuna is similar to that of the theory of the dharmas as
maintained in the Hlnayana: we perceive something different from what
really exists, things are not as they appear.
It could be said that the theory of the instantaneity of the dharmas and
Nagarjuna's conception of reality, which stress the separation between
what is outside our mind and our mental representations, are the two
DRAGONETTI 173
more important factors for the forthcoming of the idealistic theory that
there is nothing apart from the creations of our mind.
The philosophical inference
Given the preceding historical, philosophical conceptions - the impor-
tance of the mind conceived as the determining principle of human
conduct and of man's and world's destiny; nominalism which transforms
the reality in which we exist in a collection of names and labels and
undermines the consistency of beings and things; the awareness of the
existence of many cognitions which are cases of representations without
a real external object; the experience of meditation which has both
powers: to visualize objects at will and to suppress the surrounding
reality and the contents of the mind, leaving the mind empty and
isolated; the instantaneity of the dharmas (the constituent factors of what
exists, the sole existing true reality that remains concealed to our normal
knowledge which is limited to perceive something that is not there and
unable to perceive what is really there); and Nagarjuna's conception of
reality which dissolves all that exists into aVoid, depriving beings and
things of real existence, making cognition an instrument condemned to
grasp only illusions and falseties, and positing the impossibility for
normal knowledge to reach reality - given these conceptions, it was not
difficult for philosophically very well trained minds, as were Buddhist
thinkers, to ask themselves: if what we perceive is not outside (the realm
of the object), wherefrom does it come? and to answer: from the mind
(the realm of the subject). Thus they rounded an inference whose
premises originated in the beginnings of Buddhism. Only Mind was the
logically valid conclusion for a reasoning that had lasted for centuries.
nABS 23.2 174
References
Asailga: Mahayiinasiitriilarikara (1907). 13dite et traduit ... par Sylvain Levi, Tome I.-
Texte, Paris: Libraririe Honore Champion.
Bareau, A. 1954: "Trois Traites sur les Sectes Bouddhiques attribues it Vasumitra,
Bhavya et Vinltadeva;" 10umaiAsiatique 242 (Paris): 229-266.
Cox, C. 1988: "On the Possibility of a Nonexistent Object of Consciousness:
Sarvastivadin and Theories," JIABS 11.1: 31-87.
Hamlin, E. 1988: "Magical Upiiya in the Vimalakfrtinirdeiasiitra," JIABS 11.1:
89-121.
Harivarman: Satyasiddhisiistra (Vol. I 1975, Vol. IT 1978), Baroda: Oriental Institute.
Lamotte, 13.1962: L'Enseignement de Vimalakfrti (Vimalakfrtinirde.sa), traduit et annote
par ... Louvain: Institut Orientaliste.
Mahiivyutpatti (1962): Ed. by Ryozaburo Sasaki, Tokyo.
Masuda, Jiryo 1925: "Origin and Doctrines of Early Indian Buddhist Schools," Asia
Major IT: 1-78.
May, Jacques 1971: "La philosophie bouddhique idealiste," Asiatische Studien XXV
(Bern: Francke Verlag): 265-323.
Milindapaiiho (Vol. I 1963, Vol. IT 1964) = Milinda's questions, translated by I. B.
Horner, Oxford: Pali Text Society.
Nagarjuna: Miilamadhyamakakiirikas (Madhyamakasiitras) (1903-1913). Publie par
Louis de la Vallee Poussin, Osnabrock: Biblio Verlag 1970.
Nagarjuna: Bhavasarikriintiparikathii (1938). Ed. N. Aiyaswami Sastri, Adyar: Adyar
Library.
Pali texts are quoted from the editions of the Pali Text Society.
Ramanuja: of the Brahmasiitra (1959-1964). Edited by R.D. Karmarkar,
Poona: University of Poona.
Saddharmapu/Jrj.arfkasiitra (1908-1912): Ed. by Prof. H. Kern and Prof. Bunyiu
Nanjio, Osnabriick: Biblio Verlag 1970.
Sa1'Jldhinirmocanasiitra (1935) = Sa1'Jldhinirmocana Siitra. L 'Explication des Mysteres,
Texte Tibetain edite et traduit par Etienne Lamotte, Louvain-Paris:
Adrien Maisonneuve.
Sa1'Jldhinirmocanasiitra (1995) = Wisdom of Buddha. The Sa1'Jldhinirmocana Siitra,
translated by John Powers, Berkeley, USA: Dharma Publishing.
Santideva: = 9ikshasamuccaya. A Compendium of Buddhistic
Teaching (1897-1902). Ed. Cecil Bendall, Gravenhage: Mouton & Co.
1957.
DRAGONETTI 175
Shastri, D.N. 1964: The Philosophy of Nyaya- V a i s e ~ i k a and its Conflict with the
Buddhist Dignaga School (Critique of Indian Realism), Delhi-
Varanasi: Bharatiya Vidya Prakashan 1976.
Ta tche tou louen, Mahaprajnaparamitopadeia (1970). Le Traite de la Grande Vertu
de Sagesse de Nagarjuna (Mahaprajnaparamitasastra), avec une
nouvelle Introduction par Etienne Lamotte, Tome Ill, Louvain: Institut
Orientaliste.
Taish6 The Taisho Shinshu Daizokyo (The Tripitaka in Chinese) (1925).
Edited by Prof. Dr. J. Takakusu, Prof. Dr. K. Watanabe, Tokyo: The
Taisho Shinshu Daizokyo Kanko Kai (Society for the Publication of
the Taisho Tripitaka) 1971.
Tola, F. and Dragonetti, C. (1983): ''The Trisvabhilvakarika of Vasubandhu," Journal
of Indian Philosophy 11: 225-266.
Vasubandhu: Abhidharmakosa & Bharya (1971). Critically edited by Swami Dwarika-
das Shastri, Varanasi: Bauddha Bharati.
Vasumitra's I pu tsung lun [un = Samayabhedoparacanacakra.
Vasumitra: Samayabhedoparacanacakra, etc. (1936). Ed. Teramoto Enga and Hira-
matsu Tomotsugu, Tokyo 1975.
Vi'!lsatika (1925): Vijnaptimatratasiddhi. Deux Traites de Vasubandhu. ViIp.satika ... et
TriIp.sika ... Original Sanscrit pUblie ... par Sylvain Levi, Ire Partie -
Texte, Paris: Librairie Ancienne Honore Champion.
JOHN KIESCHNICK
Blood Writing in Chinese Buddhism
One of the most common forms of ascetic practice in Chinese Buddhism
involves pricking one's tongue or finger to draw blood which is then
mixed with ink and used to copy Buddhist scriptures. At first glance, the
practice of copying scriptures in human blood seems to be a classic
example of what the prominent Chinese scholar Hu Shih (1891-1962)
once termed the "Indianization" of China. Speaking in the early part of
the twentieth century, HU lamented the infiltration of wild, irrational
Indian ascetic practices into what he saw as an essentially reasonable and
practical Chinese religion.! A closer look at the origins of the practice of
blood writing shows that, while its roots can be traced in part to Indian
Buddhism, blood writing was sustained in China by a mixture of foreign
and indigenous beliefs and traditions. Further, once we appreciate these
beliefs and traditions, we see that far from the irrational masochistic acts
of disturbed minds, blood writing, like other related ascetic practices,
was a reasoned, understandable act.
The earliest description of a Chinese figure copying a scripture in
blood appears to be a passage in the History of the Chen Dynasty which
states that the prince Chen Shuling copied the Nirvana Sutra in
his own blood in 579. The History of the Chen Dynasty was compiled in
the early seventh century and so the story may have been invented at that
time.2 In any event, certainly by the seventh century Chinese monks and
laymen were pricking their fingers and tongues, mixing the blood with
ink, and copying out Buddhist scriptures in the murky mixture. The
seventh-century monk Haishun mJII&, for instance, in addition to
shedding his own blood in veneration of some relics, also "mixed his
blood with ink and wrote out the Scripture of the Precepts of the Seven
Buddhas"} The monk Wen' gang :zM, also ofthe seventh century, was
1. "The Indianization of China: A Case Study in Cultural Borrowing," in Inde-
pendence, Convergence, and Borrowing in Institutions, Thought, and Art
(Cambridge: Harvard University Press 1937), pp. 219-247.
2. Chenshu (Beijing: Zhonghua Shuju 1972) 36, p. 495.
3. Xu gaoseng zhuan 30, J. TAKAKUSU Taishi5 shinshu
daizokyo (Tokyo: TaishO issaikyo kankokai 1924-1932), here-
Journal of the International Association of Buddhist Studies
Volume 23. Number 2.2000
JIABS 23.2 178
said to have copied some six hundred fascicles of Buddhist scriptures in
his blood.
4
Dozens of such references in medieval sources testify to the
popularity of blood writing at that time. The poet-monk Gu;mxiu l{f*
(832-912) even composed a poem in praise of a fellow monk who
completed a copy of the A vata'!lsaka in his blood, "his ten fingers
drained [of blood] to complete seven scrolls."5 In the medieval period,
the practice seems to have been admired by more than Buddhist
devotees, even attracting the attention of an occasional emperor. For
instance, two fascicles of unidentified scriptures written in the blood of
Zengren a ninth-century monk, were submitted to the emperor
upon the monk's death.
6
Nor does the practice seem to have waned in later periods. Extant
prefaces and letters of prominent monks in the sixteenth and seventeenth
centuries contain many references to blood writing, carried out by
everyone from leading Buddhist thinkers to humble farmers. The
practice crossed boundaries dividing monastic and lay, famous and
obscure, man and woman. Scholars of Chinese Buddhism in the first
half of the twentieth century have documented the continuing popUlarity
of blood writing among Chinese monks.? This large body of evidence
provides ample material for exploring the motivations behind the
practice.
Scriptural Sources
One important source for blood writing was scriptural. Several Buddhist
scriptures refer to the practice of writing in one's blood. The reason the
A vata'!lsaka Sutra was frequently copied in blood can be traced to a
after "T", no. 2060, vol. 50, p. 525a. I have been unable to locate other references
to the Scripture of the Precepts of the Seven Buddhas (Qifo jie jing
4. Song gaoseng zhuan T 2061, vol. 50, p. 792a.
5. Quan Tang shi (Beijing: Zhonghua Shuju 1960) 837, p. 9438.
6. Song gaoseng zhuan 26, p.877b. Several texts that refer to Zengren's blood-
written scriptures survived at Dunhuang, indicating the popularity of Zengren and
his exploits in the Tang. For references, see Stephen TEISER, The Scripture on
the Ten Kings and the Making of Purgatory in Medieval Chinese Buddhism
(Honolulu: University of Hawaii Press 1994), p.127, n.19.
7. J. PRIP-M0LLER: Chinese Buddhist Monasteries. Their Plan and its Function as
a Setting for Buddhist Monastic Life (1937; rpt. Hong Kong: Hong Kong
University Press 1982), p.323; Holmes WELCH: The Practice of Chinese
Buddhism 1900-1950 (Cambridge: Harvard University Press 1967), p. 323.
KIESCHNICK 179
passage near the end of the scripture in which the bodhisattva Samanta-
bhadra explains to the pilgrim Sudhana that the Buddhist scriptures are
so vast that one can never hope to master them all. Instead, the bodhi-
sattva advises, Sudhana should make ten great vows, including vows to
give reverence to the buddhas, to propagate Buddhist teachings, and so
forth. Under the vow to "Imitate the Buddhas," the bodhisattva explains
that the buddha Vairocana was " ... willing to give even his life. He
peeled off his skin for paper, broke off a bone for a pen, and drew his
own blood for ink. The scriptures he copied in this manner stacked up as
high as Mount Sumeru. All of this he did out of respect for the Law."8
Next to the Avatal?1saka passage, the most influential reference to blood
writing for Chinese Buddhists was an injunction in the Brahma's Net
Scripture, a text that claims to be a translation of an Indian original, but
was probably in fact compiled in China in the fifth century.9 At one
point, the Brahma's Net enjoins the Buddhist devotee to "keep, read and
recite the scriptures and monastic regulations of the Great Vehicle with a
single mind. Cut away your skin for paper, draw your blood for ink and
use your marrow for water. Break off a piece of your own bone for a
pen and copy out the Buddhist precepts."10 Note that in these texts the
procedure includes removing a piece of one's skin and breaking off a
piece of bone in addition to writing with one's blood. The same is true
for several other early scriptural references to the practice. I I
8. Da fangguang fo huayan jing *1:J (Skt. * Buddhiivata1'J'!sakasutra)
T 293, vol. 10, p. 845c.
9. See Paul GRONER, "The Fan-wang ching and Monastic Discipline in Japanese
Tendai," in Robert E. Buswell ed., Chinese Buddhist Apocrypha (Honolulu:
University of Hawaii Press 1990), p. 254.
10. Fan wang jing T 1484, vol. 24, p. 1009a.
11. The Treatise on the Peifection of Great Wisdom refers to the practice in the story
of a brahmaciirin named Dharrnarakta. According to the text, after years of fruit-
less practice, Dharmarakta encounters a Brahman who tells him '''I possess a
stanza of the sacred Law. If you truly love the Law, I will give it to you.'
Dharrnarakta replied, 'I truly love the Law.' The Brahman then said, 'If you truly
love the Law, you should make paper of your skin, a pen of your bones, and
copy it in your blood.'" (Da zhi du [un [Skt. *Mahiiprajiiiipiirami-
topadea] 16, T 1509, vol. 25, p.178c; translated with notes in Etienne
LAMOTTE, Le traite de la grande vertu de sagesse [Louvain: Institut Orientaliste
1944-1980], p. 975). A text purportedly translated in the Eastern Jin (317-420)
entitled the Original Acts of the Bodhisattva recounts a similar tale of a master
who "on account of a single stanza sliced off his own skin for paper, broke off a
bone for a pen and mixed his blood with ink." (Pusa benxing jing i?Fi\iI*1''Jt.i
nABS 23.2 180
There is no evidence that devotees in India ever took such passages as
literal guides to practice rather than marvelous accounts of distant
Buddhist heroes whose incredible actions are to be revered .rather than
copied.
12
Indeed, short of an extreme ascetic trance, it is next to impos-
sible to carry out the full practice, breaking off a piece of one's own
skeleton, dipping it in one's own blood and writing on a piece of skin
stripped from one's own body. As far as I can tell, it was only in China
(and later in East Asian cultures directly influenced by China) that the
practice of mixing blood with ink and copying scriptures with a brush
on paper was extracted from scattered textual references, and
carried out, reflecting the determined zeal of devotees who saw them-
selves on the margin of the Buddhist world.
These texts provided the scriptural basis for blood writing in both
canonical Indian texts and indigenous Chinese scriptures. But we should
resist the temptation to stop here. It is not enough to cite a few scriptural
sources, however influential, to explain a given practice, especially one
as demanding as blood writing. The scriptures leave many basic
questions unanswered. They do not, for instance, specifically explain the
advantages to be derived from writing a scripture in one's blood. How
did Chinese Buddhists expect to benefit from the practice? Further, the
huge corpus of Buddhist scriptures describes all manner of beings
performing all manner of deeds. What was it about this particular act
that inspired Chinese Buddhists to ignore the parts of the practice that
are impractical (writing with a piece of one's own bone), and adopt the
part of the description that is possible, taking up brush and knife and
trying it themselves? Below I explore the cultural and historical factors
that made the practice of copying Buddhist scriptures in blood
acceptable and even appealing to Chinese Buddhists. Specifically, I
[Skt. *Bodhisattvapiirvacarya] T 155, vol. 3, p. 119b). Similar accounts appear in
theXianyujing T 202, vol.4, p. 351b, and in the Luoyang qielanji
ftmlftrl 5, T 2092, vol. 51, p.1020b; Yi-t'ung Wang (trans.), A Record of
Buddhist Monasteries in Lo-yang (Princeton: Princeton University Press 1984),
p.232.
12. James BENN has demonstrated a similar pattern at work in the practice of burning
the body. That is, in Indian texts the practice is carried out by supernormal
beings, whereas among Chinese Buddhists it is taken literally. See "Where Text
Meets Flesh: Burning the Body as an Apocryphal Practice in Chinese
Buddhism," History of Religions, vol. 37, no. 4 (1998), pp. 295-322.
KIESCHNICK 181
discuss the roots of blood writing in notions of merit, asceticism, and
Chinese attitudes towards blood in general.
Merit
One key factor behind the popularity of copying scriptures in blood was
the prevalence of the idea that one could derive merit from copying
Buddhist scriptures in the more conventional medium of ink. From at
least the third century B.C., Indian Buddhists adhered to the notion that
one can gain "merit" from either giving or making certain objects, and
that this merit can then either be used to one's benefit in this life or the
next, or be transferred to one's intimates to improve their lot. For our
purposes here, however, the key development in Indian Buddhism carne
much later, in the first centuries of the Common Era, with the emer-
gence of what has been termed the "cult of the book." The Perfection of
Wisdom literature, the Lotus Sutra and the A vata'flsaka Sutra all extol
not just the propagation of Buddhist teachings, but specifically the
propagation of the physical Buddhist books that contain these teachings,
and assure their readers of the merit accruing to thbse who copy
Buddhist texts. 13
By the time Buddhism entered China in the first century of the
Common Era, China could already boast a long tradition of book
making, book learning, and book collecting.l
4
Not only were books
respected for their content and the erudition they represented; they were
also admired for their calligraphy, for the way they looked. All of these
factors carne together in the reading and copying of texts. In particular,
13. See, for instance, the Anasahasrika translated by Edward CONZE, in The
Perfection of Wisdom in Eight Thousand Lines and Its Verse Summary (Bolinas,
California: Four Seasons Foundation 1973), pp.104-106. The passage is
discussed in Gregory SCHOPEN, "The Phrase 'sa prthivlpradesas caityabhiito
bhavet' in the Vajracchedika: Notes on the Cult of the Book in Mahayana," Indo-
IranianJournall7.3/4 (November-December 1975), pp.l47-87; see also, Leon
HURVITZ, translator, Scripture of the Lotus Blossom of the Fine Dharma (New
York: Columbia University Press 1976), p.l74.
14. For surveys, see TSIEN Tsuen-hsuin, Paper and Printing, Part 1 of Volume 5,
Chemistry and Chemical Technology, in Joseph Needham ed., Science and
Civilisation in China (Cambridge: Cambridge University Press 1985); Jean-
Pierre DREGE: Les bibliotheques en Chine au temps des manuscrits (Paris: Ecole
d'Extreme-Orient 1992).
nABS 23.2 182
copying manuscripts was a way to at once read, study, and memorize a
text while at the same time practicing the art of calligraphy. 15
In the medieval period, the felicitous confluence of Buddhist notions
of merit with the Chinese passion for books produced an unprecedented
outpouring of Buddhist manuscripts. The production of many if not
most of these manuscripts was motivated at least in part by a desire to
acquire merit. Medieval collections of Chinese miracle tales tell stories
of men rewarded for copying Buddhist scriptures with birth in a pure
land, or, for instance, of a recently deceased woman saved from a
tortured stay in hell by a sister in the world of the living who
scriptures on her behalf.
16
More concretely, colophons to medieval
manuscripts state explicitly that the copier or the donor who commis-
sioned the copy of a Buddhist scripture expected to derive merit from it.
In most cases, the colophon indicates that this merit is to be directed
towards the benefit of a relative.
What evidence we have for the motivation behind the practice of
writing scriptures in blood indicates that the belief in merit was funda-
mental to this practice as well. Take, for example, the blood-written text
of the prominent sixteenth-century monk Hanshan Deqing fI2; [lj In
his autobiography, Deqing explained his reasons for taking on the
project as follows:
In the Spring of my thirty-second year, I returned from Yanmen. At this time I
recalled the benevolence of my [deceased] parents and the care they had given
me. I also thought of all of the obstacles that stood between me and the [Buddha]
Law. On reading the vow of the great master Huisi of Nanyue, I vowed to make
a copy of the Scripture of the Expanse of Buddhas of the Flower Adornment (i.e.
the Avata1]'lsaka) by mixing my own blood with gold. Above, this would tie me
to the karma of prajiili, and below it would repay my parents for their benevo-
lence.17
15. Jean-Pierre DREGE, "La lecture et l'ecriture en Chine et la xylographie," Etudes
chinoises 10, no. 1-2 (1991), p. 92.
16. lin 'gang banruo jiyanji in Xu zang jing (Taipei: Xin-
wenfeng 1968-1970), a reprint of NAKANO Tatsue "P!fJ!lI\f, ed., Dai Nihon
zokuzokyo :k B (Kyoto: Z6ky6 shoin 1905-1912), vol. 149, p.49a;
lin 'gang banruo jiyan ji B, p. 46a.
17. Hanshan laoren mengyou ji 53, pp.37-38 in Xu zang jing,
vol. 127. See also Sung-peng Hsu, A Buddhist Leader in Ming China: The Life
and Thought of Han-shan Te-ch'ing, 1546-1623 (University Park: The Penn-
sylvania State University Press 1979), p. 72. The vow of Huisi that Deqing refers
to here is a piece by the eminent sixth-century monk in which he states his vow to
KIESCHNICK 183
Deqing's statement that his action would "repay my parents for their
benevolence" refers to the transfer of the merit from copying the
scriptures to his parents in the afterlife: In accordance with well-attested
Buddhist noti()ns of merit, Deqing's parents could receive a better
rebirth with the merit earned for them by their son, including perhaps
eons in a pure land, or a shorter stay in a less pleasant destination.
Deqing's statement is not surprising, as parents are one of the most
common recipients of merit in all such dedicatory inscriptions, whether
for Buddhist statues or scriptures - a practice with roots in Indian
Buddhism. IS
With this Buddhist background in mind, references to blood writing in
the medieval secular dynastic histories begin to make sense. The Tang
History, for instance, describes an eighth-century official named Yuan
Dexiu who on his mother's death "ate neither salt nor kumiss,
did not sleep on rushes or mats, and pricked himself, using the blood to
paint images and copy Buddhist scriptures."19 When the father of the
ninth-century figure Wei Shou died, he too "pricked himself, using
the blood to copy Buddhist scriptures."20 Although neither of these
passages makes reference to merit - not surprising in these secular
sources which grant mention of Buddhist matters only with great reluc-
tance - it is almost certain that Yuan and Wei copied scriptures for their
parents in the belief that doing so would assist them in the afterlife.
copy the Prajiitlptlramittl with gold leaf, a practice that was often repeated at
great expense by devotees of later periods as well. For Huisi's vow, see Nanyue
Si Da Chanshi lishi yuanwen T 1933, vo1.46, p.786c,
translated and discussed in Paul MAGNIN, La vie et ['oeuvre de Huisi (Paris:
Ecole Fran"aise d'Extreme-Orient 1979), pp. 192-238.
18. See Gregory SCHOPEN, "Filial Piety and the Monk in the Practice of Indian
Buddhism: A Question of 'Sinicization' Viewed from the Other Side" in his
Bones, Stones, and Buddhist Monks. Collected Papers on the Archceo[ogy,
Epigraphy, and Texts of Monastic Buddhism in India (Honolulu: University of
Hawaii Press 1996), pp. 56-71.
19. Jiu Tangshu 'lfmif (Beijing: Zhonghua Shuju 1975), 190b, p. 5050. A tenth-
century Buddhist banner with an inscription claiming that it was painted in blood
survived at Dunhuang: Stein painting 196, reproduced in Roderick WIDTFIELD,
The Art of Central Asia: The Stein Collection in the British Museum (Tokyo:
Kodansha 1982-), vol. 2, plate 32, with discussion on pp. 323-324. WHITFIELD
notes that the pigment on the banner does not appear to be blood. It may be that
the amount of blood mixed with other pigment was so small that the blood is not
apparent.
20. Jiu Tangshu 162, p. 4244.
JIABS 23.2 184
At times the merit for blood writing was applied to the living as well
as the dead. For instance, when Tang emperor Suzong (r. 756-761) took
ill, his empress was said to have copied Buddhist scripturesjn her own
blood, apparently in the belief that the merit from her actions would
assist her husband's recovery.21Sirnilarly, a Song-era figure named Gu
Xin ,Il'i:Ff was said to have brought eyesight back to his blind mother by
copying out several fascicles of Buddhist scriptures in his blood.
22
The
range of references to blood writing shows just how prevalent both this
practice and the notion of Buddhist merit were; blood writing was not
confined by boundaries of gender, religious or social status: m<;:>nks,
nuns, humble laymen and powerful empresses all found occasion to copy
a Buddhist scripture or two in their blood.
The curious colophon of an anonymous devotee preserved at
Dunhuang provides us with an example of the more subtle, sophisticated
uses to which merit derived from blood writing could be put. The man's
dedication to a copy of the Diamond Sutra reads as follows:
On the fifteenth day of the fourth month in the third year of Tianyou, bingyin in
the sequence of years [30 April 906], an old gentleman of eighty-three pricked
himself to draw blood, which he mixed with ink. He wrote this scripture with his
own hand for propagation to all of the believers in Sha Prefecture. May the state
and the land be still and peaceful; may the wheel of the Law turn forever. Should
I die in writing it, I ask only that I quickly pass out of this world. I have no other
prayers.23
Another colophon by the same figure appended to yet another copy of
the Diamond, written a few months earlier, is equally humble and
generous. He writes:
On the third day of the second month in the third year of Tianyou, bingyin in the
sequence of years [28 February 906], an old man of eighty-three pricked the
middle finger of his left hand. He drew blood to make the ink fragrant and wrote
this Dia[mond] Scripture for transmission to people of believing hearts. He is
entirely without prayers. [Since] original nature is truly empty, there is no
pleasure for which to pray.24
21. Jiu Tangshu 261, p. 260.
22. Songshi *.\t'. (Beijing: Zhonghua Shuju 1977) 456, p. 13394.
23. P. 2876, reproduced in HUANG Yongwu ii7.k:lt\, Dunhuang baozang : ~ l l J . l n l ~
(Taipei: Xinwenfeng 1985), vol. 124, p. 622a. The translation is from TErSER,
The Scripture on the Ten Kings, p. 126, which also includes discussion of the
passage along with annotation that I have omitted here.
24. Or.821O/S.5669, reproduced in HUANG, Dunhuang baozang, vol. 44, p.300b;
translation from TEISER, The Scripture on the Ten Kings, p. 127. Curiously,
KIESCHNICK 185
The phrase he "drew blood to make the ink fragrant" should not be
taken literally, and probably means that he "ornamented" the writing
with his physical suffering, a sign of his sincerity and a topic to which
we will return.below. For now, let us focus on the copiest's understand-
ing of merit. In the first of these colophons, his merit is directed toward
lofty, abstract targets: "the state and the land," and "the [Buddha] Law."
In the second, he makes a point of noting that his merit was directed
nowhere in particular. Evidently, this particular copiest had taken to
heart the principles of emptiness expounded in the Diamond Sutra: as all
is ultimately devoid of permanence (including the self), there is, in the
end, no tangible object to receive merit. The "Old Gentleman" was,
however, an exception: not surprisingly, most who copied the scripture
in their blood did so expecting to receive tangible, direct benefits from
it. Indeed, a carefully hidden expectation of merit for the deed lurks
behind the words of even this devotee. Why else would he go to the
trouble of copying out the scripture in his blood, when, for conveying
its teachings, ink would have served the same purpose? As the great poet
Bai Juyi put it in the early ninth century in a dedication to a set of
scriptures carved in rock, a scripture "drawn in blood mixed with ink on
paper made of one's skin deteriorates and disappears, like painting with
a brush on water."25 If one's goal was simply widespread distribution of
texts, copying with ink made more sense.
Yet even if we recognize that one of the major motivating factors for
copying scriptures was not only to propagate the teachings they contain
but also to gamer merit, what was to be gained by copying scriptures in
blood? The Diamond Sutra asks its readers to reproduce the text, but
says nothing about blood. The notion of Buddhist merit contributed to
the birth and of printing in China because of the belief that the
production of huge amounts of texts would naturally produce huge
amounts of merit.
26
Yet by the sixteenth century, when Deqing lived,
the practice of blood writing continued despite the fact that printing was
initial tests on this manuscript by specialists working for the British Library have
failed to reveal any traces of blood. Presumably, and as said before (note 19), the
blood used was in such small quantities that it can no longer be detected, though
the possibility that the donor lied in his inscription cannot be ruled out. Susan
Whitfield, personal communication.
25. "Suzhou Chongyuansi Fahuayuan shibijing beiwen"
in Quan Tangwen (Beijing: Zhonghua Shuju 1959) 678, p. 2a.
26. DREGE: Les bibliotheques en Chine, p. 266.
JIABS 23.2 186
by then widespreadP Clearly, there was more to merit-making than
sheer numbers. This brings us to the importance of ascetic ideals in the
practice of blood writing; for in calculations of merit, in- addition to
numbers, sincerity (as expressed through the willingness to endure
physical pain) was also factored into the karmic equation.
Asceticism
Behind the choice of blood over ink was an admiration for self-sacrifice
and specifically for self-inflicted physical suffering. Though primarily a
Buddhist concept, this notion was not entirely foreign to pre-Bw;ldhist
China. Texts from the Warring States period (475-221 B.C.) like the
Analects and the Mazi express a general suspicion of sensuality -
Confucius warning that "The gentleman seeks neither a full belly nor a
comfortable home," Mozi inveighing against decorative clothing.
28
And
sons were praised for depriving themselves of comfort in service to their
parents.29 But with the entry of Buddhism to China, self-abnegation was
taken to greater extremes, and harsher forms of asceticism involving
self-mutilation and even ritual suicide became increasingly common.
From the early medieval period on, monks, nuns and laypeople
embraced a number of ascetic practices, including the burning of marks
on the body and scorching or slicing off fingers. Stories of fierce monks
who cut off ears, gouged out eyes and chopped off arms circulated
widely in collections of miracle tales and biographies of eminent monks.
The motivations for these practices were various. Most Chinese
Buddhists believed that such acts were supported by the teachings of the
Buddha. Many were inspired to follow the example of the Buddha who
in previous incarnations committed great acts of self-sacrifice with no
regard for his own comfort. Many believed that acts of self-abnegation,
27. Indeed, Deqing himself participated in several projects devoted to printing
Buddhist texts.
28. Lunyu 1.14; D.C. LAU (trans.), The Analects (1979; rpt. Middlesex: Penguin
Books 1986), p. 61. Mozi 20; Burton WATSON (trans.): Basic Writings of Mo
Tzu, Hsiin Tzu, and Han Fei Tzu (New York: Columbia University Press 1967),
p. 62. See also Stephen ESKILDSEN, Asceticism in Early Taoist Religion (Albany:
State University of New York Press 1998), pp.l-13.
29. The focus of these deprivations was usually on mourning rites. Keith KNAPP
argues that the new emphasis on self-deprivation was part of a larger transforma-
tion of attitudes towards "filiality" (xiao ) in the Warring States period. Keith
KNAPP: Accounts of Filial Sons: Ru Ideology in Early Medieval China (Ph.D.
dissertation for the University of California at Berkeley 1997), p. 61.
KIESCHNICK 187
including self-mutilation, would improve their chances of rebirth in a
pure land. Elsewhere I have made the case that self-mutilation was also
inspired by a more nebulous wish to purify the body.3o
The practice. of blood writing was informed by these ascetic practices
with which it was closely tied. Dinglan YEIi, a ninth-century monk, in
addition to writing scriptures in his blood, was also said to have branded
his arms and even to have plucked out his eyes to feed wild animals.
31
His contemporary Daozhou mA-, in addition to painting an image of
A valokitesvara in his blood, also cut off a piece of his left ear when
praying for rain.
32
The Tang History records that when the parents of a
certain Wang Iingru died he both "copied Buddhist books in his blood"
and "cut off two fingers from his hand."33 Blood writing, then, was one
practice among many, arrayed on an ascetic scale running from minor
acts of self-mortification to practices involving intense physical pain.
The language of these accounts of blood writing further reveals that the
practice was respected not so much for any magical properties of the
blood itself, but because of the physical pain of cutting oneself.34 The
description of the seventh-century monk Haishun's austerities ends with
the line "acts of his in which he overcame himself and tested his will
were all of this sort."35 Tenth-century monk Zhenbian's ~ m blood
writing is described as a reflection of his ability to "overcome suffer-
ing."36 And the description of the austerities of Wen'gang, are capped
with the phrase "He suffered greatly, with great perseverance, and
entered into the sea of selflessness."37 In sum, blood writing was
30. John KIESCHNICK: The Eminent Monk: Buddhist Ideals in Medieval Chinese
Hagiography (Honolulu: University of Hawaii Press 1997), pp. 35-66. See also,
BENN, "Where Text Meets Flesh."
31. Song gaoseng zhuan 23, p. 856b.
32. Song gaoseng zhuan 23, p. 859a. See also BENN, "Where Text Meets Flesh,"
pp.31O-12.
33. Xin Tangshu *Hm'B (Beijing: Zhonghua Shuju 1975) 195, p. 5591.
34. This contrasts with the Tibetan practice of making paintings with blood from the
nosebleeds of holy men. See David JACKSON, A History of Tibetan Painting
(Vienna: Verlag der Osterreichische Akadernie der Wissenschaften 1996), p. 251;
257, n. 574.
35. Xu gaoseng zhuan 13, p. 525a.
36. Song gaoseng zhuan 7, p. 747a.
37. Song gaoseng zhuan 14, p. 792a.
JIABS 23.2 188
perceived to be one of many possible ascetic acts, all based in part on a
shared belief iri the power of corporal self-sacrifice.
Some references disclose more specific details about tl1e point of
inflicting pain on oneself. Ming prefaces to scriptures copied in blood
emphasize the transformative potential of blood writing. Deqing, for
instance, vows, through copying a scripture in his blood, "to exchal1ge
this illusory body for one that is permanent and adamantine."38
Elsewhere, he notes that "by drawing his blood to copy this scripture,
the blood of his illusory body will drip into the sea of the Dharma
nature."39 But above all, Ming prefaces to blood-written scriptures,
couched in elegant, flowery language and redolent with Buddhist erudi-
tion, repeatedly emphasize the superiority of fierce, selfless practice over
the empty talk of exegesis. Ouyi Zhixu (1599-1655) speaks
derisively of "pretentious Chan monks with their lofty talk of philosoph-
ical principles who know nothing of actual practice" compared to the
"inconceivable merit" of writing the Lotus Sutra in blood.
4o
Zibo
Zhenke (1543-1603) in a postface to a blood-written scripture
tells the story of a monk, "dull by nature," who nonetheless came to
comprehend the principles of the A vatarrzsaka not through recourse to
teachers or commentary, but by copying the text out in his blood.
41
Even the most sophisticated, literate monks were always ready to dispar-
age intellectual effort when compared to manly physical practice.
Indeed, it was the erudition of these monks that afforded them the
opportunity to make such claims in a peculiar genre of writing -
prefaces to blood-written scriptures - that combined a reverence for both
bookish doctrine and ascetic practice.
38. "Cixue he jin Huayanjing fayuanwen" in Hanshan
[aoren mengyouji 40, p. 401b.
39. ''Ti Gongquan Chanren xueshu Fahuajing hou,"
Hanshan [aoren mengyouji 31, p. 326a.
40. "Xue shu Fahuajing ba" Ifilif7:t#Ji1WZ in Lingfeng Ouyi dashi zong [un
7.1, p. 9 in Jiaxing Da zang jing (Taipei: Xinwen
feng 1987), vo1.36, p. 372b. On Zhixu's blood writing, see CRO Sh6gon
(Shih Sheng-yen), Minmatsu Chiigoku bukkyo no kenkyii - tokuni Chigyoku 0
chiishin toshite 0 -C (Tokyo: Yama-
kiM Butsushorin 1975), p. 238.
41. "Ba Qi Chanren xueshu Huayan jing," in Zibo Zunzhe
quanji 15, p. 446b-447a in Xu zang jing, vo1.126, p. 892.
KIESCHNICK 189
A curious motif in the records of blood writing, tangentially related to
Buddhist asceticism, is filial devotion. Recall that Deqing's decision to
copy out a scripture in blood was inspired by memories of his deceased
parents. A number of the examples of blood writing I have already cited
were prompted by the death of a parent. The connection between parents
and this practice was founded on the belief that copying Buddhist books
produced merit, that copying scriptures in one's blood produced particu-
larly potent quantities of merit, and that this merit could then be used to
assist the dead. Equally important, however, was a strong tradition of
practices and stories associated with filial sons. In early China, stories of
the exploits of filial sons developed largely independent of Buddhist
influence, at least during their formative period.
42
In these stories, men
are lauded for their devotion to their parents and rewarded for their
virtues by miracles that occur spontaneously, according to the Chinese
principle of "resonance." These elements creep into Buddhist accounts of
blood writing as well. One Tang collection of miracle tales writes, for
instance, of a layman named Li Qianguan *J:ttl who, on the death of
his father, copies out in his blood the Diamond Sutra and the Heart
Sutra. Immediately thereafter, "an unusual fragrance filled the court-
yard. "43 The same text tells a similar story of a man named Qiao Qing
~ J ] J l P who, on the death of his mother, made a copy of the Diamond in
his own blood, whereupon auspicious grasses sprouted up around his
mourning hut.
44
Outside of the references to blood writing, these
accounts of filial miracles read like classic stories of devoted sons, a
tradition from which they drew.
45
Similar accounts in a range of
sources, including the dynastic histories, reveal that the practice of blood
writing extended beyond didactic stories to the lives of real people who
would copy scriptures in their blood as an expression of piety for
deceased parents. What was true for lay people was equally true for
monks. Whether in sixth-century biographies or private writings in the
seventeenth century, monks expressed feelings of remorse on the deaths
of their parents and concern that they had not been as filial as they
42. That is, during the Han and Six Dynasties period. See KNAPP, Accounts of Filial
Sons. Later accounts of filial sons do betray a Buddhist influence, particularly
with respect to self-mutilation.
43. lin' gang banruo jiyan ji B, p. 46b.
44. lin 'gang banruo jiyan ji B, p. 46b-47a.
45. In a note to the story of Qiao Qing, the compiler of the text explains that the story
appears also in accounts of filial sons.
nABS 23.2 190
should have been. This powerful combination of Buddhist notions of
merit, asceticism and filial devotion made the practice of blood writing
accessible and compelling when a parent died.
Blood
Modern attitudes towards blood are the product of a long of
development in which we have become increasingly sensitive to bodily
fluids in general. In seventeenth-century England, when James I hunted
stag, "ladies of quality" would wait anxiously for blood from the hunt to
wash themselves with, believing this would whiten their skin.46 This is
all a far cry from the customs and mores of modern England, where one
buys one's meat in neat, bloodless packets, and gladly leaves the
handling of blood to butchers and surgeons. A similar shift has occurred
across Europe and in the United States where, for instance, blood has
made a slow retreat from the dinner table.
47
These new attitudes shape
our reaction to the practice of blood writing, evoking in modern
scholars, myself included, a feeling of uneasiness and morbid fascina-
tion. In general, the same pattern of development holds true for China,
where people have become increasingly sensitive to the sight of blood.
One can take the point too far: pre-modem Chinese were far from
indifferent to blood. The eighth-century monk Jiaoran for instance,
instance, objected to blood writing, arguing that writing a holy scripture
in one's blood is wrong. because the body is a vile, unclean thing.48
Buddhist scriptures at times invoke blood along with puss, phlegm and
bile when attempting to provoke in their readers disgust for the body.49
46. Keith THOMAS: Man and the Natural World: Changing Attitudes in England
1500:1800 (Oxford: Oxford University Press 1983), p. 29 and pp. 290-291.
47. Stephen MENNELL: All Manners of Food: Eating and Taste in England and
France from the Middle Ages to the Present (Chicago: University of lllinois
Press 1996), p. 309.
48. "Tang Huzhou Dayunsi gu Chanshi Yugong bei ming bing xu"
in Quan Tangwen 918, p.4b. A disgusted ninth-century
official made the same point just before banishing a local monk for copying scrip-
tures in blood. Beimengsuoyan by Sun Guangxian $%7\:;1( (d. 968)
(Beijing: Zhonghua Shuju 1985) 9, pp. 74-75.
49. Consider for instance a passage in the Majjhima Nikiiya (10.4) that states "Again,
bhikkus, a bhikkhu reviews this same body up from the soles of the feet and
down from the top of the hair, bounded by skin, as full of many kinds of impu-
rity thus: 'In this body there are head-hairs, body-hairs, nails, teeth, skin, flesh,
sinews, bones, bone-marrow, kidneys, heart, liver, diaphragm, spleen, lungs,
KIESCHNICK 191
And Daoists as well as Buddhists at times objected to bloody sacrifices
to the gods.
50
Nevertheless, it is safe to say that pre-moderns had more flexibility in
using blood in certain prescribed contexts. We know, for instance that
blood was smeared on ritual implements in ancient China, and that
menstrual blood was used in certain medical remedies and magical
potions.
51
More important for our purposes was the use of blood for
sealing oaths. The dynastic histories are replete with references to the
practice of drinking an animal's blood (shaxue lIlXlfTI.) to seal a covenant,
whether between two rulers or between two generals. 52 Similarly, one
fifth-century Daoist text describes a covenant "sealed with the blood of a
white horse"53 In these cases, blood had a magical property, ensuring
that the oath would be kept.
More relevant still to the practice of blood writing is the practice of
using human blood to mark a vow. The Zuo zhuan, compiled in
approximately the third century B.C., recounts the story of the King of
Chu who fled his enemies, taking refuge in the kingdom of Sui. In part
out of respect for the King's brother, the Sui agreed to protect the King,
rather than turn him over to his enemies. The King then "made an
incision in the flesh over [his brother's] heart and with the blood
large intestines, small intestines, contents of the stomach, feces, bile, phlegm, pus,
blood, sweat, fat, tears, grease, spittle, snot, oil of the joints, and urine.'" Bhikkhu
NANAMOLI and Bhikkhu BODHI, trans., The Middle Length Discourses of the
Buddha, A New Translation of the Majjhima Nikaya (Boston: Wisdom
Publications 1995).
50. Terry F. KLEEMAN, "Licentious Cults and Bloody Victuals: Sacrifice,
Reciprocity, and Violence in Traditional China," Asia Major, Third Series, vol.7,
part 1 (1994), pp.185-211. See also Alan COLE, Mothers and Sons in Chinese
Buddhism (Stanford: Stanford University Press 1998), p. 228.
51. On the smearing of blood on ritual implements, see YAMADA Sumeru WEB#.iIt;,
"Kin" JJ in Yamada Sumeru chosaku sha vol.4 (Tokyo: Meiji
Shoen 1984), pp. 301-338. On menstrual blood in traditional China, see JIANG
Shaoyuan It.mJffi:, "Zhongguo ren de tiangui guan de ji fangmian" q:r
in Chenbao fukan 1926, March 8, 10 and 13.
52. On blood covenants in early China, see Mark Edward LEWIS, Sanctioned
Violence in Early China (Albany: State University of New York Press 1990),
pp.43-50.
53. Santian neijie jing =::Kfl3!yIH&1 1, HY 1196, p. 6a; translation in Stephen BOKEN-
KAMP, Early Daoist Scriptures (Berkeley: University of California Press 1997),
p.216.
JIABS 23.2 192
concluded an alliance with the men of Sui."54 The reason the King used
his brother's blood rather than his own probably relates to the particular
role his brother had played in winning over the sympathies of the Sui.
Moving forward several centuries, a Tang commentary to this same Zuo
zhuan passage notes "By drawing blood from over the heart and forging
an alliance with it, he expressed his supreme sincerity."55 The same
notion appears in a number of episodes in the Tang. In 880, in the midst
of a rebellion, the loyal minister Zheng Tian marked a pledge of
loyalty from an officer by "sealing a paCt with his blood."56 Similarly,
the late ninth-century figure Zheng Ranzhang pledged allegiance
to a bandit leader by "taking out wine, cutting his shoulder, and making
a pact with his blood."57 In a more frivolous episode, the heir apparent
under emperor Gaozu (r.618-626) agreed to give his brother a talented
court musician if his plot to usurp the throne succeeded. Resealed the
vow by "cutting his shoulder and making a blood pact."58 The Zuo
zhuan was compiled well before the entry of Buddhism to China, and
even in the later episodes there is not so much as a whiff of Buddhist
influence. Similar accounts can be gleaned from later dynastic histories
in which individuals seal various sorts of pacts or vows with their blood.
These passages reflect a separate tradition in which blood, and
particularly one's own blood, was drawn in small quantities as a sign of
sincerity when making solemn oaths.
This tradition of drawing blood in oaths of sincerity divorced from
Buddhist concerns is evident even in the writings of the monk Ouyi
Zhixu, who once composed a letter of devotion to his mother in his own
blood.5
9
In general, however, while Buddhist ideas of asceticism and
merit are absent from the accounts of princes and generals in the Zuo
zhuan and Tang History, the opposite does not hold true. That is,
Buddhist practitioners could easily accommodate the idea that drawing
one's own blood was a noble sign of sincerity and determination, in
54. Chunqiu Zuo zhuan zhengyi Duke Ding, year five, in Shi san
jing zhu shu (Beijing: Zhonghua Shuju 1979) 54, p.435; Burton
WATSON (trans.): The Tso chuan. Selections from China's Oldest Narrative
History (New York: Columbia University Press 1989) p. 185.
55. Chunqiu Zuo zhuan zhengyi 54, p. 435.
56. Xin Tangshu 185, p. 5403.
57. Xin Tangshu 224b, p. 6399.
58. Xin Tangshu 79, p. 3549.
59. Lingfeng Ouyi dashi zonglun 5.1, p. la.
KIESCHNICK 193
addition to its ascetic value, utility in acquiring merit, and resonance
with filial devotion.
Conclusion
Various factors should have worked against the development of blood
writing in China. Before the entry of Buddhism, there was no strong,
pervasive ascetic tradition in China. Later, even though influenced by
Buddhist teachings, Daoist asceticism emphasized celibacy and fasting,
and in general eschewed all forms of self-mutilation.
6o
Early notions of
filial piety encouraged sons to keep their bodies whole, while Buddhist
texts described the body as filthy and vile - hardly a fitting medium for
sacred scriptures. In the case of blood writing, however, whatever reser-
vations these factors may have produced in the minds of practicing
Buddhists, they were overwhelmed by the doctrine of merit, veneration
for physical suffering, and the powerful symbolism of blood as a mark
of sincerity. In the end, the decline of the practice of blood writing was
brought on not by traditional concerns for the preservation of the body
or Buddhist notions of the body as unclean, but rather by the introduc-
tion of new sensibilities over the past hundred years.
Modernity has been accompanied by an increasing uneasiness with
asceticism, coupled with greater squeamishness about bodily fluids.
61
The Catholic church is now reluctant to grant sainthood to holy men
who display stigmata, and accounts of the grisly ascetic practices of
Christian saints that once provoked awe and wonder, now elicit shock
and disgust.
62
What is true for the West, has been true as well in China,
though perhaps the threshold of tolerance for bloody religious practices
has lowered more slowly in China than in Europe and the United States.
Twentieth-century Chinese intellectuals like HU Shih have generally
considered ascetic practice a distasteful example of religious fanaticism,
and more recently Buddhist monastic organizations have attempted to
60. See ESKILDSEN, Asceticism in Early Taoist Religion.
61. On this point, see in particular Norbert ELIAS' riveting analysis of changing atti-
tudes towards spitting and nose-blowing in his The Civilizing Process: The
History of Manners and State Formation and Civilization (1939; Oxford:
Blackwell 1994).
62. See Kenneth L. WOODWARD, Making Saints. Inside the Vatican: Who Become
Saints, Who do not, and Why (London: Chatto & Windus 1991), particularly
Chapter 5.
JIABS 23.2 194
curb some more gruesome forms of asceticism.
63
Chinese leaders no
longer make incisions over their hearts when pledging loyalty to allies,
and few Chinese mothers would now appreciate a letter in l;>lood from a
devoted son.
While blood oaths disappeared from the Chinese political arena by the
late nineteenth century, the practice of blood writing continued arr;tong
Buddhist devotees for some time, and was still quite common in the
thirties.
64
Because blood writing was embedded in a network of associa-
tions that imbued it with an aura of tradition, it was more resistant to the
general trend of increased aversion to blood than pacts between political
leaders.65 At present, however, blood writing is quite rare in China, and
seems to have vanished completely from Taiwanese Buddhism. This is a
part of a more general trend in which Buddhist ascetic practices, includ-
ing blood writing, are on the decline as Buddhists absorb the values of a
society increasingly ill at ease with self-inflicted pain.
63. Hu Shih: "The Indianization of China"; CAl Huirning "Feizhi 'shaoxian-
ba' de xiguan hao" in Fayin $1f (1984.3), p. 43.
64. PRIP-M0LLER: Chinese Buddhist Monasteries, p. 323; WELCH: The Practice of
Chinese Buddhism, p. 323.
65. The parallel with fox -hunting in England is suggestive. While the idea of washing
themselves in blood from a hunt no longer appeals to English women, English
boys are still smeared with the blood of a fox at their fIrst hunt. Evidently because
of its status as a revered tradition, fox-hunting has remained resistant to a more
general shift in public mores away from contact with blood. See Norbert ELIAS
and Eric DUNNING, Quest for Excitement: Sport and Leisure in the Civilising
Process (Oxford: Blackwell 1986), pp.150-174.
KLAUS-DIETER MATHES
Taranatha's Presentation of trisvabhiiva
in the gtan ston sfiin po*
Abbrevations used:
MAV
MAVBh
MAVT
MSA
MSABh
RGV
RGVV
1. Abstract
Madhyantavibhaga
Madhyantavibhagabharya
Madhyantavibhagapkii
MahayanasiitralalJ'lkara

Ratnagatravibhaga
Ratnagatravibhagavyakhya
The doctrine of trisvabhiiva plays a central role in the formulation of the
gian ston ('empty of other') Madhyamaka. Normally any positive asser-
tion on the level of ultimate truth, except that all phenomena are empty
of an own-being, would not be accepted by a Madhyamaka school.
1
Taranatha (1575-1634), a follower of the gian ston exegesis of the Jo
nan pa school of Tibetan Buddhism, applies this proposition, however,
only to the apparent truth, which he equates with the imagined and the
dependent natures (parikalpita- and paratantrasvabhiiva). The ultimate
truth, or the perfect nature is empty of other
* Paper read at the Xllth Conference of the International Association of Buddhist
Studies, Lausanne 1999.
1. See WILLIAMS 1989: 62.
2. The equation of the Yogacara terms parikalpita and paratantra with apparent
truth and with the ultimate truth is quite common in the gian stan
Madhyamaka of the Jo nan pas. Contrary to parikalpita, which is merely
imagined and does not exist at all, paratantra is admitted to exist on the level of
apparent truth. exclusively exists in terms of ultimate truth. So
Taranatha says in his gZan stan siiin pa: "Even though the two, dependent and
imagined, are equal [in the sense that] they do not exist in reality, equal in being
delusive appearances, and equal in being apparent [truth] and false, they should
be distinguished in terms of their respective marks: The imagined does not exist
even [on the level of] apparent [truth] whereas the dependent does. Since
does not exist [on the level of] apparent [truth], [only] ultimately, it
exists in reality. Likewise the imagined exists as imagination, the dependent as
[mental] substance, and the perfect as something whichis [even] free from the
Journal of the International Association of Buddhist Studies
Volume 23. Number 2.2000
JIABS 23.2 196
(gian stan), that is, the imagined and the dependent, but not empty of its
own-being.
3
This presentation of trisvabhava, which is typical of the
Jonangpas, can be traced back to bTsan Kha bo che (born 1021). It
mainly follows the Brha{!fka, a commentary on some of the Prajfia-
paramitasutras.
4
Taranatha sets forth this particular Madhyamaka under-
standing of the two truths in his "Essence of the Empty of Other" (Tib.
gtan stan sfiin pa) in a short and precise way. In its essentials, the
doctrine was taken over by KOIi sprul blo gros mtha' yas
5
and is still
adhered to by modem gian stan proponents like Khenpo Tsultrim
Gyamtsho.
The gian stan sfiin pa is a typical Tibetan presentation of the four
tenets of Buddhist philosophy Sautrantika, Cittamatra and
Madhyamaka). In this important genre of Tibetan hermeneutics the four
tenets (Tib. grub mtha ') and their subdivisions are usually defined along
the lines of the Madhyamaka distinction into apparent and ultimate truth,
the Madhyamaka being considered the ultimate mode of analysis by all
Tibetan schools. What makes Taranatha's text on the four tenets special
is the fact that he divides the fourth tenet into "ordinary" and "Great
Madhyamaka" (dbu rna chen po). In the latter the two truths are
presented in a typical gian stan way by relating them to the Yogacara
concept of trisvabhava.
This particular trisvabhava theory is largely based on the two
introductory verses of the first chapter of the Madhyantavibhaga, where
mental fabrication 'it does not exist in terms of these two'" ( ... gian dban dan
kun brtags giiis bden par med miiam dan / 'khrul snan yin miiam dan / kun
rdzob dan rdzun pa yin miiam yin kyan / mtshan iiid so sor dbye dgos pa ni /
kun brtags kun rdzob tu yan med / gian dban kun rdzob tu yod pa yin cin / yons
grub ni kun rdzob tu med La don dam du yod pas bden par yod pa' 0 / de biin du
kun brtags ni brtags pas yod pa dan / gian dban ni rdzas su yod pa dan / yons
grub ni de giiis su yod pa min yan spros med du yod pa' 0, Tiiranatha: "gZan stoil
sfiiil po ces bya ba Mugs so," rIe btsun Tarantitha'i gsun 'bum biugs so (Leh:
publ. by Namgyal & Tsewang Tarn, 1982-5, Vol. 4, p. 505,11. 1 - 3).
3. "Because that wisdom, [or] true nature of phenomena, (both are equated with the
perfect nature) is established in its own right (Skt. * svabhavataft) since begin-
ningless time, and neverchanging, it is not empty of its own-being (Skt. * sva-
bhavena) and exits permanently" (ye ses chos iiid de ni ran gi no bos gdod rna
nas grub cin nam du yan 'gyur ba med pa'i phyir ran gi no bos ston pa rna yin
iin rtag tu yod pa' 0, Tiiranatha, op. cit. (see footnote no. 2), p. 504, 11. 2 - 3).
4. See STEARNS 1999: 89.
5. Cf. Ses bya kun khyab, Vol. 2, pp. 546-9.
MATHES 197
the central concepts of this Yogacara work - duality, false imagining
and emptiness - are introduced. Taranatha further refers to the three
types of emptiness presented in MSA XIV.34. In a way typical of the
gian ston tradition, the presentation of the perfect nature is then
combined with the tathagatagarbha theory of the Ratnagotravibhaga on
the grounds of an equation of purified suchness with the state of the
Tathagata in MSA IX.37. Finally, Taranatha comes to the conclusion
that distinguishing all phenomena on the basis of the three natures
amounts to the same as differentiating such phenomena under the aspect
of consCiousness on the level of apparent truth, and the aspect of wisdom
on the level of the ultimate truth.
In the following, I attempt an evaluation of Taranatha's trisvabhava-
interpretation against the background of the pertinent passages of the
Indian treatises adduced, espeCially the Madhyantavibhiiga, Mahayana-
siltrala'!lkara and Ratnagotravibhiiga.
2. The Initial Stanzas of the Madhyantavibhaga
2.1. Taranatha's interpretation
The dbu rna chen po chapter of the gian ston sfiin po starts, after a
general introduction, a second subchapter with a Citation of the first two
verses of the first chapter of the Madhyantavibhaga, which define the
right middle way in the Yogacara works of Maitreya. As an alternative
to, or better, a further development of Nagarjuna's Madhyamaka, they
playa central role for the proponents of the gian ston, distinguishing as
they do the categories of the existing and non-existing. One has to bear
in mind that the root text, which does not make much use of the
trisvabhava terms, equates the perceived object with the imagined
nature, false imagining with the dependent nature, and the absence of
duality, or emptiness, with the perfect nature (cf. MA V 1.5).
In the following, I give a translation of the initial stanzas. The
additions in brackets follow Vasubandhu's commentary.
False imagining exists. (l.la)
Duality is not found in that. (1.1 b)
But emptiness is found there, (l.lc)
[And false imagining] is found in relation to [emptiness] as well. (Ud)
Therefore everything is taught (I.2b)
As neither empty nor non-empty, (l.2a)
Because [false imagining] exists, because [duality] does not exist, and because
nABS 23.2 198
[false imagining] exists [in relation to emptiness, and emptiness in relation to
false imagiillng]. (l.2c) .
And this is the Middle Path. (I.2d)6
Taranatha starts by explaining that false imagining - being consciousness
which takes the form of a perceived object and perceiving subject - only
exists on the level of apparent truth. Duality, however, does not exist at
all, since it is a pure mental creation. Thus apparent truth is free of the
two extremes of nihilism and etemalism. The first extreme is avoided by
asserting false imagining on the level of apparent truth, the second by
negating the existence of the object-subject duality.
Emptiness, equated by Taranatha with wisdom,7 really exists as the
true nature of phenomena in false imagining.8 In a state where mental
stains still prevail, false imagining also exists in relation to the true
nature of phenomena or emptiness. It is to be understood, however, that
false imagining exists only as something (ultimately) unreal (bden med
kyi lio bor yod pa). Being consciousness which consists of accidental
stains, it must be given up eventually.9 Since the existence of emptiness
6. Cf. NAGAO 1964: 17-18: abhutaparikalpo 'sti dvayan tatra na vidyate / sunyatii
vidyate tv atra tasyiirn api sa vidyate // na sunyarrz niipi ciisunyarrz tasrniit sarvarn
vidhfyate / sattviid asattviit sattviic ca rnadhyarnii pratipac ca sii II.
7. Khenpo Tsultrim Gyamtsho explained that the equation of emptiness with
primordial wisdom follows from the fact that we are talking here about the empti-
ness or true nature of false imagining, which is mind.
8. "The wisdom of (genitive of identification?)a emptiness free of mental fabrication
really exists as the true nature of phenomena in consciousness, [i.e.] false
imagining" (spros bral ston pa flid kyi ye ses de ni / rnarn ses yan dag min rtog
de la chos flid kyi tshul du bden par yod cin / ... , Taranatha, op. cit. (see footnote
no.2), p. 503, ll. 5-6).
a. In this context Sthiramati merely explains emptiness as the dharrnatii (chos
flid) of false imagining: "False imagining is found in emptiness in the form of
phenomena (lit. "as something possessing a quality (dharmin)," i.e., possessing
dharmatii), in the sense that emptiness exists in false imagining as its true nature"
(Siinyatiiyiis tu sattvarn (text: sarvarn) abhataparikalpe taddharmateti krtvii
sunyatiiyiirn apy abhutaparikalpo dharmirupelJa vidyate, YAMAGUCHI 1934: 15,
ll. 17-9). Cf. Tib.: ston pa flid ni yan dag pa rna yin pa kun rtog pa la de'i chos
flid du yod de / ston pa flid la yan yan dag pa rna yin pa kun rtog pa chos can gyi
no bor yod do (Karmapa Tanjur, sems tsam, bi, p. 392, ll. 2-3).
9. "When stains [still] prevail, that consciousness exists in the true nature of
phenomena in terms of phenomena (lit. "something possessing a quality," i.e.,
phenomena possessing dharrnatii), accidental stains which can be separated,
which must be given up, [being] stains without real existence" (dri bcas kyi skabs
na chos flid de la / rnarn ses de chos can dri rna glo bur ba 'bral run / span bya
MATHES 199
and the non-existence of phenomena (false imagining and so forth) are
asserted on an ultimate level, the extremes of nihilism and eternalism are
avoided with regard to the ultimate truth.
Apparent truth, dualistic appearances, delusions, etc. do not exist in
their own right; hence they are empty of an own-being. The true nature
of phenomena, that is, emptiness or wisdom, exists from beginningless
time. It is never-changing and therefore not empty of an own-being.
This, according to Taranatha, does not contradict the SUtras, where it is
said that even the dharmadhtitu is empty. Being empty does not neces-
sarily entail emptiness of an own-being. It may be understood as empty
of other factors, for example, mental fabrications, which are different
from wisdom or dharmadhtitu.
What now follows is Taranatha'sdefinition of the three natures based
on what has been said:
The imagined [nature] is [like] the sky etc., [like] all non-entities. [It consists of]
all object-appearances such as: visible forms appearing to the [false] imagining,
[all] relations between names and things, [which arise by] clinging to names as
things and mistaking things for names, and [every] object grasped by a superim-
posing intellect - outside and inside, extremes and middle, big and small, good
and bad, space and time.lO
The dependent [nature] is mere consciousness, which appears as the subject-
object relationship, because it appears by being dependent on something else, viz.
the habitual imprints of ignorance.!!
The perfect [nature] is self-awareness, clarity in its own right, free from all mental
fabrications. It is synonymous with the true nature of phenomena, the sphere of
qualities (dhannadhiitu), suchness and ultimate truth.1
2
dri ma bden med kyi no bar yod .... , Taranatha, op. cit. (see footnote no. 2),
p. 503, 1. 6.
10. Taranatha, op. cit. (see footnote no. 2): ... kun brtags ni nam mkha' la sags pa
dnos med thams cad dan I rnam rtog La sar ba'i gzugs sags yuL gyi snan cha
rnams dan I min La don du ien pa dan I don La min du ' khrul pa 'i min don gyi
'brei pa dan I phyi nan mtha' dbus I che chun bzan nan phyogs dus sags bios
sgro btags kyis gzun bya thams cad do (p. 504, ll. 5-6).
11. Ibid.: gian dban ni gzun ' dzin gyi dnos par snan ba' i rnam par ses pa tsam ste I
ma rig pa 'i bag chags kyi gian dban du gyur ( text: grur) pas snan ba yin pa'i
phyir ro (p. 504, ll. 6-7).
12. Ibid.: yons grub ni ran rig ran gsal spros pa thams cad dan bral ba de yin te I
chos fiid dan chos dbyins dan I de biin fiid dan / don dam bden pa rnams min gi
rnam grans so / (p. 504, 1. 7 - p. 505,1. 1).
JIABS 23.2 200
Taranatha then elaborates the relations between these natures. Neither
the imagined nor the dependent exist in reality: they are both deceptive
appearances, apparent truths and false. They need to be distinguished,
however, in terms of their respective features: The imagined does not
even exist on the level of apparent truth, whereas the dependent does:
The imagined exists as mere imputation, the dependent as mental
substance. In a way typical of the Jonangpas, the perfect nature is taken
to not exist on the level of apparent truth, but only on the ultimate level.
Taranatha must have seen the problems of this statement and added that
the perfect nature is also without any mental fabrication, even w,ithout
the mental fabrication that the perfect nature does not exist as the
imagined or dependent natures.
2.2. The Initial Stanzas in the Light of the Commentaries by Vasu-
bandhu and Sthiramati
13
Based on the initial stanzas (see above), we can describe three philosoph-
ical propositions which undergird the entire treatise of the Madhyiinta-
vibhiiga:
(a) False imagining exists.
(b) Subject-object duality, created by false imagining, is not found in
that.
(c) False imagining is found in relation to emptiness in the sense that
emptiness is found in false imagining as its true nature.
The initial stanzas introduce at the same time the three main philosoph-
ical terms of the whole corpus - false imagining, duality and emptiness
14
- which are related to the three natures in MA V 1.5.
Vasubandhu explains false imagining in his commentary as the
construction of the perceived object and the perceiving subject. Duality
is the perceived object and the perceiving subject. Emptiness refers to
the fact that this false imagining is devoid of any subject-object relation-
ship. Taking up the canonical formula on being empty as it is found in
the Majjhimanikiiya,15 Vasubandhu then shows that he understands
13. I am particularly grateful to Prof. Schmithausen, who gave me some fruitful
suggestions for this chapter, especially with regard to the problem of the two
unbalanced trisvabhiiva models in the MAV.
14. See ECKEL 1985: 35.
15. See CHALMERS 1899; and SEYFORT RUEGG 1969: 319-320. MAVBh on I.1
runs as follows: "Thus one truly sees that something is empty of that which does
MATHES 201
empty or emptiness as the absence of something, namely duality, in
something else which exists (false imagining and emptiness). Sthiramati,
too, defines the emptiness of false imagining as the absence of duality.
He cites the common example of a rope which is falsely perceived as a
snake: the rope is there; it is merely empty of "snakehood". The
canonical formula on being empty (the absence of something in some-
thing which exists) is then explained as meaning that duality does not
exist in false imagining. What is left over are false imagining and
emptiness, both of which exist.l
6
Consequently, one is left wondering
how Taranatha can explain false imagining to be self-empty (ran stan)
and as ultimately not existing (a point we will come back to later).
Sthiramati gives four alternative commentaries directly on the initial
propositions of MAV 1.1. Summarizing these different explanations, one
can say that the first two padas serve the purpose of (i) repudiating
complete denial and wrong superimposition, (ii) establishing that all
phenomena are only mind and (iii) explaining the marks of defilement.
The explanations differ considerably in their way of defining the onto-
logical status of false imagining, which exists "in terms of own being"
expl. no. 1), "substantially" expl. no. 2), by
nature as an [ongoing] modification of consciousness (vijfiiinapari-
1}iimiitmanii, expl. no. 3), and as "defilements" (abhiltaparikalpasva-
expl. no. 4). Emptiness is described as "absence of
duality in false imagining" (griihyagriihakarahitatii 'bhiltaparikalpasya,
expl. nos. 1+2), as a "referential object [conducive] to purification"
(visuddhyiilambana, expl. no. 2), and as a "true (or existent) lack of
self' (bhiltanairiitmya, expl. no. 3). Further, it is said that its own-being
is purification (vyavadiina, expl. no. 4) and that it is by nature the non-
existence of duality.
In his commentary to the next stanza, Vasubandhu says that the proper
middle path (madhyama pratipat)17 is followed by asserting that every-
thing is neither empty nor non-empty:
not exist in it, and one truly realizes that that which remains there is present, [and]
hence exists there. The marks of emptiness are thus explained correctly" (yad
yatra nasti tat tena sunyam iti yathabhutaTJ1 samanupasyati yat punar atra-
bhavati tat sad ihastfti yathabhutaTJ1 prajanatfty aviparftaTJ1 sunyata-
Zalqal)am udbhavitaTJ1 bhavati, NAGAO 1964: 18,11.4-7).
16. NAGAO 1964: 14,11.4-14.
17. Cf. MA V V.23-26, where a number of pairs of extremes are listed to make sure
that the middle is correctly distinguished from all possible forms of dichotomiz-
JIABS 23.2 202
"Neither empty" [means not empty of] emptiness and false imagining, "nor non-
empty" [m'eans empty of] duality, that is, perceived [object] and perceiving
[subject). "Everything" [means] the conditioned, which is called false imagining,
and the non-conditioned, which is called emptiness.
IS
'
The paradoxical triple clause because it exists, it does not exist and it
exists (MA V 1.2c) is then explained as:
Because false imagining exists, because duality does not exist and because
emptiness exists in relation to false imagining and false imagining in relation to it
(i.e., emptiness).19
The third causal clause shows that the last two lines of MA V 1.1 have to
be taken together. The double locative relationship between false
imagining and emptiness, the two entities of which all phenomena are
not empty, suggests a mutual existence; and it is precisely this mutual
existence the second sattviit ("because it exists") of MA V 1.2c refers to.
In his commentary on MA V 1.2, Sthiramati clarifies the relationship
between false imagining and emptiness by pointing out that false
imagining is conditioned, because it depends on causes and conditions;
emptiness does not and is therefore a non-conditioned entity. The
conditioned is not empty, on account of the own-being of false
imagining, but it is empty of duality. False imagining is found in
relation to emptiness as something possessing a quality (i.e., phenomena
possessing dharmatii), in the sense that emptiness exists in relation to
false imagining as its true nature.
20
The relationship between emptiness and false imagining is further
elucidated in MA V I.13, where the own-being of the non-existence of
duality is taken as the mark of emptiness. The latter is described as
neither existent nor non-existent. Existence does not apply because of
the noncexistence of duality. Non-existence is not accurate either,
because of the existence of the non-existence of duality. The latter term
indicates that emptiness is not the mere non-existence of a perceived
ing concepts, similarly to what one finds, for instance, in the works of Nagfujuna
(see NAGAO 1964: 69-70).
18. MA VBh on 1.2: na siinyaTJ1 siinyataya cabhiitaparikalpena ca / na casiinyaTJ1
dvayena grahyena grahakerza ca / sarvaTJ1 saTJ1skrtaTJ1 cabhiitaparikalpakhyaTJ1 /
asaTJ1skrtaTJ1 ca siinyatakhyaTJ1 / (NAGAO 1964: 18,11. 10-12).
19. MAVBh on 1.2: sattvad abhiitaparikalpasya / asattvad dvayasya / sattvac ca
siinyataya abhiitaparikalpe / tasyaTJ1 cabhiitaparikalpasya / (NAGAO 1964: 18,11.
13-4).
20. MA Vf on 1.2 (see YAMAGUCm 1934: 15,11.12-20).
MATHES 203
object and a perceiving subject; rather, this absence constitutes a quality
in the sense of a true mode of being. This becomes even clearer in the
following, where it is concluded that emptiness is neither different from
nor the same as false imagining. It cannot be different because the true
nature or emptiness of a phenomenon, which appears as a result of false
imagining, cannot be really different from this very phenomenon. In
other words, it is due to the absence of duality that false imagining and
emptiness are not different.
21
On the other hand, under the aspect of
own-being (or the existence) of the non-existence of duality, emptiness
is not exactly the same as false imagining; otherwise it would not be the
only referential object conducive to purification.22 The point is that even
though the own-being of emptiness is defined in a negative way, it is not
understood as the mere absence of something like a hare's horn, but as
the true nature of phenomena, a general state of being)3
From what has been said above, it becomes clear that the initial
propositions (a) and (c) can be interpreted in different ways. The main
question is, in what way false imagining exists (a) and how the relation-
ship between false imagining and emptiness is to be explained (c). In
21. Cf. the Dharmadharmattivibhiiga, where dharmas (explained as false imagin-
ing) and dharmatti (=emptiness) are similarily taken as being neither different
nor the same. In the Dharmadharmattivibhtigavrtti their non-difference is
explained on the grounds that the dharmatti is constituted by the mere non-
existence of phenomena and is, as such, not different from what phenoniena as a
result of false imagining really are: non-existent. See MATHES 1996: 122.
22. By virtue of this difference, emptiness gives rise to supramundane dharmas. See
SCHMITHAUSEN 1987, Vol. I: 78.
23. Cf. MA V'f on I.l3ab: "[The word bhavain ... abhiivasya bhiivaJ:t sunyasya
(MA V I. 13ab)] is not superfluous. If only non-existence/non-entitya
of duality was taught to be the mark of emptiness, it would be understood as the
mere 'being in its own right' of the non-existence/ non-entity of duality, like the
non-existence/non-entity of a hare's hom and not - as is the case with the state
of suffering etc. - in terms of true nature (dharmatti). Therefore, it is said: 'The
non-existence/non-entity of duality is emptiness'. And this non-existence/non-
entity exists in false imagining" (Sanskrit in brackets reconstructed: [ntidhiko
dvaytibhavaJ:t] itfyati nirdisyamtine dvaytibhiivasya svti-
tantryam evtivagamyate / na duJ:tkhattidivad dharmatti-
rupatti / tasmtid evam ucyate dvaytibhavaJ:t sunyatti / tasya ctibhavasytibhata-
parikalpe bhiiva[J:t] ... , YAMAGUCm 1934: 47, 11. 6-11.)
a. The Tibetan has dlios po med pa for abhava (Karmapa Tanjur, sems tsam, Bi,
212a6-212bl).
nABS 23.2 204
Sthiramati's commentary it is possible, as we have seen, to identify at
least two interpretations within his four explanations.
The central focus of the first interpretation lies on a falae imagining
which, according to Sthiramati's first explanation, does exist ulti-
mately.24 Emptiness is then defined as the absence of duality in false
imagining. In other words, false imagining possesses or is the carrier of
emptiness, which is understood as an abstract quality.25 It is what false
imagining really is, empty of duality. Duality and emptiness are then
just two different aspects of false imagining, namely the way it appears
and the way it really is. With such an interpretation of false
and emptiness, however, a literal translation of the locative in MA V LId
(tasyiim, i.e., sunyatiiyiim) becomes problematic. Whereas the explana-
tion that false imagining exists in relation to emptiness as "something
possessing or bearing the quality [of true nature, emptiness]"
(dharmin)26 fits well the model centred on false imagining, a literal
understanding of the locative sunyatiiyiim in this sentence requires that
emptiness be taken as something which pervades all phenomena like
space (as is explained in MA V I.16c) and that it be able to exist even in
its own right. The latter quality of emptiness is also needed to explain
the fact that false imagining must be given up to attain liberation (as
indicated by Sthiramati's four explanations of MA V Ld and as directly
stated in MA V L4d
27
). In MA VBh on 1.5, where the terms of trisva-
24. Cf. MAVT 1.1, where the pada abhutaparikalpo 'sti is glossed with svabhliva-
taJ:!. A little further down Sthiramati endorses the objection of an opponent:
"[Opp.:] If thus duality was entirely non-existent like a hare's horn, and false
imagining existed ultimately in its own right, then the non-existence of emptiness
would follow. [Answer:] It is not like that, because emptiness is found there"
(Sanskrit in brackets reconstructed: [yadi eva'!! dva]ya'!! sarvathli
niisti I abhataparikalpas ca paramarthataJ:! svabhavato 'sty eva'!! sunyatiibhliva-
prasmigaJ:! (text: sunyata 'bhliva-) I naitad eva'!! yasmiic chunyatii vidyate tv
atra, YAMAGUCHI 1934: 10-1). In other words, there is nothing wrong with the
ultimate existence of false imagining, because emptiness, i.e., the absence of
duality, is found in it.
25. Cf. MAVT on 1.13ab, where dharmata, a synonym of emptiness, is compared
with the state of suffering or impermanence.
26. ... sunyatiiyiim apy abhataparikalpo dharmirupeIJa vidyate I (YAMAGUCHI 1934:
15,11. 1920).
27. This follows also from MA V 1.4: "Therefore its (=consciousness in its different
. forms, as explained in the preceding verse ) [existence] is established as false
imagining. For it is not as [it appears], nor is it completely non-existent, [since] it
is assumed that from its cessation liberation results" (abhutaparikalpatva'!!
MATHES 205
bhtiva are introduced, Vasubandhu explain:; that the three natures,
namely the perceived object, false imagining and the absence of duality
(see below), are included in false imagining, as long as the latter exists.
This may indicate that the model centred on false imagining describes
only a samsaric state of mind.
The second interpretation of the initial propositions focuses on a
space-like understanding of emptiness. False imagining, being only a
flow of modifications of consciousness (Sthiramati's third and fourth
explanation of MA V Lla), is not said to exist ultimately. Emptiness as
all-pervading true nature does not depend on false imagining and must
be more than the mere absence of duality in it. This becomes clear in
Sthiramati's fourth explanation of MAV 1.1, where he says that the
own-being of emptiness is purification because it has the own-being of
the non-existence of duality. The path and cessation are also included in
it, because they are constituted by emptiness.
28
From this one could infer
that the qualities of the path etc. exist (in the sense of gian stali) since
beginningless time, or else they could not be included in false imagining
as its true nature.
29
If the path is explained as in MA V TILll, namely as
consisting of gradually cultivated and thus conditioned qualities like
non-conceptual wisdom, it is difficult to subsume it under false
imagining30 and the non-existence of duality (=emptiness in the first
siddham asya bhavaty ata/:t I na tatha sarvatha 'bhavat (the na seems to refer to
both tatoo and sarvatOO 'bhavat, cf. Tib. I de biin rna yin ye med minI)
muktir II, NAGAO 1964: 19, 11. 5-6 & 10). Vasubandhu elaborates:
"because its (i.e., false imagining's) existence is not the way it appears. And it is
[also] not completely non-existent, since it arises as mere deception. For what
reason, again, is its mere non-existence not asserted? Because it is assumed that
liberation results from its cessation. Otherwise, neither bondage [in samsara] (Le.,
defilements) nor liberation (Le., purification) would be brought about." (yasman
na tatoo 'sya bOOvo yatOO pratibOOsa utpadyate I na ca sarvatOO 'bOOvo bhranti-
matrasyotpadat I kimarthaT(! punas tasyabhava eva I yasmat I
muktir II anyatOO na bandho na prasidhyed ... ,NAGAO 1964: 19,
11.7-12).
28. MA VT on 1.1 (Sanskrit in brackets reconstructed): [siinyatasvabhiivo hi
vyavadanaT(! dva]yabhavasvabhavatviit I atra ca siinyatiiprabhavitatvad marga-
nirodhayor api grahaIJaT(! veditavyam I (YAMAGUCm 1934: 13,11.9-11).
29. Cf. MAVT on III.22b-d, where the path is considered to be "non-conditioned"
because it is constituted by emptiness and not conditioned by karmaklea defile-
ments.
30. Cf. Vasubandhu's commentary on the transformation of the basis (asraya-
parivrtti) in the Dharmadharmatavibhaga, where the samsaric state of mind is
nABS 23.2 206
interpretation) alike.
3
! Problematic in this model is therefore the expla-
nation that false imagining is something which possesses emptiness
(endowed with the path etc.) as its true nature.
In the MA VT on LId (fourth explanation) Sthiramati explains that the
locative atra in MA V LIe (siinyatli vidyate tv atra) is used to express the
notion that emptiness is not something altogether different from one's
mind-stream of false imagining. One rather proceeds along the path
from the side of defilements (false imagining) to that of purification
(non-conceptual wisdom).
described as false imagining, and the remedy as non-conceptual wisdom. Without
a transformation of the basis, the root text goes on to say, there would be no
underlying support for designating a person passed into nirvana. Vasubandhu
explains that it cannot be the antecedent (still samsaric) states of mind, because
the cessation of that which is opposed to liberation (i.e., false imagining
according to MA V lAd) coincides with the arising of non-conceptual wisdom
(remedy). False imagining and non-conceptual wisdom cannot occur at the same
time, because two opposite processes do not have the same basis. This means that
they cannot occur in the same citta (see MATHES 1996: 151).
Against this background, URBAN and GRIFFITHS' (1994: 13) statement that
"the MV-! 1.1.1. (=MA VT on I. La) makes this distinction (false imagining as an
undefiled nonerroneous flow of experience and as a defiled and mistaken set of
percepts), identifying abhiitaparikalpa with error (bhranti) when it is defiled
and with emptiness (siinyata) when it is not" seems problematic to
me. In MA VT on 1.1 the own-being of false imagining is defined as defilements
(saTJ1kleSa) because its defining mark is error. And the own-being of emptiness is
taken to be purification, which includes the path and cessation, on the grounds
that these are constituted by emptiness (cf. YAMAGUCHI 1934: 13,11.2-3 & 9-
11). Since the path is included in the perfect nature on account of its being an
unmistaken perception of reality (i.e., non conceptual wisdom), it is difficult to
see how it can be called abhiitaparikalpa.
a. Even though the exact location is not given, this can refer only to Sthiramati's
fourth explanation of MA V 1.1. .
31. Cf. MA VT on ill.11, in which conditioned dharmas called the path are included
in the perfect nature (defined as the non-existence of duality in MA V 1.5) on
account of their being correct perceptions of the ultimate truth. Sthiramati justifies
this by pointing out that "the non existence of duality in [the dependent nature] is
exclusively the perfect [nature]. But this does not rule out the possibility that [the
perfect] is [also] something else than that, on the [wrong] assumption that (iti) the
perfect [nature] is exclusively the non-existence of dUality." (Sanskrit in brackets
reconstructed: ... tatra eva I na tu dvayabhava eva
iti YAMAGUCm 1934: 126,11.18-9). Trans-
lated into the language of mathematics, this means, "non-existence of duality" is a
proper subset of "perfect nature".
MATHES 207
In this context, it is interesting to note t.hat in Sthiramati' s four
different commentaries on MA V 1.1. emptiness is only taken as the
absence of duality in false imagining, when a more ontological status
(expl. 1 and 2) of the latter is emphasized.3
2
When false imagining is
understood in a more epistemological sense (expl. 3 and 4), however,
the absence of duality (emptiness) is also described as the "real lack of
self" (bhiitanairlitmya) or as being constituted by the path and cessation.
This positive connotation of emptiness also explains why MAV I.1c is
not a redundant repetition of MA V I.1 b.
33
In fact, emptiness is not only
defined as "the non-existence of duality", but also as "the existence of
this non-existence" in MAV I. 13ab. That emptiness is also taken as
something more than the mere absence of duality becomes clear in the
presentation of the sixteen types of emptiness in the second part of the
first chapter
34
and MA V I.22c, where emptiness is equated with the
luminous nature of mind (see also 3.2. further down).
2.3. The Doctrine of Trisvabhiiva in the Madhyantavibhaga and its
Commentaries
Faithful to the text they are commenting on, neither Vasubandhu nor
Sthiramati translate the initial passage (MA V I.1-2) into the terms for
trisvabhliva. We have to tum our attention therefore to MAV 1.5, where
the root text itself introduces the three natures.
MAV 1.3-4 is a comment on the initial passage. False imagining is
specified as a consciousness that takes the form of objects, living beings,
a personal self and perception. In a world outside of consciousness,
however, there are no such objects. For this reason, a perceiving subject
does not exist either. Nevertheless, consciousness is established as false
imagining, and its cessation is taken to be liberation. Only after this
32. In the second explanation emptiness is also taken to be the referential object
conducive to purification.
33. If emptiness was only the absence of duality, MA V I.1c would merely repeat that
duality is not found in false imagining.
34. In MA V 1.20 the last of the sixteen types of emptiness, i.e., the emptiness whose
own-being is non-existence (abhavasvabhavasunyatil)a is explained as being
different from the other fifteen (the fifteenth being a summary of the first four-
teen), since they describe the non existence of a perceiving person and perceived
phenomena (i.e., duality).
a. See Vasubandhu's b h a ~ y a on MAV I 16c-20d (NAGAO 1964: 24-26). The
root text is, as always, very cryptic, but doubtlessly confirms the above observa-
tion.
nABS 23.2 208
elaboration of the initial propositions are the terms for trisvabhiiva -
imagined, dependent and perfect (parikalpita, paratantra, parini/fpanna)
- introduced, in order to show that they correspond to what was stated
above:
The imagined, dependent and perfect [natures] are taught because of the
[imagined] object, false imagining and the non-existence of duality (MA V 1.5).35
If one relates the central terms emptiness, which is defined as (i) the
non-existence of duality and (ii) the existence of this non-existence (cf.
MAV I.13ab), to the perfect nature, and false imagining to the depen-
dent nature, the entire treatise can be understood in terms of the doctrine
of trisvabhiiva without the help of a commentary. Especially in the third
chapter on reality, the three natures are directly implied. In the first part
of this chapter, ten aspects of reality are introduced. They reflect certain
older concepts relating to truth/reality, such as the four noble truths of
early Buddhism or the apparent and ultimate truths of the Madhyarnikas.
Each of them is explained in terms of three aspects. This exposition
implies the three natures so clearly that it seems to be a mere question of
style that they are not mentioned expressis verbis throughout.3
6
The ten
aspects of reality are thus an attempt to explain older concepts ontruthl
reality from the point of view of the three natures, and thereby to show
the continuity between mainstream Buddhist thought and Yogacara.
Important for us is that the MA V clearly relates the two truths of the
Madhyamikas to the trisvabhiiva, explaining that only the perfect nature
is a fit candidate for the ultimate truth (see below). If one applies this to
the definition of madhyamii pratipat in the initial two stanzas, it would
be safe to say that the non-existence of duality and the existence of false
imagining refer to the level of apparent truth, and that the mutual exis-
tence or false imagining and emptiness defines the relation of apparent
and ultimate truth. This, in fact, supports interpretation of
trisvabhiiva, that is, Yogacara-Svatantrika-Madhyamaka.
35. MA V 1.5: kalpitaJ:t paratantras ca eva ca I arthlid abhatakalpiic ca
dvayiibhliviic ca deSitaJ:t (NAGAO 1964: 19).
36. The three natures are clearly mentioned at the beginning of the presentation of the
ten aspects of reality in MA V III.3: "The threefold nature: permanently non-
existent, existent, but not in reality, and in reality [both] existent and non-existent
are asserted as the three natures" (svabhlivas trividhaJ:t ... asac ca nityalJ'! sac ciipi
atattvataJ:t I sadasattattvatas ceti svabhlivatraya see NAGAO 1964: 37-8).
MATHES 209
Given the fact that the author of the extremely cryptic root text
obviously tried to avoid the terms for trisvabhiiva in the first place,
however, it is quite a difficult task to come up with a consistent theory
of it. The problem is that terms which are loosely related to each other
are not always' completely identical. The perfect nature, for example, is
related to emptiness only through the aspect of the non-existence of
duality. On the other hand, the second aspect of emptiness (existence of
the non-existence of duality) can only be accommodated in the perfect
nature.
In his commentary on the stanza which introduces the trisvabhiiva
terms (MA V 1.5), Vasubandhu says that false imagining includes the
three natures, if mere false imagining exists. This is an interesting point
because the previous stanza concludes by stating that false imagining has
to disappear completely in order for one to obtain liberation (MA V
l.4d). In other words, the three natures are included in false imagining
only in a samsaric state. This may be also the reason why it is difficult
to find room for the pure dependent nature (the non-conceptual wisdom
etc., of MA V IIUI) within this presentation: in a samsaric state of mind
the path has not been cultivated yet. Vasubandhu equates the imagined
nature with the (perceived) object, false imagining with the dependent
nature and the non-existence of a perceived object and perceiving subject
with the perfect nature.
Sthiramati does not elaborate on Vasubandhu's locativus absolutus
(abhataparikalpamiitre sati) and the probably intended restriction of the
presented trisvabhiiva model to a description of the samsaric state of
mind. The imagined nature is explained as being entities like visible
forms, sense faculties, a personal self and perceptions. As such, that is,
as they appear to be, they do not exist. False imagining is taken to be the
dependent nature on the grounds that it depends on causes and condi-
tions. The perfect nature is defined as the absence of duality in false
imagining. Sthiramati concludes by stating that the three natures are not
different entities, but rather different aspects of false imagining. The
aspect of false imagining which must be known is the imagined nature,
the aspect to be known and given up (defilements) is the dependent
nature, and the aspect to be known and actualized (its being free of
duality) is the perfect nature.
The perfect nature "as a perfection in the sense of being an unmistaken
[perception of the ultimate reality]" occurs in
the root text and Vasubandhu's commentary only in the third chapter on
nABS 23.2 210
reality, when ultimate truth is explained in terms of trisvabhiiva (MA V
IILlI). In fact, this perfect nature is the path, described in MAV III.ll
as conditioned and thus as a kind of "pure dependent nature". As has
been already observed above, it does not really fit the trisvabhiiva model
of MAV I.5. The reason for Sthiramati having introduced the two
meanings of the perfect nature already in his commentary on MA V I.5
may be that he saw and tried to remove tensions resulting from an un-
balanced presentation of trisvabhiiva. Also in MA VT on 1.1. he presents
different explanations of the initial propositions as if they were com-
patible alternatives. The difference between the first two and the l ~ s t ex-
planation reflects the same tension between MA V I.5 and MA V III. 1 1.
Sthiramati's fourth explanation (MA VT on Ll) of false imagining as
defilements, and of emptiness (which includes the path and cessation) as
purification, corresponds to the presentation of the path and the ultimate
truth in the third chapter: the dependent nature has to be finally given up
(MA V III.9c) and only the perfect nature is a fit candidate for the ulti-
mate truth (MAV IILlOd). Furthermore, in the presentation of the ulti-
mate truth as the pure object, it is said that the dependent (equated with
false imagining)37 cannot be any such pure object, since it occurs
together with stains. Only the perfect nature - being the exclusive candi-
date for ultimate truth - qualifies as a pure object of wisdom (MA VT on
IILlOb-d and III.I2cd). In the MAVT's presentation of three types of
emptiness (MA V III.6cd), Sthiramati says that the dependent nature is
not completely non-existent. It exists in the way it is perceived by pure
"mundane wisdom" (laukikajiiiina), namely wisdom acquired after
meditation.
The tension between two different trisvabhiiva models is also evident
in Sthiramati's commentary on the third initial proposition: False
imagining is found in relation to emptiness as "something possessing the
quality [of true nature]" (dharmin), in that emptiness exists in relation to
false imagining as its true nature (dharmatii). According to the model
focusing on the dependent nature, which is defined as false imagining in
MA V I.5, the phenomena of the dependent nature possess or bear
(dharmin) the dharmatii. The latter is an abstract quality of the depen-
dent, namely its being free of duality. This explanation fits the trisva-
37. In MA VT on llI.12cd it is explained that the dependent and the imagined can not
be pure referential objects of wisdom on the grounds that false imagining is
endowed with stains and that the imagined does not exist.
MATHES 211
bhtiva model of MAV 1.5 well. As already shown above, the perfect
nature, or emptiness, is held in different parts of the MA V to possess
qualities other than the mere absence of duality. As the luminous nature
of mind (MAV L22c) or natural emptiness (MA V IIL6d), the perfect
nature can exist on its own, and from the point of view of the path, the
dependent, or false imagining, even has to be given up. The perfect
nature pervades the dependent nature as its true nature (dharmata) like
space. In his fourth explanation on MA V I.ld, Sthiramati says that the
path, that is, mental factors like non-conceptual wisdom, is itself consti-
tuted by emptiness. If these factors are then also contained in false
imagining as its true nature, as MA V LIe would suggest, they must exist
since beginningless time as a part of the non-conditioned perfect nature.
This would contradict MAV III. 11, however, where it is said that
conditioned dharmas, called the path, are included in the perfect nature
on account of their being correct perceptions of the ultimate truth. On
the other hand, Sthiramati says in his MA VT on III.22b-d, where the
meaning of the conditioned and non-conditioned dharmas is explained,
that the truth of the path can be called conditioned and non-conditioned
at the same time. It has to be brought out, but is nevertheless constituted
by the non-conditioned and not caused by the karmakleia defilements. In
other words, the dharmas of the path are not caused like ordinary
phenomena, but arise from a correct and non-conceptual perception of
emptiness.
It would to be a difficult task to construct a consistent theory of
trisvabhava based on what has been said up to now. Right from the
beginning,38 where the initial propositions can be interpreted in different
ways, two trisvabhava models can be identified. They come close to
what SPONBERG (1981: 99) calls the pivotal and progressive exegetical
models of trisvabhava.
The model centred on the dependent nature (false imagining), or the
pivotal trisvabhava model, starts from an ultimately existing or all-
inclusive dependent nature. The imagined and perfect natures are just the
way the dependent nature appears to be and really is. The dependent
nature is thus a receptacle of the perfect nature, which is understood as
something abstract, like the state of suffering or impermanence. When
the dependent nature is purified it stops being false imagining and mani-
fests in itself, and thus in everything, the perfect nature. The latter
38. Based on the trisvabhiiva definitions in MA V 1.5.
nABS 23.2 212
becomes then, a kind of "purified dependent nature", which in this
model can be only thought of as the conditioned dharmas called the path
- such things as non-conceptual wisdom, which is included iq the perfect
nature on account of its being a correct perception of ultimate reality.
The ordinary world, that is, the defiled dependent nature, still remains,
of course, in the non-enlightened mind-streams of other sentient beings.
In the progressive model, the focus lies more on an emptiness which
pervades all phenomena of the dependent nature like space. This all-
pervading emptiness possesses positive qualities and can exist, contrary
to the pivotal model, in its own right, without the stream of false
imagining or the dependent nature. The dharmas of the path participate
in the non-conditioned perfect nature and are only caused in the sense
that they have to be brought out. Still, a "pure" dependent nature seems
to be necessary to explain where the manifestation of the perfect nature
can occur. The three natures represent three levels, each revealing a
progressively deeper degree of reality.
These incompatible passages in the Madhyantavibhaga probably
reflect, as I have already pointed out in a paper read at the IATSin
Bloomington,39 different strands of thought not yet completely harmo-
nized. This is fairly typical of the early Yogacara school, which not only
incorporates early Mahayana thought, but also has a rich background of
Abhidharma analysis. The commentaries' uncertainty about the onto-
logical status of false imagining may thus reflect the Abhidharmic back-
ground of this early Yogacara material. Thus, it is generally asserted in
the Hlnayana schools that conditioned, dependently arising entities really
exist. Such a stance would of course be incompatible with a Madhya-
maka understanding of the Y ogacara, which is at least attempted in some
passages.
It is not surprising, then, that Tibetan schools had different views on
these texts and interpreted them in the context of their own hermeneu-
tical systems.
39. "Taranatha's 'Twenty-one Differences in respect to the Profound Meaning': A
Possible Starting-Point for Studies in the gZhan stong Madhyamaka," (to be
published in the Proceedings oj the International Associationjor Tibetan Studies,
Bloomington 1998).
MATHES 213
2.4. Conclusion
Comparing Taranatha's presentation of trisvabhava with what has been
observed in the Madhyantavibhaga and its commentaries, one first of all
wonders how, based on MA V 1.1-2, the dependent nature can be
explained as self-empty (ran ston) and as existing only on the level of
apparent truth. This seems to be at odds with Vasubandhu's and Sthira-
mati's commentary, where the initial propositions of the treatise are
explained by recourse to the canonical formula for the mode of empti-
ness (i.e., the absence of something in something which exists). Even
though this fonnula supports a distinction between self-empty and empty
of other, one would prefer to say that - contrary to the Jonangpa posi-
tion - the dependent is empty of other, namely the imagined. Another
proponent of the gian ston, gSer mdog pal). chen Sakya mchog ldan
(1428-1507), sees this point and contradicts what is obviously the stance
of the Jonangpas by explaining that the basis which is empty of some-
thing other (ston pa'i gii) is the dependent, the negandum the duality of
the imagined, and the "actual entity of emptiness" (ston pa'i dnos po)
non-dual wisdom.
4o
The problem with Vasubandhu's interpretation of the initial stanzas,
however, is not only, that an existing false imagining that is left over in
emptiness has to be completely given up, but also that false imagining is
equated with the dependent nature, and that the pure dependent factors
of MA V III. 11 , such as nirvikaZpajnana, which are cultivated on the
path, must also be accommodated within it. It is inconceivable, for
example, how false imagining and nirvikaZpajnana can simultaneously
exist within the same citta. Vasubandhu excludes such a situation in his
commentary on the Dharmadharmatiivibhaga, where the disadvantages
of not postulating an asrayaparivrtti are elaborated. If Vasubandhu's
comment on the initial stanza describes, as indicated above, only a
samsaric state of mind, however, there is no problem at all: in a samsaric
state of mind the pure factors of the path have not been cultivated yet!
Based on this and the restriction of ultimate truth to the perfect nature in
MA V lILlO, it makes sense to follow Taranatha and see in the fIrst two
padas of the initial madhyama pratipat definition only a description of
apparent truth.
40. gSer mdog PaI.1 chen Siikya mchog ldan, "Zab zi spros bral gyi Mad pa ston iUd
bdud rtsi'i lam po che zes bya ba bzugs so," p.1l7. See also the discussion of
Siikya mchog ldan's position, MATHES in print.
JIABS 23.2 214
Whereas Taranatha's commentary on the fIrst part of the initial stanza
reflects a reasonable and possible understanding of the whole treatise, his
second step of equating emptiness with primordial wisdom is more diffi-
cult to follow. There is not a single passage in the Madhyiintavibhiiga
and its commentaries to support this notion. On the other hand, empti-
ness is not only taken as the non-existence of duality in false imagining,
but also as "the existence of this non-existence" in MAV I. 13ab. That
emptiness is here something more than the mere absence of duality
becomes clear in the presentation of the sixteen types of emptiness in the
second part of the first chapter and in MA V I.22c, where it is equated
with the luminous nature of mind. As we have seen above, Sthiramati
knows also of a commentary on the initial stanza where emptiness is
explained as being constituted by the path and cessation. From this one
could infer that the qualities of the path etc. exist (in the sense of gZCln
stan) since beginningless time, or else they could not be included in false
imagining as its true nature. Constituting the ultimate truth, they are in
reality beyond space and time and independent of false imagining. And
since it is possible to transcend space and time at any place and at any
time, each instant of false imagining possesses (dharmin) emptiness and
its qualities as true nature. Thus Taranatha restricts the perfect nature to
its unchangeable aspect by saying that the perfect in terms of being
unmistaken actually belongs to the pure dependent nature. Wisdom
arises in this interpretation only in the sense that it becomes manifest in
a mind-stream while one is removing the hindering defilements. To
some extent this is supported by Sthiramati's commentary on the initial
propositions (MA V'f on I.1, 4th expl.) and on the meaning of the
conditioned and non-conditioned in MA VT on Ill.22b-d, where it is said
that the path is non-conditioned, being constituted by the non-
conditioned, that is, emptiness, and not fabricated by ordinary karma-
klea defilements. It is also conditioned, because it has to be brought
out. In MAV III.ll, however, the path is only defined as being
conditioned.
To sum up, it has become clear that Taranatha's interpretation of the
madhyamii pratipat passage of the Madhyiintavibhiiga profits from
tensions between two trisvabhiiva models.
MATHES 215
3. The Three Types of Emptiness and Absence of Own-Being
3.1. Taranatha' s Interpretation
Having elaborated his initial description of trisvabhava on the basis of
the Madhyantavibhaga, Taranatha goes on to quote MSA XIV.34,
where three types of emptiness are mentioned: the emptiness of non-
existence, the emptiness of phenomena not exactly existing as they
appear to be and natural emptiness.
41
According to Vasubandhu these
three correspond to the three natures. It should be noted that we have the
same threefold classification of emptiness also in MA V III.6cd,42 where
emptiness, as one of the four traditional antidotes to the four mistaken
views, is explained in terms of the trisvabhava doctrine.
43
The three
natures are also explained in relation to the three types of n*svabhavata
(i.e., absence of own-being in terms of phenomenal attributes, the
absence of own-being in terms of arising and the ultimate absence of
own-being).44 According to the tradition of the Saf!ldhinirmocanasiitra,
which explains three types of nil:zsvabhavata, all knowable objects are
pervaded by emptiness and by the quality "absence of own-being," and
therefore the gian ston tradition asserts that all phenomena are empty
and non-empty. Thus everything lacks an own-being in terms of
phenomenal attributes and arising, which is in tune with the normal
Madhyamaka understanding of emptiness (ran ston). The ultimate
absence of own-being (paramarthan*svabhavata), however, constitutes
a positive quality, a kind of all-pervading true nature.
Khenpo Tsultrim Gyamtsho, a modem proponent of the gian ston and
follower of Taranatha, compares the paramarthanil:zsvabhavata with the
41. MSA XN.34: "[First] one realizes the emptiness of non-existence and then the
emptiness of not [exactly] existing as [what the forms of consciousness appear to
be], [but only] after having realized natural emptiness is one considered to be
someone who realizes emptiness" (abhtivasiinyatti1J1 jfititvti tathtibhtivasya
siinyatti1J1/pralq'tyti siinyatti1J1jfititvti siinyajfia iti kathyate 1/, S. LEVI 1907: 94).
42. Cf. MA V IIl.6cd: abhtivas ctipi atadbhtival; prakrtil; siinyatti matti (NAGAO
1964: 39). Emptiness of non-existence means that the imagined does not exist at
all. Emptiness of consciousness not existing as it appears means that the depen-
dent is not completely non existent, however, it does not exist as it is imagined.
Natural emptiness is related to the perfect, the own-being of emptiness.
43. A full correspondence is drawn, of course, only in Vasubandhu's b h t i ~ a .
44. This is in accordance with the Sa1J1dhinirmocanasiitra (see 1. POWERS 1995:
98-105).
JIABS 23.2 216
genuine realization that "the true nature of thoughts, [such as worries
that] there might be difficulties in [one's] future, is [nothing else than]
the luminous nature [of mind], which is really free of mental fabrica-
tions .. "45 In other words, the third niJ:!svabhilvatil refers to the fact that
the ultimate luminous nature of mind is really free or empty of every-
thing (e.g., mental fabrications) other than its own-being (gian stan).
This contradicts the so-called ran stan tradition which has it that
everything - including the Buddha qualities - is empty of any kind of
own-being or inherent existence. Pure ran stan, however, according' to
Taranatha's understanding of what the Buddha taught in the
turning of the wheel of the dharma, means that only the apparent truth is
empty of an own-being (ran stan). And since the three types of absence
of own-being refer to this very emptiness, they teach correct ran stan
and are in accord with gian stan. The truly existent perfect nature which
is expressed by the paramilrthaniJ:!svabhilvatil does not contradict the
Madhyamaka, since it is not created, does not become extinct, does not
abide, neither comes nor goes, and transcends space and time.
3.2. The Three Types of Absence of Own-Being in the SaIp.dhinir-
mocanasfitra and the Three Types of Emptiness in the Madhyanta-
vibhaga and Mahayanasiitralrupkara
According to Sa'!1dhinirmacanasiitra VII.7,46 the ultimate absence of
own-being (paramilrthaniJ:!svabhilvatil) is entirely different from the
first two, the niJ:!svabhilvatil of phenomenal attributes and
arising (utpatti) - which means, different from the normal Madhyamaka
understanding, that phenomena and their arising lack an own-being. The'
ultimate absence of own-being, on the other hand, is taken to be con-
stitutedby the fact that everything lacks a true self (dharmanairiltmya)-
which is considered to be an all-pervasive positive quality.
A similar notion is attributed to the natural emptiness (prak.rtisiinyatil)
of the Y ogacara, which is contrasted with an emptiness of non-existence
and an emptiness of phenomena not exactly existing as they appear. In
MSABh XIV.34 natural emptiness is explained as the own-being of
45. "rna 'OIlS pa la dka' las khag po yad pa'i rnarn rtag gi gnas lugs dlias gnas
spras bral gyi 'ad gsaf' - Khenpo Tsultrim Gyamtsho's joyful comment on
while explaining the gian stali sfiili po passage
quoted by Kon sprul blo gros mtha' yas in Ses bya kun khyab, Vol. 2, p. 549.
The teachings were given at Pullahari Monastery, Nepal, in January 1999.
46. LAMOTIE 1935: 69.
MATHES 217
emptiness (sunyatiisvabhiiva) and in MA Yf III.6cd as the own-being of
the non-existence of duality. That natural emptiness, which is at times
also expressed as dvayiibhiivasvabhiivasunyatii or abhiivasvabhiiva-
sanyatii in the MA V, does have a positive connotation, becomes particu-
larly clear in the presentation of the sixteen types of emptiness in MA V
1.17 -20. In MA V 1.20 the last two, abhiivasunyatii and abhiivasvabhiiva-
sunyatii are clearly distinguished from each other. Whereas abhiiva-
sunyatii merely summarizes the first fourteen types of emptiness, that is,
the non-existence of a personal self and phenomena, abhiivasvabhiiva-
sunyatii is something entirely different: it is the existence of the non-
existence of a personal self and phenomena.
47
Thus also in the MSA and
the MA V the existence of an absence is seen as something different from
just the absence of something.
48
3.3. Conclusion
The three different types of emptiness or absence of own-being
(niftsvabhiivatii) provide a strong argument in favour of a ran ston /
gian ston distinction. Whereas the first two types of emptiness, the ones
in relation to the imagined and the dependent natures, define the absence
of own-being in line with a normal ran ston understanding of emptiness,
the natural emptiness or paramiirthaniftsvabhiivatii leaves ample space
fora gian ston interpretation of the ultimate. One has to keep in mind,
however, that such an understanding requires adherence to Tar-anatha's
trisvabhiiva theory. Moreover, the explanations of natural emptiness etc.
do not imply anything further than that the existence of the absence of
duality is something more than the absence itself (see above). It is only
Tar-anatha's combination of these Yogacar-a elements with the tathiigata-
garbha of the Ratnagotravibhiiga that fully underpins gian ston.
47. See MATHES 1998: 462-3.
48. One could compare this with the law of gravity in physics, where the concrete
attraction of two masses and the fact that masses always follow the law of gravity
in the whole universe are two different things entirely. Whereas the first element
represents a concrete example, a concrete object of negation or (in the example
borrowed from physics) two concrete masses, the latter refers to the very validity
of a law or fact, be it the all-pervasive quality "absence of an own being" or the
universality of the law of gravitation.
nABS 23.2 218
4. The Combination of the tlisvabhiiva with the tathagatagarbha
In a following subchapter Taranatha explains the uncommon meaning of
gian ston generated by combining his doctrine of trisvabhiiva with the
tathiigatagarbha theory. In order to do so, he quotes MSA IX.37:
Even though suchness is undifferentiated in all [living beings], in its purified
form it is the state of the Tathagata. Therefore all living beings have the
seed I nature (garbha) of the [Tathagata].49
It is this same Mahiiyiinasiitriila1pkiira, however, which distinguishes in
its third chapter sentient beings according their basic spiritual potential
(gotra). In MSA III.ll it is said that there is a group of persons with no
potential at all (agotra), and among those there are some who will not
attain nirvana for some time, and some who will not attain it at all
(tatkiiliiparinirviifladharmii atyanta'!l ca). This gotra theory, which is
quite common in the Y ogacara, is in sharp contrast with the statement in
MSA IX.37. There are only two solutions. One either takes mainstream
Y ogacara literally and declares the tathiigatagarbha stanza as a teaching
with a provisional meaning within the Yogacara, or one sticks to the
tathiigatagarbha theory and explains away the limitations of the agotra-
stha.
50
Taranatha chose to do the latter. This fits his trisvabhiiva model
well: the different gotras, which function as a cause (hetu), or better, a
basis for a spiritual career either as a Sravaka, Pratyekabuddha or Bodhi-
sattva, belong to the dependent nature and are thus restlicted to the level
of apparent truth. The omnipresent perfect nature, which is constituted
by Buddha qualities since beginningless time, pervades even those
without a gotra. It is thus only a question of time till they become aware
of their innate qualities, even if they have no gotra for the time being.
Taranatha's hermeneutics have thus reached the heights of the Ratna-
gotravibhiiga where suchness endowed with inseparable qualities is
hindered only by exterior accidental stains. Suchness, as we have seen
above, is to be understood as a synonym for the perfect nature. And in
order to show that it is endowed with the immeasurable Buddha
qualities, he quotes RGV II.5:
49. MSA IX.37: tathatti suddhim tigata I tathagatatvaJ'!! tasmtic
ca tadgarbhal:z sarvadehinal:z II (LEVI 1907: 40).
50. See SEYFORTRUEGG 1969: 82.
MATHES 219
[The state of Buddha] is endowed with Buddha qualities, which surpass the
number of sand grains in the banks of the Ganga. [These qualities] all radiate
light, are uncreated and occur inseparably [from the true nature].51
Thus all the imcreated qualities of the ultimate Buddha exist, and the
dharmadhiitu wisdom belongs exclusively to the category of the ultimate
truth. In view of their having existed since beginningless t i m ~ , the other
four wisdoms (mirror-like wisdom, wisdom of equality, discriminating
and all-accomplishing wisdoms) are mainly part of the ultimate. But
given that they are attained through meditation, these four wisdoms are
also involved to a small extent in apparent truth.
52
This holds true also
for the ten powers and the four types of fearlessness. The qualities of the
form-bodies and speech of the Buddha pertain in equal part to both
truths. Accordingly the sViibhiivikakiiya (equivalent of dharmadhiitu
wisdom) belongs only to the ultimate, the dharmakiiya mainly to the
ultimate, and the sa'!lbhoga- and nirmii1}akiiya - if one does not distin-
guish what is real and imputed - in equal parts to both truths. Insofar as
it appears to others, Buddha activity participates in the apparent truth.
Nevertheless, since wisdom, its supporting force, is ultimate, the bodies,
wisdom, qualities and activity of the Buddha are contained in the
ultimate and have existed since beginningless time.
Interesting for us is Taranatha's final summary of the three natures.
He starts off by saying that the imagined nature is usually differentiated
into the imagined of the perceived object and the imagined of the
perceiving subject. The dependent is distinguished into impure and pure,
and the perfect into an unchangeable perfect nature and the perfect
nature constituted by unmistaken perception. In fact, the imagined
nature is only the perceived object, whereas the real perfect nature is the
unchangeable one. The perfect nature constituted by unmistaken percep-
tion is included under the pure dependent nature. The imagined nature
of the perceiving subject is by nature the same as the dependent. Upon
careful analysis, therefore, the dependent must be included under the
imagined, and since its true nature is the perfect, all phenomena are
included under the imagined and the perfect. Thus, distinguishing all
phenomena on the basis of the three natures amounts to the same as
51. RGV II.5: ganglitfrarajo '!ftair buddhadharmaiJ:t prabhiisvaraiJ:t / sarvair akrta-
kair yuktam avinirbhiigavrttibhiJ:t II (JOHNSTON 1950: 80).
52. This means that when the accidental stains are removed on the level of apparent
truth, the four types of wisdom are shining forth, like a crystal whose surface has
been cleaned (oral explanation of Thrangu Rinpoche, Bauddha, Kathmandu).
nABS 23.2 220
differentiating such phenomena under the aspect of consciousness on the
level of apparent truth, and the aspect of wisdom on the level of ultimate
truth. The consciousness of visible form etc. does not really exist; its
true nature, however, does.
5. Final Conclusion
Taranatha's abandoning of the dependent nature and restriction of the
perfect nature to its unchangeable aspect entails a sharp distinction
between the perfect and imagined, which is hardly compatible with the
main parts of the Madhyantavibhaga. The dependent nature undou,btedly
plays a central role as an inexpressible reality, which can be either
misunderstood, becoming the imagined nature, or realized just as it is,
abiding as the perfect nature. This explanation, however, is not in
harmony with all strands of the treatise. And it is from these tensions
that the Jonangpas profit, in that they offer a creative interpretation
fitting their gian ston view.
Taranatha's particular understanding of the Madhyiintavibhaga is also
a prerequisite to his using the three types of emptiness of MSA XIV.34
and the three niJ:tsvabhavata of the Sa'!ldhinirmocanasiitra towards his
end, and reading his ran stan / gian stan distinction of emptiness into it.
To complete this move, Taranatha equates the perfect nature of the
Yogacara with the tathiigatagarbha theory of the Ratnagotravibhiiga.
He thereby tacitly overlooks the problem that this identification (in
MSA III.II) contradicts the typical Yogacara classification into different
potentials (gatra), including explicitly an ultimate cut-off potential, and
leaves it to the reader himself to apply his hermeneutics and restrict the
teaching of different gatras to the level of apparent truth.
To sum up, without the Tathiigatagarbhasiitra and the Ratnagotra-
vibhiiga there would be little that directly supports Taranatha. Even
though there are fundamentally different interpretations in the Tibetan
traditions regarding the Ratnagotravibhaga, one can say without a doubt
that the ultimate is not self-empty (ran stan) but endowed with insepa-
rable supreme qualities.
53
Thus wisdom is explained as being already
present in ordinary sentient beings as an integral part of the Buddha
53. RGV 1.155: sanya agantukair dhatuJ:! savinirbhagaZalqQ/}aiJ:! / asanyo 'nuttarair
dharmair / "The Buddha element is empty of accidental
[stains], whose mark is that they can be separated. It is not empty of the supreme
qualities, whose mark is. that they cannot be separated [from the Buddha
element]." (Johnston 1950: 76). See also RGV IT.5, cited by Tiiranatha above.
MATHES 221
element.
54
It only has to be cleaned of the accidental stains of defile-
ments.
A problematic feature of the yogiieiira-tathiigatagarbha synthesis is
that Taranatha also applies the equation "dharmadhiitu = wisdom" to the
Madhyiintavibhiiga, and refers to "emptiness" in the third initial propo-
sition of MAV 1.1. as "wisdom of (i.e., which is) emptiness" (ston pa
ilid kyi ye ses). As we have seen, however, this is merely one legitimate
and interesting exegesis of a problematic stanza.
The Third Karmapa Rail bymi rdo rje (1284-1339) had earlier enun-
ciated a similar position in his rNam par ses pa dan ye ses 'byed pa'i
bstan beos, which represents his understanding of the Yogacara works
attributed to Maitreya. Unlike the Jonangpas, however, he does not
present dharmadhiitu as wisdom when describing the transformation of
the eight types of consciousness into the four kinds of wisdoms in accor-
dance with the Yogacara works. Only in his final summary does Ran
byun rdo rje indicate his understanding of the dharmadhiitu as one of the
five ever-present wisdoms in a gian ston sense.
55
In fact, it was Dol po
pa who started to use this terminology freely according to his gian ston
interpretation of the Buddhist literature. In other words, Dol po pa and
later Taranatha took dharmadhiitu or emptiness in the Y ogacara works
as a kind of wisdom, and therefore felt free to call it that, a license
against which other schools reacted strongly.56
54. See RGVV on 1.28: "It is said that all living beings have Buddha nature on the
grounds that all (lit. "the multitude of') sentient beings are included in the Buddha
wisdom." (buddhajfiiiniintargamat sattvariises ... uktiiJ:! sarve dehino buddha-
garbhiiJ:!, JOHNSTON 1950: 26,11. 1-4). See also RGV 1.102-104, where in the
second of nine examples which illustrate the ever existing Buddha nature, honey
is compared to undefiled wisdom and the swarm of bees concealing the honey, to
defIlements (JOHNSTON 1950: 26, 11. 1-4 & 61,11.5-15).
55. Having enunciated the transformation into four kinds of wisdom and three kayas
(dharmakiiya, sa/]'Ibhogakiiya and nirmiil}akiiya), he explains these three kiiyas
of transformed consciousness as the sViibhiivikakiiya, which rests in the dharma-
dhiitu. In his final summary, Rail byuil rdo rje says: ''The actualization of the
own-being of the five wisdoms and four kayas is Buddha-[hood]. Endowed with
the stains of the mind, intellect and consciousness, it is the iiIayavijfiiina. Free
from stains, it is called the essence of the victorious one" ( ... ye ses Ina dan sku
bzi yi I ran biin mnon gyur sans rgyas te I sems yid mam par ses pa yi I dri mar
ldan gan kun gii yin I dri med rgyal ba'i sfiin por brjod, Rail byun rdo rje: rNam
par ses pa dan ye ses 'byed pa'i bstan beos biugs so, p. 4b4).
56. See STEARNS 1999: 48-50.
nABS 23.2 222
Bibliography
Tibetan Texts
Karmapa Tanjur. Rumtek/Delhi 198? (no date)
Kon sprul Blo gros mtha' yas: Ses bya kun khyab, 3 Vols. Beijing: Mi rig dpe skrun
khail 1982.
TaraniUha "gian ston siiin po ces bya ba Mugs so," in rie btsun Taranatha'i
gswi 'bum biugs so. Leh: publ. by Namgyal & Tsewang Tam 1982-5,
Vol. 4, pp. 491-514.
Rail byun rdo rje (Third Karmapa): rNam par ses pa dari ye ses 'byed pa'i bstan beos
biugs so. Rumtek: Rumtek Par khail, no date.
Sakya mchog Idan, gSer mdog pan chen: "Zab ii spros bral gyi Mad pa ston fud bdud
rtsi'i lam po che ies bya ba Mugs so," in gSer mdog chen sa kya
mehog ldan gyi gsuri 'bum legs Mad gser gyi bdud rtsi glegs bam.
Collected Works. Thimphu: Kunzang Topgey 1975, Vol. 4.
Other Works
Chalmers, R. (ed.) 1899: "Cu!asuiiiiatasuttllI)l", Majjhimanikaya, Vol. m. Text Series
62. London: Pali Text Society, pp.104-109.
Eckel, D. 1985: "Bhavaviveka's Critique of Yoga car a Philosophy in Chapter XXV of
the Prajiiapradlpa," Indiske Studier 5: 25-75.
Johnston, E. H. (ed.) 1950: The Ratnagotravibhaga Mahiiyanottaratantrasastra.
Patna: The Bihar Research Society.
Levi, S. (ed.) 1907: Asariga: Mahiiyana-Sutrala1!lkara, Vol. 1: Textes. Bibliotheque de
l'Ecole des Hautes Etudes 159. Paris: Librairie Honore Champion.
Lamotte, E. (ed.) 1935: Sa1!ldhinirmocanasutra: L'Explication des Mysteres. Louvain,
Paris: Universite de Louvain.
Mathes, K.-D. 1996: Unterscheidung der Gegebenheiten von ihrem wahren Wesen
(Dharmadharmatavibhiiga). Indica et Tibetica 26. Swisttal-Odendorf:
Indica et Tibetica Verlag.
1998 "Vordergriindige und hochste Wahrheit im gian ston-Madhya-
maka," in H. Preiss1er & H. Stein (eds.): Anniiherung an das Fremde,
XXVI. Deutscher Orientalistentag yom 25. bis 29.9.1995 in Leipzig,
ZDMG-SUPPL. 11, pp.457-468.
in print: "Taranatha's Twenty-one Differences in respect to the
Profound Meaning: A Possible Starting-Point for Studies in the
gZhan stong Madhyamaka." To be published in the Proceedings of
the International Association for Tibetan Studies, Bloomington 1998.
Nagao, G.M. (ed.) 1964: Tokyo: Suzuki Research
Foundation.
Powers, J. 1995: Wisdom of Buddha: The Sa1!ldhinirmocana Sutra. Berkeley: Dharma
Publishing.
MATHES 223
Schmithausen, L. 1987: Alayavijfiana: On the Origin and the Early Development of a
Central Concept of the Yogacara Philosophy. 2 vols. Studia Philo-
logica Buddhica, Monograph Series IV. Tokyo: The International
Institute for Buddhist Studies.
Seyfort Ruegg, D. 1969: La Theorie du Tathagatagarbha et du Gotra. Paris:
Publications de l'Ecole d'Extreme-Orient (LXX).
Sponberg, A. 1981: "The Trisvabhava Doctrine in India & China: A Study of Three
Exegetical Models." Bukkyo bunka kenkyujo kiyo 21: 97-119.
Stearns, C. 1999: The Buddha from Dolpo: A Study of the Life and Thought of the
Tibetan Master Dolpopa Sherab Gyaltsen. New York: SUNY.
Urban, H. B. & Griffiths, P. 1. 1994: "What Else Remains in Sunyatii?: An Investi-
gation of Terms for Mental Imagery in the Madhyantavibhaga-
Corpus." JIABS 17.1: 1-25.
Williams, P. 1989: Mahayana Buddhism. London: Routledge.
Yamaguchi S. (ed.) 1934: Sthiramati: Madhyantavibhaga!fka, Vol. 1: Texte. Nagoya:
Librairie Hajinkaku.
SARA McCLINTOCK
Knowing All through Knowing One:
Mystical Communion or Logical Trick in the
Tattvasaf!lgraha and Tattvasaf!lgrahapafijika*
It is well known that Buddhists of various stripes have had a difficult
relationship with the concept of omniscience. In the earliest surviving
corpus of texts, we find passages both for and against the idea that the
Buddha knew all things, or dharmas, without exception. Nonetheless,
over time it became commonplace for both philosophers and sutra
compilers to refer to their founder not only by the well-worn epithets of
Buddha, Tathi'igata, and Bhagavan, but also as Sarvajfia, the All-
knowing or Omniscient One. This was so despite the fact that according
to many Buddhist philosophers - especially the Madhyamikas of later
centuries and others who followed the Prajfiiipiiramitii collection of
Mahayana sutras - there ultimately was nothing to know. Starting with
Nagarjuna and his peers, one observes an increasing dedication to the
idea of an omniscient Buddha; yet simultaneously, one also encounters a
relentless attempt to dismantle the very "individual nature" (svabhiiva)
that would seem (at least according to many of the realist ontologies
then current in India) to allow for any knowledge at all. Why this occurs
is a question that I cannot address in this short essay. What I can do,
however, is present one Mahayana approach to the problem of omni-
science - what I call the knowing all through knowing one approach -
and show how it was elaborated in an ingenious fashion by
and KamalasTIa, two Indian Buddhist philosophers of the eighth century.
The idea that one should be able to know everything simply by
knowing one single thing appears counterintuitive. For such a maxim to
make sense, it would seem to be necessary that all things are in fact one
single thing, as in certain Vedantin positions. A passage from the
* This paper was first presented at the 1999 annual meeting of the American
Academy of Religion at a panel on yogic perception. For this revised version, I
would like especially to thank Dan Arnold of the University of Chicago and John
Dunne of the University of Wisconsin-Madison for numerous insightful
comments and criticisms.
Journal of the International Association of Buddhist Studies
Volume 23 Number 2.2000
JIABS 23.2 226
Brhadaral}yaka suggests such a position when it states
"Through the seeing, hearing, contemplation, and knowing of the Self,
everything is known."! Here, direct knowledge of the atman or Self
somehow is or entails knowledge of everything, which in tum suggests
that everything is just the iltman. Advaita Vedantin philosophers, with
their extreme forms of monism, might be able to make sense of this
dictum,2 but, as is commonly known, Buddhists generally reject monis-
tic metaphysical formulations, such as that of the world-spirit brahman
or the universal atman. So, when Buddhist philosophers invoke the
knowing all through knowing one approach to omniscience, it seems
likely that they have something else in mind. .
Perhaps the most obvious difference between monists and Buddhists in
the formulation of the knowing all through knowing one thesis is the
nature of the one thing that is known. For Buddhists, rather than a
single, underlying and absolute reality, such as brahman or atman, it is
precisely the lack of such a reality (i.e., 'emptiness' or sunyata) that
constitutes the single thing, knowledge of which is equated with
knowledge of all. A.ryadeva, the presumed student of Nagarjuna,
presents an early Mahayana formulation of this principle in his Catuf:z-
sataka:
bhavasyaikasya yo sarvasya sa smrtal; I ekasya iiinyatii yaiva
saiva sarvasya iiinyatii 1/ "Who sees just one thing is said to see all; the empti-
ness of one is just the emptiness of all." [CS 8.16]
A.ryadeva's purpose in this verse is to show that a particular (and
crucial) requisite of liberation is possible. That is, in the Buddhist
formula of liberation that A.ryadeva advances here, it is necessary that
one become definitively aware that one has stopped, prevented, or
turned back - the Sanskrit is varalJam - first everything that is non-
meritorious, then the iltman or Self, and finally all dharmas, or things.
This formula, which A.ryadeva lays out in the verse immediately prior to
CS 8.16, stresses three necessary reversals.
3
While the fIrst two reversals
evince a general Buddhist emphasis on moral rectitude and the principle
1. Brhadaral).yaka 2.4.5.: iitmano vii are darianena iraval)ena matyii
vijiiiineneda'!l sarva'!l viditam.
2. See especially the early chapters of PHILLIPS (1995) for details on the varieties of
Advaitin monism.
3. CS 8.15: viiral)a'!l prag apul)yasya madhye varal)am iitmanal; I sarvasya
viiral)a'!l paiciid yo janfte sa buddhimiin II.
McCLINTOCK 227
of no-self (anatman), the third reversal reflects the specifically
Madhyamaka position that one should destroy all notions of indepen-
dent, inherent, or ultimately real existence in relation to all things. In
response to the question of how, even before liberation, an ordinary
practitioner can come to know all things, such that she might then be
able to verify that she has indeed 'stopped' or 'turned them back' with-
out exception, Aryadeva says "Who sees just one thing is said to see all;
the emptiness of one is just the emptiness of all."
In his commentary on this verse, CandrakIrti explains that the empti-
ness of one dharma is not different from the emptiness of all dharmas.
4
Thus, to know the emptiness of all dharmas, one must know only the
emptiness of a single dharma. In this way, emptiness is similar to space.
As space remains unchanged regardless of the changing forms - such as
ewers and bowls - that surround it, emptiness too remains unchanged
regardless of the multitude of dharmas that possess or 'hold' its But
4. CandrakIrti starts with the standard Buddhist categories of the five skandhas, and
then proceeds to the twelve liyatanas and the eighteen dhlitus to show that all
infinite and various entities have a single nature: emptiness. See CSV ad CS 8.16
(D ya 139a6-l39b1): gzugs kyi ran biin ston pa iiid gan yin pa de iiid tshor ba
la sogs pa phun po rnams kyi ran biin ston pa iiid do I de biin du mig gi skye
mched gyi ran biin ston pa iiid gari yin pa de iiid skye mched bcu giiis char gyi
yari yin no I de biin du mig gi khams kyi ran biin ston pa iiid gari yin pa de iiid
khams bco brgyad char gyi yari yin no I de biin du drios po dari yul dan dus dan
rten gyi dbye bas tha dad cin rab tu dbye ba tha' yas pa rnams las drios po gcig
gi rari biin ston pa iiid gari yin pa de iiid drios po thams cad kyi rari biin stori pa
iiid do I.
5. CandrakIrti makes this point both in his co=entary on CS 8.16 and in a related
passage in his independent treatise, the Madhyamaklivatlira (MAv). In the CSV
he states (D ya 139b1-2): bum pa dari ' khar giori la sogs pa tha dad kyan rnam
mkha' tha dad med pa biin no I gzugs la sogs pa'i srios po tha dad kyari gzugs
la sogs pa rnams kyi ran biin gyis ma skyes pa las tha dad pa med pa'i phyir
chos gcig kho na'i rari biin gyis ma skyes pa yons su ses na chos thams cad kyi
ran biin gyis ma skyes pa yan yoris su ses par ' gyur te I. The MA v 12.2 states
(D 'a 216a5-6): ji ltar snod kyi dbye bas mkha' la dbye ba med de ltar I drios
byas dbye ba ' ga' yari de iiid la med de yi phyir I ro miiam iiid du yari dag thugs
su chud par mdzad gyur na I mkhyen bzari khyod kyis skad cig gis ni ses bya
thugs su chud II. In his own commentary on this verse, CandrakIrti explicitly
names the knowledge of the single flavor of emptiness as "omniscient wisdom."
He says (D 'a 330a3-5): dper na bum pa dari 'khar giori la sogs pa tha dad
kyari sgrib pa med pa iiid du mtshuris pa' i phyir der Hogs pa' i nam mkha' la tha
dad pa med pa de biin du I gzugs dari tshor ba la sogs pa'i dnos po tha dad
kyan der gtogs pa' ide kho na iiid skye ba med pa' i mtshan iiid can la tha dad pa
med pas de kho na iiid ni ro gcig (D adds pa) kho nar ses par bya' 0 I de yari
JIABS 23.2 228
emptiness is more significant than space, because, unlike space, empti-
ness can be shown to be the ultimate nature of all things. Thus, by
knowing emptiness in relation to one thing, one also knows the ultimate
nature of all things. Since one knows the ultimate nature of all things,
one also eliminates ignorance entirely.
Two points are important here. First, we should recognize that the
single thing, knowledge of which equals knowledge of all, is both a
universal and an abstraction: it is empti-ness. This might again lead one
to think of Vedantin monists like Mal.l<;ianamisra, who hold that only
brahman, understood as universal Being, or satta, is ultimately rea1.
6
But emptiness differs from the ultimate reality of Vedantins like
Mal.l<;ianamisra, in that it is generally seen as a negation possessing no
greater reality than all other abstractions and universals. And this brings
us to the second point we should note: on the Buddhist view, when it is
said that all dharmas are qualified by the universal emptiness, this does
not mean that there is some single entity or thing, 'emptiness,' which
truly exists, underlying or otherwise applying to all appearances. Empti-
ness cannot be a real entity, for, like everything, it is subject to the same
arguments that show that all dharmas without exception are ultimately
devoid of real and independent existence.? Even on the level of conven-
mkhyen pa'i skad Gig gGig kho nas thugs su chud pas bcom ldan ' das kyis
mkhyen pa'i skad cig gcig kho nar thams cad mkhyen pa'i ye ses brfies so II.
6. For a discussion of how MaJ).<;ianamisra construes the universal (siimiinya) as
that which is non-different (abheda) and the particular as that which is
different (bheda), see THRASHER (1978). According to THRASHER, for
MaJ).<;ianamisra, only the non-different universal, Being or sattii, is true, and only
that is known in perception. This non-different universal is the nature (svabhiiva)
of all things, and it is the same in all things (136). All things are thus ultimately
one, and differences are all a result of ignorance. The similarity to the Buddhist
view is striking, as is the crucial difference between them: namely, that for the
Vedantin the single nature ofall things is real, and for the Buddhists (generally) it
is a negation and hence unreal.
7. In his Bodhipathapradfpapa/ijikii (D khi 279a), the eleventh-century Bengali
Buddhist scholar AtIsa delineates four great logical reasons (gtan tshigs chen po
biz) used by Madhyarnikas to establish lack of independent existence: the reason
which refutes production of an existent, a nonexistent, a both existent-and-
nonexistent, and a neither-existent-nor-nonexistent entity (mu bii skye ba 'gog
pa'i gtan tshigs); the so-called vajraka!/a-reason, or reason that refutes produc-
tion from self, other, both, or neither (rdo rje gzegs ma'i gtan tshigs); the neither-
one-nor-many reason (gcig dan du ma brai ba'i gtan tshigs); and the reason of
dependent arising (rten Gin ' brei ba ' byun ba'i gtan tshigs). In addition to their
primary function of refuting independent natures in general, Madhyamikas also
McCLINTOCK 229
tional reality, emptiness is not real, both because it is a negation and
because it is a conception. Furthermore, as a negation, emptiness can
also be understood to escape the problem of whether it is the same as or
different from the dhannas that instantiate it.
8
Such is the general thrust
of the Madhyamaka argument, as I understand it.
Now, the fundamental question that presents itself here is: What is the
nature of the 'all' that one knows in knowing emptiness? In his com-
mentary on the CatuJ:tSataka and in a related passage in the Madhya-
makiivatiira, CandrakIrti refers to the 'one flavor' or ekarasa metaphor,
in which knowledge of the emptiness of all dharmas is compared with
knowledge of the saltiness of seawater.
9
Just as one does not need to
use these reasons to show the lack of independent existence of the universal
emptiness itself. For an emphatic statement of the "emptiness of emptiness," see,
e.g., CandrakIrti's MAv 6.185-6.
8. This way of phrasing things is an application of the Buddhist epistemological
tradition's understanding of universals as exclusions (vyiivrtti) or negations
(nivrttz) that have no real relation to the entities they qualify. For a succinct state-
ment of how universals as exclusions are ultimately unreal, see TS 1188: na
bhiivo niipi ciibhiivo ' niisritiiniisrito 'poho naikiinekas
ca vastuta/:lll. The same principle can be seen as work in the context of a
Madhyamaka discussion of emptiness; see, for example, MAY and MAP ad
MA 70. For more on universals as exclusions (i.e., apoha), see below.
9. The elucidation of the 'one flavor' doctrine through the metaphor of the saltiness
of seawater comes in CSY ad CS 8.16 just after the citation of two verses from
the Samiidhiriijasiitra, which CandrakIrti also quotes in PrasP ad MMK 4.9 in
what seems to be a related context. In MMK, Nagiirjuna appears to argue that
since all things are equally devoid of independent existence (svabhiiva) - in other
words, all things are equally empty - no fault can ever be urged against
Nagfujuna's position. Whatever entity the opponent tries to put forth as an
example of something that does possess independent existence is equally
questionable (siidhyasama) and hence not suitable to be used in a proof state-
ment. In his commentary, CandrakIrti further explains that, just as the indepen-
dent existence of form cannot be found, neither can that of the other four
skandhas. In each case, the skandha can be shown to be neither the same as, nor
different from, its cause. Seeking apparently to back up this reasoning with
scripture, CandrakIrti then cites the following two Samiidhiriijasiitra verses:
yathii jfiiita tayiitmasa7J1.jfia tathaiva sarvatra buddhi/:ll sarve ca tatsva-
bhiivii dharmavisuddhii gaganakalpii/:lII l2.7 II ekena sarva7J1. jiiniiti sarvam
ekt;na paryati I kiyad bahu pi na tasyotpadyate mada/:lllll.l611. In the
CSY, CandrakIrti again cites the verses (in reverse order) and then presents the
helpful example of the saltiness of seawater as an illustration for knowing all
through knowing one. The Tibetan translation of the passage states (D ya 139b
5-6); rgya mtsho'i chu'i thigs pa 'thUlis pa biin no I ji ltar ro mtshUlis pa'i phyir
JIABS 23.2 230
drink all the waters of the sea to know that all of it is salty - it suffices
merely to drink a drop - likewise, one knows the emptiness of all
dharmas simply by knowing the emptiness of one. Candraklrti states in
the that it is in this manner that through a
single instant of awareness a Buddha can attain omniscient wisdom
(thams cad mkhyen pa'i ye ses).1
0
All of this seems to indicate that, for
these Buddhists at least, the Buddha's omniscience consists in the
knowledge of emptiness alone, but does not imply any further
knowledge of the infinite and various individual (and ultimately unreal)
dharmas of conventional reality. In other words, knowing all here means
simply knowing the ultimate rea,lity of all - it is a metaphorical
omniscience rather than a literal one. 11
gcig tu rgya mtsho' i ro myafzs na thams cad du ro myafz bar' gyur pa de biin du
dnos po rnams kyi 'dus byas ston pa Hid kyan yin te I mtshan Hid mtshun pa' i
phyir ro II.
N.B. The precise interpretation of MMK 4.8-9 and PrasP ad cit. has been the
subject of some dispute. In particular, B.K. MATILAL (1974) and K.BHATTA-
CHARYA (1974) have argued against the interpretation of the notion of siidhya-
sama as petitio principii, an interpretation often taken by modem scholars, both
in the present context and more generally. MATILAL (221-222) states that rather
than being an instance of question begging, Nagarjuna's reasoning in:MMK 4.8
can be summed up as follows: "If a reason is adduced to refute the Madhyarnika
thesis of emptiness in a philosophic dispute, it will not constitute a refutation, for
it will enjoy the same predicament along with the proposition to be proved."
Nagarjuna's fundamental point here seems only to be that nothing can be
assumed to have svabhiiva and that all things can and must be subject to the type
of tests for svabhiiva that he provides in the MMK before they can be used as
reasons or justifications to prove svabhava in something else. Insofar as he has
shown the emptiness of each of the five skandhas, he believes that he has shown
the emptiness of all things - this is in keeping with the Buddhist notion that the
five skandhas comprise all that dependently arises. Nagiirjuna does not seem to
make any explicit claims for knowing the emptiness of all things through
knowing the emptiness of one; it is Candraldrti who introduces this idea in the
commentary. In the case of Nagarjuna' s student Aryadeva, however, the theme of
knowing all through knowing one is clearly present even in the root text.
10. See n. 5 above.
11. That is, it is a metaphorical omniscience as long as one maintains that conven-
tional objects are in fact objects that can be known (i.e., jHeyas). If, instead, one
were to hold that emptiness is the only truly knowable thing (because it alone is
ultimately true), then knowing the emptiness of a single thing would be a literal,
not a metaphorical, omniscience. Such a perspective appears unlikely, however,
since emptiness qua object of knowledge is usually considered to be conven-
tional. CandrakIrti's position on the question of whether and how conventional
McCLINTOCK 231
But the drive to attribute 'maximal greatness' to the Buddha appears to
have been strong, and it seems that some Mahayanists (or perhaps most)
envisioned a Buddha whose omniscience was m ~ c h more than metaphor-
iCld.
12
While maintaining the general structure of the knowing all
through knowhig one approach to omniscience, certain Buddhists appear
to have understood the 'all' that is known in this fashion to be far more
expansive than just knowing emptiness - encompassing literally every-
thing, whether real or unreal, ultimate or conventional. To illustrate this
tendency, consider the following passage from the ITa ba'i khyad par,
an eighth-century work by the Tibetan author Ye ses sde. John
MAKRANSKY has translated the passage as follows (additions are those
of the translator):
A Tathiigata knows all things free from conceptualization: both their individual
characteristics and their general characteristic [emptiness]. Prior [to attaining
Buddhahood,] bodhisattvas' actions are not yet spontaneous, their obstructions
are not yet fully purified, their power is not yet inconceivable. [They alternate
between sessions of meditative equipoise and activity.] When they abide in
meditative equipoise, they see emptiness, the general characteristic of all things,
with nonconceptual awareness (rnam par mi rtog pa'i ye ses). When they arise
from that equipoise [for postmeditation session activity], they know the indi-
vidual characteristics [of things] to be merely an illusion, through their purified
awareness of the world (dag pa 'jig rten pa'i ye ses) which is obtained sub-
sequent to that [nonconceptual awareness].
By proceeding to higher and higher levels through that training, [they attain]
its fruit, the stage of Buddhahood where [all] conceptuality is cleared away and
obstructions are [fully] purified. At that point meditative equipoise occurs of
objects and emptiness are knowable objects (jfieya) is a contentious issue. There
is some evidence that he considers at least statements concerning knowledge of
the ultimate to be metaphorical; see, e.g., MAvBh ad MAv 12.4 (D 'a 330bS),
where he remarks that when knowledge occurs in relation to an image (rnam pa
= ilkilra) of reality (de kho na fiid = tattva), then the statement that "there is
knowledge of reality" is a metaphorical application (fie bar btags pa = upacilra):
ses pa de kho na fiid kyi rnam pa' i rjes su byed pa skye ba na de kho na fiid ses
so ies fie bar gdags so I. The real import of this passage, however, may be to
show that knowledge of the ultimate (and not just statements about that
knowledge) is itself metaphorical, since shortly after the above remark, we find
CandrakIrti insisting that in reality there is no knowledge of anything, as neither
knowledge nor any knowable object is actually ever produced (D 'a 330b6-7):
de'i phyir btags pa las de kho na fiid rtogs so zhes rnam par bzhag gi / dnos su
na 'ga' zhig 'ga' zhig gis ses pa ni ma yin te I ses pa dan ses bya gfii ga yan ma
skyes pa fiid kyi phyir ro I.
12. The term 'maximal greatness' is that of Paul GRIFFITHS. See especially
GRIFFITHS (1994: passim).
JIABS 23.2 232
itself. Then, without ever arising from that equipoise, all the cognitive objects that
exist are mamfest at once. .
... As it says in the Buddha-avataT[lsaka-sutra: "The mind of the Bhagavan
Buddhas, having been purified through an inexpressible period 'of hundreds of
billions of trillions of aeons, is free of obstruction. [It perceives] all realms of the
cosmos without exception within its cognitive sphere. [It perceives] all of the
dharmadhiitu (universal emptiness) without exception within its cognitive
sphere. It knows all with respect to past, present, and future, with one cognition
(dgOlis pa gcig), free of obstruction. It possesses all in its comprehension."13
Clearly, for the author of this text, the Buddha's omniscience is much
more than a metaphor. Although neither Yeses sde nor the sutra passage
that he quotes states that the Buddha. knows all things through kllowing
one single thing (one jiieya), both similarly stress that the Buddha knows
all things through one single cognition (one jiiiina), in one single
moment. Since a single cognition is generally understood to have a
single object, the structure of omniscient knowledge can still be charac-
terized as knowing all through knowing one; the claims, however, appear
much greater. Buddhas not only know the ultimate nature of all things,
they somehow actually know all myriad things without exception. Yet
the passage does not clarify how this can be. Indeed, the shift from
meditative equipoise centering on emptiness to full-blown omniscient
13. MAKRANSKY (1997): 351-52. In addition to being the author of this text, Ye ses
sde was also a leading translator under the Tibetan king Khri sroil Ide btsan.
Most notably for us, he was also the co-translator of main
Madhyamaka texts (the MadhyamakiilaT[lkiira and its Vrtti) along with Kamala-
SIla's commentary, the MAP (MadhyamakiilaT[lkiirapafijikii). For more on the
author and his work, see SEYFORT RUEGG (1981). The Tibetan for the above
passage is as follows (D jo 219b6-220a4): mam par mi rtog biin du ran dan
spyi'i mtshan fiid thams cad mkhyen te I snon byan chub sems dpa' mdzad cin
sgrib pa ma byan lhun gyis ma grub I dban bsam gyis mi khyab pa ma brfies
pa'i tshe I miiam par giag pa' i dus na ni mam par mi rtog pa' i ye ses kyis dnos
po thams cad kyi spyi'i mtshan iiid ston pa iiid du gzigs la I de las biens [D:
biins] nas rjes las thob pa dag pa 'jig rten pa'i ye ses kyis sgyu ma tsam du ran
gi mtshan iiid mkhyen pa de gon nas gon du sbyans pas rtog pa bsal te sgrib pa
byan ba'i ' bras bu sans rgyas kyi sa la lhun gyis te mfiam par 'jog pa dan I
biens pa mi mna' bar ses bya ji siied pa de thams cad dus gcig tu mnon du gyur
te I ... sans rgyas phal po che'i mdo las kyan I sans rgyas beam [dan' das mams
kyi dgons pa ni chags pa mi mna' te I bskal pa bye ba khrag khrig brgya ston
brjod du med pa'i yan brjod du med par yons su sbyans pa' 0 I 'jig rten gyi
khams thams cad ma Ius par yul du gyur pa' 0 I chos kyi dbyins ma Ius pa thams
cad yul du gyur pa' 0 I dus gsum thams cad la dgons pa gcig gis chags pa med
par mkhyen iin thugs su chud pa'i yul dan ldan pa' 0 ies bya ba la sogs pa rgya
cher gsuns so I.
McCLINTOCK 233
knowledge of all things whatsoever takes on an air of mystery, like an
experience of mystical communion that cannot be explained through
rational means.
Quite possibly some Buddhists would be content to let the matter rest
there. After all, the Buddha's omniscient awareness is frequently said to
be inconceivable. How could a person who has not experienced full-
blown omniscience be expected to explain or prove it? Nonetheless, for
Buddhists of a rational bent, this 'non-explanation' is less likely to be
satisfying. and KamalaSila are two examples of this latter
type of Buddhist. Using the logical and epistemological tools of instru-
mental awareness or valid cognition (pramti"(la) that they inherited from
their predecessors Dignaga and DharmakIrti, these philosophers appear
set on providing a rationally coherent explanation of how a Buddha can
literally know all things at once. Not only do they proffer a proof of the
Buddha's omniscience, it also seems likely (given the space devoted to it
and its prominent position at the end of the work) that the demonstration
is one of their most cherished and original contributions in the two texts
that I will consider here: the Tattvasa'!!graha (TS) by and
its commentary, the Tattvasa'!!grahapafijikti (TSP) by his student
Kamalaslla. Indeed, in the opening pages of the monumental TSP,
KamalaSila states his opinion that for a person who is inclined toward
faith, it may be sufficient simply to hear about the Buddha's great
qualities in order to set out on the path. But for a person inclined toward
wisdom, it is necessary to use valid cognitions to demonstrate that such
great qualities can actually occur. Only when one is sure that one's goal
is possible does it make sense to engage in practices designed to lead to
that goal. As KamalaSila clarifies, the person inclined toward wisdom -
in other words, the rational and judicious human being - can definitively
determine that the Buddha's omniscience is possible, and hence worth
striving for.l
4
The TS and TSP are apparently intended, at least in part,
14. TSP ad TS 1-6: ye 'pi prajJianusari1)a[t te 'pi badham apayantal;
prajnadfnaT]'! ca gU1)anam abhyasat avagacchanto apy
atzndriyarthadrksadhakat pram(1)at niinam jagati sambhavyanta eva tatha-
vidhal; siiraya[t iti II "But those who are inclined toward wisdom, not seeing any
refutation (badha) of these kinds of [qualities], and understanding that good
qualities such as wisdom and so on [can reach] perfection through
habituation (abhyasa), [will determine] - through the valid cognition that will be
stated [in the final chapter] which proves the seeing of supersensible objects-
that 'certainly these kinds of sages are possible in the world.'"
nABS 23.2 234
to lead their readers to the conclusion that the Buddha can and does have
such extraordinary powers and knowledge.
The long final chapter of the TSIP is justly famous for its detailed
discussion and attempted demonstration of the Buddha's omniscience.
Somewhere around the middle of the chapter, two obscure objectors,
Sama!a and Yajfia!a, are made to give voice to the same doubt con-
ceming the knowing all through knowing one approach to omniscience
that we have hinted at here.
ls
One of the strongest statements is the
following:
svabhavenavibhaktena yaf:t avabudhyate / bhavanam
na sa budhyate II One who understands everything in terms of a non-
different nature does not know the individual natures of all things.
[TS 3250]
When and Kamalasila get around to addressing this objec-
tion nearly two hundred verses later, they make it quite clear that the
Buddha's knowledge of a non-different, (i.e., a single and universal)
nature does imply knowledge of all individual natures. In a deliberate
mirroring of the objectors' statement, quips:
svabhavenavibhaktena yaf:t sarvam avabudhyate / svarupal}y eva bhavana'!l
so 'vabudhyate II One who understands everything in terms of a non-
different nature understands precisely the individual natures (svarupa) of all
things. [TS 3631]
How do and Kamalasila defend this claim? To understand
their answer, it is necessary to understand something of Dignaga and
Dharmaldrti's theory of exclusion, or apoha. According to this theory,
only particulars are truly real, while universals are unreal mental con-
structS.
16
These mental constructs are fabricated by the mind through a
. process of exclusion, which these philosophers feel allows them to
15. The objections of Slimata and Yajfiata are presented in TS 3246-3260, just after a
much longer section in which, KamalasTIa tells us, the views of the
Kumiirila are given as the These two otherwise unknown authors
may be as well, since they are shown to argue against the possi-
bility that knowledge of Dharma can arise through the speech of a human being.
It is unfortunate that nothing more is known about them, since, philosophically
speaking, their objections are in many ways more compelling than those of the
more extensively quoted Kumarila.
16. Sources that explore the apoha theory in the works of Dignaga, Dharmaklrti and
their Indian and Tibetan successors include DREYFUS (1997), DUNNE (1999),
HATIORI (1980), and HAYES (1988).
McCLINTOCK 235
account for the validity of the conventional usage of universals while
simultaneously denying their ultimate reality. The mechanism for this
process depends upon the fact that one can class certain particulars
together, even though in reality each one is utterly unique, because one
can validly ascertain that they are all equally "excluded from everything
that does not perfonn the expected function of X."17 In this way, one is
able to use a kind of shorthand for the sake of communication and say
that certain groups of particulars possess a certain universal property, or
x-ness. Although he uses the positive language of universals and
properties, the Buddhist apohaviidin understands that the universal x
actually just indicates that the particular in question shares with certain
other particulars, a common negation - namely, the exclusion from all
other particulars that do not behave in the manner necessary for the con-
struction of that universal x. Universals are simply a way of construing
particulars, apart from which they have no independent or real existence.
In the context of the proof of omniscience, this last fact is critical.
That is, and KamalasTIa want to point out that knowledge of
emptiness cannot take place without knowledge of a particular that can
conventionally be said to be "qualified" by the exclusion which is
emptiness. In the parlance that they take over from DhannakIrti, the
exclusion (the universal) is not separate from the excluded thing (the
particular). Hence, in the verse that we just read by the
idea is that a person who knows a "non-different" or universal quality of
things must also have knowledge of the particular things themselves. But
s clever retort still does not explain knowledge of all
through knowledge of one. To clinch the argument (in their own eyes, at
any rate), Santaraksita and KamalasTIa need one further element, which
they find in the theory of yogic perception or yogipratyak!ja, drawn once
again from DhannakIrti.
DharmakIrti's theory of yogic perception, unlike some other Indian
theories, does not concern itself primarily with how yogis can see distant
objects, other times, and so on.1
8
Instead, this theory is designed to show
17. On the role of expectations in the formation of universals in DharmakIrti, see
DUNNE (1997): 136, n. 109.
18. The bibliographic sources for South Asian theories of yogic perception are
diverse. For the Buddhist tradition, see especially BUHNEMANN (1980),
PEMWIESER (1991), PREvEREAU (1994), and STEINKELLNER (1978). DREYFUS
(1997) also makes some helpful observations. Useful materials on early Nyaya
JIABS 23.2 236
how anything that one meditates upon long enough and single-pointedly
enough eventually achieves the clarity (sphutatva) and nonconceptuality
(nirvikalpatva) that DharmakIrti and his followers associate with direct
perception.
19
Whether the object of such a meditation is real or unreal
(bhuta or abhuta) is irrelevant to the meditative process; focusing long
enough and hard enough on anything leads one to experience it as if it
were right in front of one. Dharmaldrti cites a lover overcome by desire
or a father grieving for a lost son as examples of meditations on unreal
objects that can lead to visions that are the equivalent of perceptions in
terms of their clarity, though not in terms of their trustworthiJ;less.
2o
Emptiness, however, has already been established to be (conventionally)
true of all dharmas.
21
Hence, at the end of one's meditation on the
sources are found in OBERHAMMER (1984). For Kumarila's rejection of yogic
perception, see JRA (1998).
19. The locus classicus for DharmakIrti's theory of is PV3.281-287.
Other important locations include NBl.ll and PVinl. Dignaga discusses the
topic at PS1.6 and PSV ad cit. Dharmaldrti defines perception as an awareness
that is both free from conception and nonerroneous. See NB 1.4: tatra praty-
kalpaniipor;lham abhriintam I. The sign that an awareness is free from
conception is its obvious vividness or clarity (spu!atva). That it is nonerroneous
is ascertained in various manners. In the case of meditation on the emptiness, or
selflessness (nairiitmya), of all dharmas, the nonerroneous element of the aware-
ness is already definitively determined through inference. Thus, when the requi-
site clarity arises, the resulting awareness can be classed as perception.
20. In a paper delivered at the 1995 annual meeting of the American Academy of
Religion, entitled "Obsessions and Holy Truths: the Ultimate as Conventional in
DharmakIrti," John DUNNE explained how Dharmaldrti envisions the process of
Taking DharmakIrti's own example of the meditation on ugliness
(asubhii), DUNNE described how "at one stage one will directly see one's own
body as a skeleton stripped of all flesh; indeed, at a certain point, one will see
every person's body as a skeleton." On DUNNE's interpretation, this is an
example where the object of the meditation is unreal, for, as he joked, "our bodies
are not skeletons (yet)." Nonetheless, after sufficient meditative effort, the vision
of ourselves as skeletons can have the qualities of clarity and nonconceptuality
that are hall-marks of perception in Dharmaldrti' s system. The point really is that
anything one meditates on, or cultivates, in the requisite manner can eventually
attain the same vividness as ordinary perception, even if the content of the aware-
ness lacks the trustworthiness of a valid cognition (pramiil)a).
21. Although DharmakIrti generally speaks of the selflessness (nairiityma) of
dharmas, he does on occasion speak as well of their emptiness (cf., e.g., PV
2.253 and PV3.213-215), and it is clear that, if not precisely equivalent, the two
notions have a significant conceptual overlap in some contexts. In this paper, I
use the term emptiness throughout, since on and KamalaSila's
McCLINTOCK 237
mentally- constructed exclusion which is emptiness, one comes to have a
direct and trustworthy perception of just that emptiness. The direct
perception of emptiness is understood to be a powerful antidote to
beginninglessignorance, which functions through imposing individual
natures (svabhava) where none exist. This and other such meditational
practices resulting in the elimination of negative mental states constitute
what Dharmaldrti calls
We can now see how and KamalasIla put it all together.
The key move is to extend the argument based on yogic perception such
that the object of meditation is not the emptiness of a single dharma, but
rather the emptiness of all dharmas. In following the proof, it is impor-
tant to remember the specification described above: that is, since empti-
ness is an exclusion, it cannot occur in isolation from a particular.
KamalasIla expresses the proof statement (prayoga) as follows:
ye ye vibhiivyante te sambhavatsakrdekavijfllinasphu!apratibhiisaniiJ:t I yathiiriga-
niidayaJ:t I sarvadharmiiS ca bhiivyante iti svabhiivahetuJ:t II Whatever things are
meditated upon, it is possible that they will have a clear appearance all at once in a
single cognition, like a woman [who is meditated upon by a lovesick man]. Now,
all dharmas are meditated upon. The evidence is an essential property
(svabhiivahetu). [TSP adTS 3443-3445]
Underlying this proof statement is the assumption that one has already
established through inference (anumana) that the exclusion 'emptiness'
does indeed validly apply to all dharmas. Having established that fact
beyond any doubt, one may be equally assured that when meditating on
the emptiness of all dharmas, one will also be meditating on the real
particulars that are not different from the exclusion, emptiness. To be-
come omniscient, then, it seems that the key is to meditate long enough
and with enough concentration on the emptiness not of one single thing,
but rather on the emptiness of everything! In so doing, one will, ipso
facto, be meditating on all dharmas. Even though emptiness is a nega-
tion and hence unreal, it nonetheless can be said to possess a connection
to all particulars such that by fully knowing emptiness, one comes to
fully know particulars as well.
22
understanding, selflessness and emptiness can be used interchangeably, at least in
relation to See TS 3441-3442 and TSP ad cit.
22. and KamalaSUa are not explicit on this point, a fact which may
reinforce my reading of them as ultimately resisting such an interpretation. A
Buddhist contained in the NyiiyakmJikii, Vacaspatimisra's
commentary on Mal).<;lal).amisra's Vidhiviveka, however, does present the
JIABS 23.2 238
It is difficult to determine just how seriously and
KamalasTIa take their own proof. On the one hand, the demonstration
looks like a logical trick that relies upon an equivoc,ation in the
denotation of the term 'all dharmas.' That is, at the start of the yogi's
meditation, the term 'all dharmas' denotes a mental image; but at the
end of his meditation, 'all dharmas' denotes all entities (vastu or bhiiva)
without exception. Do and KamalasTIa endorse this double
signification? I think they do not, although I also suspect that the authors
may not be averse to allowing an apparent sleight-of-hand to stand in
order to bedazzle some of their readers. My reason for saying this is that
at the end of the TSIP, when summarizing the argument for omniscience
based on the authors do nothing either to refute such an
equivocation or to more fully explicate the nature of the particular that
is known when one knows emptiness of all dharmas. To the contrary,
and KamalasTIa almost seem to encourage a reading in
which the meaning of 'all dharmas' undergoes a mysterious change.
They accomplish this not only through their insistence on the funda-
mental point that meditation on the exclusion 'emptiness' entails medi-
tation on the particular from which the exclusion does not differ,23 but
argument succinctly and clearly: ca tannairiitmya-
/ na ca nairiitmya"f!l niima bhiivebhyo bhinna"f!l kiiicid asti
yad api syiit I "And the object [of the
meditation that leads to omniscience] is all things, since its nature is the direct
apprehension of the selflessness of that [i.e., of everything]. For
selflessness cannot exist apart from real things, such that it could become the
object of perception even when those [real things] were not the objects of
perception." See PEMWIESER (1991): 115 and 14l.
23. InTS 3632-3633, clarifies that although the object of the yogi's
meditation is referred to as the universal 'emptiness,' the actual object is not the
unreal universal per se, but rather a particular which can be construed as a
universal, that is, the particular as conceptually excluded (vyiivrtta) from such
properties as "possessing a self' and "being non-momentary." Although the
authors are not explicit on this point, it seems clear that the particular in this
instance is not, for example, the external cause of a perceptual image, but rather
that image itself. We can safely make this claim because it is that image which is
the direct cause of the cognition of sameness or abhinniikiira-
pratyaya) to which the authors refer in this passage. Thus, as KamalaSTIa
explains, although that mental image is a particular, it can be called a universal
insofar as it serves as a cause for that cognition of sameness. See TSP ad cit.: tad
eva hi vijiitfyavyiivrttam abhinniikiirapratyayahetutayii siistre
ity ucyate I. The cognition of sameness here is what
DharrnakIrti calls ekapratyavamarsajiiiina; see DUNNE (1999: 137-143).
McCLINTOCK 239
also through an explicit reference to the notion that omniscience entails
knowledge of such normally hidden entities as karmic causes and
effects.24
Despite this seeming ambiguity in the closing passages of the TSIP, I
think nonetheless that we can be confident that no real equivocation in
the meaning of 'all dharmas' has. occurred. For earlier in the .chapter, as
the attentive reader can be expected to recall, KamalasTIa explains that
what appears clearly in this kind of yogic perception is nothing other
than the mind (manas) or mental consciousness (manovijfzana) that takes
as its object (alambana) the 'emptiness of all dharmas' and so on; omni-
science is then equated with this clear appearance.
25
The obvious
24. Immediately after arguing that the object of yogic perception cannot be a
universal, goes on to equate omniscience with knowledge of karmic
causes and effects. See TS 3637: sahetu saphaJa,!! karmajfiiinenalaukikena yal;l
samiidhijena jiiniiti sa sarvajfiopadisyate II. This would seem to imply that a
literal and full-blown omniscience arises through the practices associated with
yogic perception; but see n. 30 below for an alternative vision of omniscience that
would account for the Buddha's display of knowledge of the details of karma
without requiring that he actually know them in any ordinary sense of that term.
25. Here, again, I have substituted the term emptiness where the authors use selfless-
ness. See TSP ad TS 3338: prayogal; I yii
bhiivanii sii karataliiyamiinagriihyiivabhiisamiinajfiiinaphalii I tad yathii kiimii-
turasya kiiminfbhiivanii I ca sarviikiirasarvagata-
nairiitmyabhiivanii kiirUlJikasyeti svabhiivahetul; I na ciisiddho hetul; I kiirUlJika-
syiirthitvena tathii pravrttisambhavasya pratipiiditatviit I niipy anaikiintikatvam I
yatal; sarvadharmagatanairiitmyiilambanasya manovijfiiinasya dharmilJo
hetunii sphu!apratibhiisi-
tva'!! siidhyam I etena ca siidhyadharmelJa yathoktasiidhanadharmasya vyiiptil;
siddhii I sphu!apratibhiisitvasya I tatas ca siimarthyiit
sarvajfiatveniipi vyiiptil; siddhii I yasmiit sarvavastugatanairiitmyiidyiilamba-
nasya manaso dharmilJo yat sphu!apratibhiisitva'!! tad eviisya sarvajfiatva,!!
niinyat I tathii hi bhiivyamiinavastusphu!apratibhiisitvena bhiivaniiyiil; siimii-
nyena vyiiptau siddhiiyii,!! siimarthyiit sarvajfiatveniipi siddhaiva I yathokta-
dharmilJo 'nyasya sphu!apratibhiisitvasyiisambhaviit I "The proof-statement is
as follows: That meditation which has the qUalities of being intense, uninterrupted
and of long duration results in an awareness in which the object appears as [if it
were] held in the palm of the hand, like the love-stricken man's meditation on his
beloved. And a compassionate person's meditation on the all-pervasive selfless-
ness in all its aspects is endowed with the above stated three qualifications. The
evidence is an essential property. And this evidence is not unestablished, since it
has been established that it is possible to engage in this kind [ of meditation] due
to the fact that a compassionate person is desirous [of engaging in it]. Nor is it
inconclusive, for the following reason: the fact of possessing a clear appearance
is that which is to be proved (siidhya); [this predicate applies to] the subject
nABS 23.2 240
implication is that the particular that the yogi knows in meditating on
emptiness is not all dharmas in a literal sense, but rather only the
momentary mental consciousness that takes the of all
dharmas' as its meditational object.
26
To be sure, the authors believe the
perfection of this form of yogic perception to be deserving of the name
omniscience, both because the emptiness that is directly known therein is
non-different from the emptiness of all things, and because the arisal of
that clear appearance eliminates all traces of ignorance.27 But no equivo-
cation in the term 'all dharmas' occurs, whereby the yogi starts off
meditating on a mentally constructed image and ends up by knowing
literally everything. .
Like Aryadeva and Candratirti, the authors of the TSIP embrace the
knowing all through knowing one approach to omniscience, where
knowledge of emptiness equals knowledge of all. The difference is that
for and Kamalaslla, knowledge of all results from the
knowledge not of the emptiness of just any dharma, but from the
knowledge of emptiness as it pertains to one dharma in particular: the
mental image 'all dharmas.'28 Since for these thinkers a mental image is
(dharmin) which is a mental consciousness (manovijfiana) that takes as its object
the selflessness of all dharmas; [this is to be proved] by means of the evidence,
namely that [the subject, the mental consciousness] is mingled with a
meditation (bhtivana) endowed with the threefold qualification mentioned above;
and the pervasion (vyapti) of the above-stated establishing property (sadhana-
dharma) by that probandum (sadhyadharma) is established, since possessing a
clear appearance does not depend on any other cause. And therefore, the perva-
sion [of that mental consciousness] by omniscience is also indirectly established,
since that clear appearance possessed by the subject, i.e., the mind (manas) that
takes as its object the selflessness and so on that are possessed by (gata) all
things, is nothing other than the omniscience of that [person]. That is, since it has
been generally established that meditation is pervaded by a clear appearance of the
thing that is being meditated upon, therefore, by implication, it is indeed estab-
lishe,d [that the mind that is meditating on the selflessness of all dharmas] is
pervaded also by omniscience, since it is not possible for that clear appearance [of
the selflessness of all dharmas] to exist in anything other than the above stated
subject [i.e., the mind meditating on that very selflessness of all dharmas]."
26. This reading of and Kamalasila's position thus corresponds with
STEINKELLNER's findings concerning Jfianasrlmitra's understanding of the par-
ticular that is known in yogic perception. See STEINKELLNER (1978): 132-133.
27. and KamalaSila model their discussion of this point on Dharmakirti' s
PV2. See PEMWIESER (1991) for an edition and translation of a relevant portion
of the TSIP.
28. On the construction of the concept 'all' (sarva), see TS 1184-87 and TSP ad cit ..
McCLINTOCK 241
not different from the mind in which it appears, one might even say that
omniscience results from meditating on the emptiness of one's own
mind, when one's mind is configured in a certain way. and
KamalasTIa do not put things like this in the final chapter of the TSIP,
but perhaps they refrain from doing so on account of their rhetorical
stance. For at the end of that chapter, KamalasTIa declares that the proof
of omniscience was not undertaken from the authors' preferred Vijfiana-
vadin perspective, but rather only having provisionally adopted an
'externalist' (bahirarthava.din) perspective in order to convince those
"addicted to external objects."29 Much earlier in the TSIP, the authors do
speak from their preferred metaphysical perspective, in which the
individual natures (svabha.va) necessary to the conventional process of
knowing - i.e., the object known, the subjective knower, and so on - are
all progressively and thoroughly dismantled. In that context, the authors
present an alternative vision of omniscience, one that I am tempted to
call a knowing all through knowing none approach to omniscience.
3D
Does this mean that the demonstration based on in the
final chapter is a mere sophism, designed solely to get opponents like
Sama!a and Yajfia!a off their backs? I don't think so. Rather, what I
think we can say is that for and KamalasTIa, knowledge of
emptiness really does produce knowledge of all - but as there are
multiple ways of conceiving the nature of 'all,' there are equally
multiple fashions of understanding the nature of the knowledge of all
and also how that knowledge comes about through the knowledge of a
single thing.
29. Cf. TSP ad TS 3645: yeyam asmabhir vijiianavadasthitair nirakaracintii priig
akiiri sa sampratalJ'l bhavato mfmiilJ'lsakan prati bahir-
artham abhyupetya sarvajiie pratipadyamiine bhavatalJ'l bahirarthaviidiniilJ'l
katham api nopayujyata eva kartum I.
30. See especially TS 2048-2049: kalpapiidapavat sarvasalikalpapavanair muniJ:t I
akampye 'pi karoty eva lokanam arthasampadam II tenadarsanam apy iihuJ:t
sarve sarvavidalJ'l jinam I anabhogena II. Cf.
also TS 1852-1855. In these passages and the commentary on them,
and Kamalasila present a vision of Buddhahood in which teachings and other
verbal communications issue spontaneously from the Buddha who remains
utterly free from concepts. For a discussion of a similar model in CandrakIrti's
MAvBh, see DUNNE (1996): 548-550. For parallels in other Mahayana sources,
both Yogacara and Madhyamaka, see GRIFFITHS (1994) and MAKRANSKY
(1997).
JIABS 23.2 242
Abbreviations
CS
csv
D
CatuMataka of Aryadeva
Catu[zsatakavrtti of CandrakIrti
MA
MAv
MAV
MAP
MAvBh
MMK
NB
PrasP
PS/PSV
PV
PVin
TS
sDe dge edition of the Tibetan bsTan gyur
Madhyamakiila1Jlkiira of
Madhyamakiivatiira of CandrakIrti
Madhyamakiila1Jlkiiravrtti of
Madhyamakiila1Jlkiirapafijikii of KamalaSila
of CandrakIrti
Malamadhyamakakiirikii of Nagfujuna
Nyiiyabindu of DharmakIrti
Prasannapadii of CandrakIrti
Pramii1}asamuccaya and Pramii1}asamuccayav.rtti of Dignaga
Pramii1}aviirttika of DharmakIrti
Pramii1}aviniscaya of DharmakIrti
Tattvasa1Jlgraha of
TSP Tattvasa1Jlgrahapafijikii of KamalaSTIa
Bibliography
Bhattacharya, Kamaleswar
1974 "Note on the Interpretation of the Term siidhyasama in Madhyamaka
Texts," Journal of Indian Philosophy 2: 225-230.
Blihnemann, Gudrun
1980 Der Allwissende Buddha, ein Beweis und seine Probleme: Ratnakfrtis
Sarvajiiasiddhi, ubersetzt und kommentiert. Wiener Studien zur
Tibeto1ogie und Buddhismuskunde 12, Universitat Wien, Vienna.
>-)
Dreyfus, Georges
1997 Recognizing Reality: Dharmakfrti's Philosophy and Its Tibetan
Interpretations. Albany: State University of New York Press.
Dunne, John D.
1996 "Thoughtless Buddha, Passionate Buddha," Journal of the American
Academy of Religion 64.3: 525-556.
1999 Foundations of Dharmakfrti's Philosophy: A Study of the Central
Issues in his Ontology, Logic and Epistemology with Special Attention
to the Svopajiiavrtti. Unpublished Ph.D. dissertation, Harvard
University.
Eckel, Malcolm David
1992 To see the Buddha: A philosopher's Quest for the Meaning of
Emptiness. San Francisco: Harper San Francisco.
McCLINTOCK 243
Griffiths, Paul J.
1994 On Being Buddha: The Classical Doctrine of Buddhahood. Albany:
State University of New York Press.
Hattori, Masaaki
1980 "Apoha and Pratibha," in Sanskrit and Indian Studies in Honor of
Daniel H.H. Ingalls, ed., M. Nagatomi, et. al. Dordrecht and Boston:
D. Reidel Publishing Co.
Hayes, Richard P.
1988 Digniiga on the Interpretation of Signs. Studies of Classical India 9.
Dordrecht: Kluwer Academic Press.
Jha, Ujjwala Panse
1998 "Kumarila Bhana on Yogic Perception," Journal of the Indian Council
of Philosophical Research 15.3: 69-78.
Lang, Karen
1985 Aryadeva's Catuf:tsataka: On the Bodhisattva's Cultivation of Merit
and Knowledge. Copenhagen: Akademisk Forlag.
Makransky, John J.
1997 Buddhahood Embodied: Sources of Controversy in India and Tibet.
Matilal, B.K.
1974
Albany: State University of New York Press.
"A Note on the Nyaya Fallacy Siidyasama and Petitio Principii,"
Journal of Indian Philosophy 2: 211-224.
Oberhammer, Gerhard
1984 Wahrheit und Transzendenz: Ein Beitrag zur Spiritualitiit des Nyiiya.
Vienna: Verlag der Osterreichischen Akademie der Wissenschaften.
Pemwieser, Monika
1991 Materialien zur Theorie der yogischen Erkenntnis im Buddhismus.
Phillips, Steven
1995
Unpublished M.A. thesis, University of Vienna.
Classical Indian Metaphysics: Refutations of Realism and the
Emergence of "New Logic." Chicago and La Salle, Illinois: Open
Court.
Prvereau, Raynald
1994 Dharmakfrti's Account of Yogic Intuition as a Source of Knowledge.
Unpublished M.A. Thesis, McGill University.
Seyfort Ruegg, David
1981 "Autour du ITa ba'i khyad par de Ye ses sde (version de Touen-
Houang, Pelliot TiMtain 814)," Journal Asiatique269: 208-229.
nABS 23.2 244
Steinkellner, Ernst
1978 "Yogische . Erkenntnis als Problem in Buddhismus," in
Transzendenzeifahrung: Vollzugshorizont des Heils: Das Problem in
indischer und christlicher Tradition, ed. G. Oberhammer. Vienna: De
Nobill Research Library, pp. 121-134.
Thrasher, Allen Wright
1978 "Mal,lgana Misra's Theory of Vikalpa," Wiener Zeitschrift jar die
Kunde Sudasiens 26: 133-157.
LINDA PENKOWER
In the Beginning ... Guanding 1-1M (561-632)
and the Creation of Early Tiantai*
The Mahe zhiguan ."i1iiJJl:.1li! (The Great Calming and Contemplation),
the comprehensive meditation text that lays out the perfect and sudden
(yuandun) approach to buddhahood as envisioned by Zhiyi
(538-597), the principal founder of Tiantai :Kt' Buddhism,! is revered
as one of three Zhiyi texts that came to encapsulate the essence. of the
tradition.
2
The work is the result of a series of lectures given by Zhiyi
* I am grateful to the National Endowment for the Humanities for assistance in
completing this article.
1. The most exhaustive study on Zhiyi remains SATO Tetsuei's Tendai
daishi no kenkyu: Chigi no chosaku ni kansuru kiso-teki kenkyu 72il::;kJiliO)liJf:G
o (Kyoto: Hyakkaen 1961) and its 1981 sup-
plement (Zoku Tendai daishi no kenkyu: Tendai Chigi 0 meguru sho-mondai
f.!Jt 0 72i:lj;:eiliO)liJfJ\; 0 :a-1!f)('Q Biographies appear in Leon
HURVITZ, "Chih-i (538-597): An Introduction to the Life and Ideas of a Chinese
Monk," Melanges Chinoises et Bouddhiques 12 (1960-1962), and KYODO Jik6
R:Fn.yc,TendaidaishinoshOgai Regurus bunko 38 (Tokyo:
Daisan bunmeisha 1975).
2. Mohe zhiguan, T 46, no. 1911. The other two Zhiyi works which together with
the Mohe zhiguan came to be described as the three great texts of Tiantai 72il
(Tiantai sandabu) or the great texts on the Lotus (fahua
sandabu) are the Fahua xuanyi (full title: Miaofa lianhuajing xuanyi
or Profound Meaning of the Lotus Sutra), T 33, no. 1716, and
the Fahua wenju $.XtJ (full title: Miaofa lianhua jing wenju
or Words and Phrases of the Lotus Satra), T34, no. 1718. Neal DONNER
and Daniel B. STEVENSON have translated the first chapter of the Mohe zhiguan
as part of their The Great Calming and Contemplation: A Study and Annotated
Translation of the First Chapter of Chih-i' s Mo-ho chih-kuan, Kuroda Institute,
Classics in East Asian Buddhism (Honolulu: University of Hawai'i Press 1993).
Paul L. SWANSON is working on a complete translation of the text called the The
Great Cessation-and-Contemplation (Tokyo: Rissho Koseikai, forthcoming),
portions of which can currently be found at the URL: http://www.nanzan-
u.ac.jp/shubunkenlrnhck.html. Significant portions of the Fahua xuanyi have also
been translated by SWANSON in his Foundations ofT'ien-t'ai Philosophy: The
Flowering of the Two Truths Theory in Chinese Buddhism (Berkeley: Asian
Humanities Press 1989).
Journal of the International Association of Buddhist Studies
Volume 23 Number 2.2000
JIABS 23.2 246
and recorded. and edited by his disciple and principal recorder and
official biographer, Guanding (561-632), at the Yuquansi
(Jade Spring Monastery) in Jingzhou prefecture, Hubei province over
the course of the summer retreat of 594, just three years prior to the
death of Zhiyi. Gracing the opening lines of Guanding's introduction to
this text is the equally celebrated, first known substantial attempt to
chronicle the religious history of Zhiyi and connect it to its Indian roots
(as it applies to text and teachings. This conception of a line of descent,
along with doctrinal teaching and cult, came to shape the way in which
Tiantai would come to think about itself and its ancestry so much so that
the history of the tradition in large part must be approached and
appreciated in terms of what it meant and means to a practitioner to
stand in the line of this transmission. The appearance of the statement of
lineage at the head of Guanding's introduction to the Mohe zhiguan is
thus considered by modem scholarship to be a watershed in the history
of religious ideas to the extent that it constitutes one of the defining
elements in the development of group identity and sectarian conscious-
ness in Chinese Buddhism during the medieval period. Developed
gradually over the course of the Tang (618-907) and solidified with the
advent of the thirteenth century Tiantai sectarian histories, which had as
their agenda the establishment of a Tiantai orthodoxy, Guanding's
rendition of the creation of Tiantai became the standard bearer for the
tradition, eventually standing at the head of a line of transmission that
extended into the Song (907-1289).3
3. The two most representative Tiantai sectarian histories that survive from the Song
are the Shim en zhengtong (Orthodox Lineage of Buddhist Schools;
XZJ 130) compiled by Zongjian *iJ: and completed in 1237, and especially the
Fozu tongji (Record [of the Lineal Transmission] of Buddhas and
Patriarchs; T 49, no. 2035) compiled by Zhipan and completed in 1269.
Patterned after the format and categories of Chinese dynastic histories, these
histories reveal their sectarian biases by organizing the various biographies into
such categories as the biographies of the patriarchs, the disciples of the patriarchs,
dharma heirs descended from a particular dharma master, Tiantai monks of
unknown lineage, dissenters of the so-called party line, and so forth. For the
basic format and structure of the Song histories see TAKAO Giken
Sodai Bukkyoshi no kenkya (Kyoto: Hyakkaen 1975),
pp. 139-48, and SHlNOHARA Koichi, "From Local History to Universal History:
The Construction of the Sung T'ien-t'ai Lineage," in Buddhism in the Sung,
edited by Peter N. GREGORY and Daniel A. GETZ, Jr., Kuroda Institute, Studies
in East Asian Buddhism, no. 13 (Honolulu: University of Hawai'i Press 1999),
pp.524-576.
PENKOWER 247
The construction of lineage of course is not particular to Chinese
Buddhism. Rather, throughout much of Chinese secular (read: Confu-
cian) history, and certainly during the medieval period, history has been
approached through and organized around genealogy and biography
imbued with moral instruction and imperative, a melding, if you will, of
history and hagiography. Nevertheless, and indeed especially because
Tiantai scholarship, following the lead begun by historians of the Chan
tIf! tradition upon the discovery of lost texts in the Dunhuang manuscript
collections, is just beginning to appreciate the inherent biases and
selective memory that went into compiling the sectarian histories and
their posthumous and somewhat arbitrary notion of lineage, it is
incumbent upon us to reconsider the various elements - soteriological,
political, polemical, and personal - that came together in the late-Sui
(589-618) and early-Tang to produce an interest in the formulation of
lines of descent and in the promotion of various mythic ethos in which
to groundZhiyi.
4
For religious narratives, from the beginning, were no
more exclusively bound by the rules of history than were later sectarian
lineal accounts, but rather were part of and responsive to contemporary
notions of the path as well as socio-cultural concerns. In the process, the
notion of lineage itself gradually took shape.
That is the real starting point of the present discussion. For this article
argues that the historical narrative that opens the Mohe zhiguan - as well
as a diversity of competing and complementary narratives that date from
the early days after the demise of Zhiyi - predate the notion of Tiantai
4. On the Chan side, see such representative works as YANAGIDA Seizan
Shoki Zenshu shisho no kenkyu (Kyoto: H6z6kan 1967);
John R. MCRAE: The Northern School and the Formation of Early Ch'an
Buddhism, Kuroda Institute, Studies in East Asian Buddhism, no. 3 (Honolulu:
University of Hawai'i Press 1986); Bernard FAURE: The Rhetoric of Immediacy:
A Cultural Critique of Chan/Zen Buddhism (Princeton: Princeton University
Press 1991); and T. Griffith FOULK: The 'Ch'an School' and its place in the
monastic tradition, Ph.D. dissertation (University of Michigan 1987). For Tiantai
see OKUBO Ry6jun "T6dai ni okeru Tendai no densh6 ni tsuite
Nippon Bukkyo gakkai nenpo
17 (June, 1951): 87-99; my T'ien-t'ai during the T'ang Dynasty
Chan-jan and the Sinification of Buddhism, Ph.D. dissertation (Columbia
University 1993); Daniel B. STEVENSON: ''The Status of the Mo-ho chih-kuan in
the T'ien-t'ai Tradition," in DONNER and STEVENSON: The Great Calming and
Contemplation, pp. 31-61; and Jinhua CHEN: Making and Remaking History: A
Study of Tiantai Sectarian Historiography, Studia Philologica Buddhica Mono-
graph Series 14 (Tokyo: The International Institute for Buddhist Studies: 1999).
nABS 23.2 248
as a discrete entity or tradition. As such, their initial use and function lay
elsewhere and their subsequent manipulation in the service of bids to
control the contemporary dialogue developed along with that of group
consciousness. How then did Guanding understand his rendition of
Zhiyi's line of descent in the Mahe Zhiguan? This article is divided into
four sections and argues that the notion of lineage presented in
Guanding's introduction to the Mohe Zhiguan is best understood as part
of a set of inter-related intra- and inter-group circumstances that
postdates Zhiyi and had as much to do with issues of property, politics,
and personality as it did with doctrinal and practical concerns. Section
one introduces Zhiyi's religious history as it appears in Guanding's
preface to the Mohe zhiguan, paying particular attention to the historical
antecedents of this statement on lineage and to the types of relationships
implied between and among monks in this early narrative. Section two
discusses this as religious paradigm, focusing on the ways in which
sacred biography works in tandem with and is itself transformed by
religious ideas of the path. Beginning with a discussion of the construc-
tion process of the lineage statement that now opens Guanding's intro-
duction to the Mohe zhiguan, sections three and four further contex-
tualize the notion of lineage against the backdrop of competing and
complementary mythic tropes that were circulating among Zhiyi's
disciples. In section three the discussion shifts to examining the reasons
for the appearance of a lineage statement at that particular juncture of
Chinese Buddhist history, juxtaposed in particular to other regional
efforts to memorialize the recently deceased Zhiyi. In section four
consideration turns to the question of why the religious history of
Tiantai.is traced back to India in the particular figuration advanced in
the Mohe zhiguan, extending outward to include the religious landscape
in which it developed and inward to include motivations of a more
personal nature. Taken together this article hopes to portray a sense,
often overlooked, of the intentionalist role Guanding played in the
creation of a Zhiyi legacy during the transitional years between the
consolidation of the Sui and the ascension of the Tang.
Guanding's Introduction to the Mohe zhiguan
The account of Zhiyi's pedigree that Guanding presents in his introduc-
tion to the Mohe zhiguan is actually comprised of two distinct yet
PENKOWER 249
related lines - one western, the other eastern.
5
The first or western line
begins with Sakyamuni Buddha and continues through a series of
monks, each of whom is guaranteed his place in the line by having
"received [predication of future buddhahoodJ from the golden mouth"
(jinkou) 'of the Buddha, the name by which this line ultimately
came to be known. Sakyamuni transmitted the teachings (dharma) $11
(fafu) to Mahakasyapa, who in turn transmitted them to Ananda and so
forth down the line through SaI).avasa -+ Upagupta -+ Dhrtaka -+
Miccaka -+ Buddhanandi -+ Buddhamitra -+ Parsva -+ PuI).yayasas -+
-+ Kapimala -+ Nagarjuna -+ KaI).adeva -+ Rahulata -+
Sailghanandi -+ Sanghayasas
6
-+ Kumarata -+ J ayata -+ Vasubandhu -+
Manorhita -+ Haklenayasas, and -+ Sirpha, for a subtotal of twenty-three
personalities.
7
Madhyantika, reputedly an older contemporary of
SaI).avasa is then inserted between Ananda and his younger dharma
brother, bringing the total number of teachers in Guanding's rendition
of the western line to twenty-four. The last member of this line was
reputedly killed by an anti-Buddhist monarch in south India during a
persecution of Buddhism. Dying without heir, the line was effectively
cut off at that time.
8
No direct connection is thus drawn or suggested
between the members of this line and the past teachers of Zhiyi who are
listed in Guanding's second and Chinese line.
5. Guanding's statement on lineage is located in the Mohe zhiguan, T46, no. 1911:
1a13-cl. His introduction to the Mohe Zhiguan is translated as part of DONNER
and STEVENSON, The Great Calming and Contemplation. The lineage account is
found on pp. 100-107. See also SWANSON's The Great Cessation-and-Contem-
plation.
6. Following DONNER and STEVENSON, The Great Calming and Contemplation,
p.103. Philip B. YAMPOLSKY, The Platform Sutra of the Sixth Patriarch (New
York: Columbia University Press 1967), p. 8, gives Gayasata; SWANSON, The
Great Calming-and-Contemplation, note 63, gives Saflgayasata.
7. For brief sketches of these personalities taken from the Fu fazang yinyuan zhuan
fi$iZlZSliMt1.\lJ: (History of the Transmission of the Treasury of the Dharma;
T 50, no. 2058) on which Guanding's line is based see SWANSON, The Great
Cessation-and-Contemplation, notes 18-75 passim.
8. The mid-Tang Tiantai luminary Zhanran (711-782), who wrote the fIrst
commentary on the Mohe zhiguan, was the first to nominalize and designate this
line as the "ancestral / patriarchal transmission of the golden mouth" :slZ D :fll.*
(jinkou zucheng) in his Fuxing lilIli'J (full title: Zhiguanjuxing chuanhong jue
or Decisions on Supporting Practice and Broadly Dissemi-
nating [the Teachings of The Great] Calming and Contemplation), T 46,
no 1912: 145a22 and 147c5.
JIABS 23.2 250
Guanding takes the western line and his abbreviated biographical
sketches of its members from the Fu Jazang yinyuan zhuan, an
apocryphal text whose translation from the Sanskrit is ,traditionally
attributed to Kekaya and Tanyao _Hi (n.d.), active in the China of the
Northern Wei during the latter half of the fifth century. Modern scholar-
ship, however, has successfully proven that the Fu Jazang yinyuan zhuan
was compiled in China in response to a brutal attack by the Confucian
literatus-official Cui Hao (381-450) that there was no historical
record of the existence of the Buddha or a living Buddhist tradition in
India.
9
This came in the wake of the Buddhist purges suffered under the
Northern Wei in 446, to which the demise of the western line at the
hands of a south Indian king (and through no fault of the members of
the line itself) was but a veiled illusion. 10
9. On the issue of the apocryphal nature of the Fu Jazang yinyuan zhuan see Henri
MASPERO, "Sur la date et l'authenticite du FouJa tsang yin yuan tchouan," in
Melanges d'lndianisme offerts par ses eleves a M. Sylvain Levi (Paris 1911),
pp. 129-149; YAMADA Ryiij6 Daijo Bukkyo seiritsu-ronjosetsu **
(Kyoto: Heirakuji shoten 1959), p.583; and MOCHIZUKI
Shink6 J'3 ffi 7, Bukkyo daijiten 5: 4493c-4494b. While some
modern scholars feel that there may have been some basis in history for a south
Indian purge of Buddhism that could have ended the line set forth in the Fu
Jazang yinyuan zhuan, YAMADA Meiji i-U B3IJF.[M (and others) has successfully
argued against the veracity that the violent but short-lived reign of terror of the
HiilJ.a (a branch of the Iranian-speaking Hephthalities then ruling in Bacteria) in
the late-fifth century extended beyond a limited area of northwest India and
Kashmir. See his "Mihirakura no ha-butsu to sono shiihen
.:cO))jIij two parts, Bukkyo shigaku 11.1 (August, 1963): 44-58 and
11.2 (November, 1963): 40-56. For a summary of this period of history see Jan
NATTIER, Once Upon a Future Time: Studies in a Buddhist Prophecy oj Decline
(Berkeley: Asian Humanities Press 1991), pp. 110-113. The best English
language source on this genre of literature is Chinese Buddhist Apocrypha, edited
by Robert E. BUSWELL, Jr. (Honolulu: University of Hawai'i Press 1990).
10. Cui Hao, along with his Daoist associate, Kou Qianzhi (d. 448), was
instrumental in instigating an anti-Buddhist campaign, which culminated in the
446 persecution of Buddhism under Emperor Tai Wu ::tj\ffl of the Northern
Wei (r. 424-451). As part of his crusade to sinicize (read: Confucianize) the non-
Chinese Toba r1iNtt, Cui not only recommended the wholesale execution of all
monks (which was even too extreme for Kou) and the destruction of all temple
property, but also insinuated that there really was no Buddhism even in India. A
summary of the events leading up to and including the persecution is presented in
Kenneth CH'EN, Buddhism in China: A Historical Survey (Princeton: Princeton
University Press 1964), pp. 147-151. The most authoritative treatment of
Buddhism under the Northern Wei remains TSUKAMOTO Zenryii
PENKOWER 251
Fu Jazang yinyuan zhuan
In this, the ethos embodied in account of the construction and demise of
the lineal succession of the teachings in India, shares the Chinese secular
(read: Confuci,;m) penchant for mapping genealogical histories and then
bringing the moral weight and consequence implied in those histories to
bear on current circumstances and vicissitudes.
Equally telling in regard to the fifth century north Chinese origins of
this Indian lineage account is the choice of personalities included within
its ranks. By the fifth and sixth centuries, Chinese Buddhists, having
reached a level of religious sophistication that had come to recognize the
limitations and potential for error inherent in the earlier practice of
approaching Buddhist scripture through the lens of Chinese indigenous
traditions, particularly dark learning (xuanxue), had turned instead
towards a direct approach to the teaching examined on its own terms.
Grappling with how best to accurately adopt and faithfully communicate
the received tradition (whether or not that was always successful is
besides the point here), early medieval Chinese Buddhists set aside such
texts as the Laozi and Zhuangzi ru:r in favor of the interpretative
literature of the great Indian and Central Asian exegetes. Among the
most notable Indian figures whose works and traditions were popular
points of departure for Chinese Buddhist discussions on doctrine begin-
ning during this period were Nagarjuna, and Vasubandhu,
all of whom hold prominent positions among the list of twenty-three
personalities in the Fu Jazang yinyuan zhuan.
11
Shina Bukkyi5shi kenkyii (Tokyo: Kobundo shobo 1942;
reprint, Tokyo: Shimizu Kobundo 1969), pp. 335-609. See also TANG Yong-
tong: Han Wei Liangjin Nanbeichao jojiao shi, vol. 2 (1938; reprint Taibei:
Shangwu yinshuguan 1979), pp. 52-93.
11. Among several characteristics that distinguish the early medieval Chinese
Buddhist traditions is that they centered around the Chinese translation of a
specific Indian text or group of texts which, for the most part, belonged to the
exegetical (lunzang) division of the canon. The Madhyamaka-oriented
Sanlun (Three Treatises) and Silun (Four Treatises) traditions were
based on the writings of Nagarjuna and respectively named after three and four
Madhyamaka treatises, all of which were reputedly translated into Chinese by
KumarajIva in the early-fifth century. The Sanlun tradition takes its name from
Nagarjuna's Miilamadhyamaka-kiirikii, known in Chinese as the Zhonglun
(Middle Treatise; T30, no. 1564), the Shiermen lun traditionally at-
tributed to NagiiIjuna (Dvadasanikiiya siistra; T 30, no. 1568), and the Bai lun
(Hundred Treatise), a commentary by a certain master Vasu on some verses
JIABS 23.2 252
While there are slight discrepancies between the names given in the Fu
fazang yinyuan zhuan and those found in the Mahe zhiguan, the only
difference between the two succession lines of any is that
Guanding, by including Madhyantika, allows for two dharma heirs of
the same generation in his line, whereas the Fu fazang yinyuan zhuan
relegates Madhyantika, as a contemporary of Sal).avasa, to a subsidiary
line.
12
The latter text hence lists only twenty-three personalities in its
line, a discrepancy that Guanding himself acknowledges, and which he
elsewhere straddles, referring to Nagarjuna in one instance as the
"thirteenth teacher," the position that monk occupies in the line of
twenty-three teachers of the Fu fazang yinyuan zhuan.l
3
by A.ryadeva (Sata-sastra; T30, no. 1569). To these, the Silun tradition adds the
Dazhidu [un *:l& J3tfffH (Treatise on the Perfection of Wisdom Siitra), attribut-
ed to Nagiirjuna (T25, no. 1S09). The tathiigatagarbha- and Yogaciira-oriented
Dilun J:{gfffH and Shelun :!l]\ffij traditions were based on the works of
Asailga, and Vasubandhu. Yogaciira teachings were introduced to Chinese
Buddhists during the sixth century through the translation of Vasubandhu's
commentary on the Dasabhumika-sutra, known in Chinese as the Shidi jing lun
(Treatise on the Ten Stages Siitra) and popularly referred to as the
Dilun (T26, no. 1522), and later through the translations of Asailga's Mahii-
yiinasa1!lgraha-siistra, called the She dacheng lun :!l**]\ffij or Shelun (T31,
no. 1593), and Vasubandhu's known in Chinese as
the She dacheng [un shi (T31, no. 1595). The apocryphal Dacheng
qixin lun **A1HEl"]\ffij (Awakening of Faith in the Mahayana) attributed to Asva-
in particular, continued to greatly influence Chinese Buddhism, and
became in effect the special property of the Huayan 'Ii I!i tradition (T 32,
no. 1666). Stanley WEINSTEIN gives an overview of the shared characteristics of
the traditions that evolved during the fifth and sixth centuries as compared to
those of the "new Buddhism" that developed during the Sui-Tang (589-907) in
his "Imperial Patronage in the Formation of T' ang Buddhism, " in Perspective on
the T'ang, edited by Arthur F. WRIGHT and Denis TWITCHETT (New Haven:
Yale University Press 1973), pp. 265-274.
12. See YANAGIDA, Shoki Zensha shisho no kenkyu, pp. 369-370, and the chart at
the back of the book comparing the names included in the Fu fazang yinyuan
zhuan line and those of the Mohe zhiguan. It should be noted of course that
Mahakasyapa and Ananda are both considered one of the ten major disciples of
the Buddha. According to tradition, after the Buddha's demise, at the first council
in Rajagrha, Mahakasyapa first asked .Ananda to recite the sermons of the
Buddha (sutra), Upali the precepts (vinaya), and Mahakasyapa himself dealt with
the treatises (abhidharma). Fufazang yinyuan zhuan, T 50, no. 2058: 300bS-6.
l3. Mohe zhiguan, T 46, no. 1911: 1b7-8 and 1b25. This discrepancy is never quite
resolved in later Tiantai. The number of masters in the western line is sometimes
given as twenty-three, at other times as twenty-four. Nagiirjuna who is singled
PENKOWER 253
Despite the obvious break with a "string of pearls" genealogical
format, which allows for only a single head per generation and is
generally favored by the Chinese secular (and later Buddhist) world, the
inClusion of Madhyantika between Ananda and SaI].avasa is not without
textual precedent. Beginning possibly as early as the fourth century but
certainly from the early-fifth century onwards, Chinese Buddhists were
familiar with the idea of transmission theories through lists found
primarily in translations of Indian and Kashmiri meditation and vinaya
texts. Judging by the gradual increase in the level of sophistication in the
language used to describe the transmission process as well as by the
number of anomalies and deviations found between and among texts
purportedly representing the same line found in these texts, the construc-
tion of lineage statements as a means by which to define and chronicle
religious history was itself only gradually taking shape among western
exegetes during this time. Thus, for example, the earlier of two Chinese
translations of the Ayuwangzhuan Ifrillf::E.1t (Legend of King Asoka),
the bulk of which was translated into Chinese as early as 306 by An
Fagin does not include Madhyantika in its main transmission,
agreeing with the Fu Jazang yinyuan zhuan.
14
The acceptance of this
monk as a member in the line of descent from the Buddha seems to have
entered or at least become popular in China through the Sarvastivadin
line of the important Kashmiri translator and meditation master,
Buddhabhadra (359-429). Madhyantika, for instance, is sandwiched
between Ananda and SaI].avasa in Buddhabhadra's Damoduoluo chan
. jing (Meditation Sl1tra of Dharmatrata), translated into
Chinese in the early part of the fifth century, as well as in the prefaces to
it written by Huiyuan rim (334-416) and his contemporary Huiguan
riD (n.d.), all of which predate the writing of the Fu Jazang yinyuan
zhuan by about half a century.1
5
The gradual increase in the acceptance
out in this line is therefore either counted as number thirteen or number fourteen,
depending upon which rendition of the line is used.
14. T 50, no. 2042: 114a26-116cl8. MOCHIZUKI, Bukky6 daijiten 1: 93a questions
the fourth century introduction of this text.
15. Damoduoluo chanjing, T 15, no. 618: 301c. This line later became an important
reference for the development of the Chan transmission. See McRAE, The
Northern School and the Formation of Early Ch'an Buddhism, pp. 80-82. The
pertinent portion of the Huiyuan list appears in the Damoduoluo chan jing, T 15, .
no. 618: 301a, and in the Chusanzangjiji (Collection of Notes on
the Translation of the Tripi!aka), T 55, no. 2145: 65c9-12. That of the Huiguan
preface is included in the latter collection; T 55, no. 2145: 66c5-9. Both prefaces
JIABS 23.2 254
of the inclusion of Madhyantika in lineage statements deriving from the
Buddha appears to have continued in China in the years between the
writing of the Fu Jazang yinyuan zhuan and the time Guanding penned
his list in the Mahe zhiguan, as can be seen by a number of early-sixth
century texts. Thus, for example, unlike the earlier Aaka translation,
the Sanghabhadra translation of the Ayu wang jing completed
in 512 includes Madhyantika in its main line of succession,16 as do two
lists (or variants of the same list) describing Buddhabhadra's Sarvasti-
vadin line of descent.!? Guanding's insertion of Madhyantika into the Fu
Jazang yinyuan zhuan line seems to have been influenced by is in
conformity with this increasingly popular early medieval convention.
Despite that the western line is cut off in India with the untimely death
of its heirless member, SiIpha, and thus no direct connection exists
between that line and the one begun in China, Guanding nonetheless
insists that the origins of the Chinese line are grounded in and can come
to be known through the western transmission of the teachings. "If a
practitioner hears the transmission of the treasury of the dharma {':Ii'!
(fu Jazang)," Guanding proclaims, "he will recognize the origins of
[our ancestral] line * (zang)."18 On the basis of what does he make this
claim?
The second or eastern line presented by Guanding in his introduction
to the Mahe zhiguan stands in sharp contrast to the first, and suggests a
relationship between the monks in this line distinct from that between
the Indian luminaries mentioned above. Unlike the western line which
proceeds outward from Sakyamuni, the eastern line follows the Chinese
secular and especially imperial genealogical style, which begins with the
founder of the clan or dynasty and works backwards through the
ancestors with one head per generation. The narrative that records the
origins of the Mahe zhiguan thus starts in China with Zhiyi, and is
omit Mahaklisyapa from their lists; Huiguan also omits Upagupta, demonstrating
the somewhat arbitrary nature of these genealogies.
16. T 50, no. 2043: 152c15-l8 and l69c14-17.
17. The fIrst, which omits Mahakasyapa and lists fIfty-three personalities, is located
in the Chu sanzang jiji, T 55, no. 2145: 89a20-b29; the second, which includes
him, follows immediately on 89c4-90alO.
18. Mohe zhiguan, T46, no. 1911: la13. This line can also be translated, "If a
practitioner hears the [History of the] Transmission of the Treasury of the
Dharma ... ," that is, as a direct reference to the Fu jazang yinyuan zhuan.
PENKOWER 255
traced backwards through Huisi (515-577)19 to Huiwen (ac-
tive during the mid-sixth century).20 Rather than rest on the transmis-
sion of dharma as does the western line; this moreover is a transmission
of meditation technique. So, we are told, Zhiyi's setting forth of the
teaching of calming and contemplation in the Mohe zhiguan is nothing
more or less than the method of cultivating mind that was on to
him by Huisi, who in tum had received it from Huiwen. All monks of
the eastern line are designated simply as "shi" or teachers. Huisi
"attended or served" (shi) Huiwen; Zhiyi did likewise under Huisi. Zhiyi
"spread l' (chuan) Huisi's three kinds of calming and contemplation -
gradual and sequential i$f* (jianci), variable ::fJE (buding), and
perfect and sudden (yuandun)"21 as handed down by Huiwen. Hui-
wen, portrayed as an unusually gifted and enigmatic ascetic, is said to
have applied his mind exclusively to the Dazhidu [un 1lRlJ (Treatise
on the Perfection of Wisdom Sutra), purportedly a work by Nagar-juna
(active in India during the mid-second century), and to have devised his
method for cultivating the mind through insight into the writings of this
long-deceased western teacher.22 Guanding thus in no way implies that
19. Paul MAGNIN has produced a full-length study on Huisi entitled La vie et
l'oeuvre de Huisi (515 577): Les origines de la secte bouddhique chinoise du
Tiantai, Publications de l'Ecole Franr;:aise d'Extreme-Orient, no. 116 (Paris:
Ecole Franr;:aise d'Extreme-Orient 1979). For more on Huisi see also SATO, Zoku
Tendai daishi na kenkyu, pp. 136-300.
20. There is no separate biography for Huiwen in the XGSZ.
21. Mahe zhiguan, T 46, no. 1911: lc1-2. Zhiyi expounded on each of these three
different approaches to the practice of calming and contemplation in three separate
works. The gradual and sequential approach is discussed in his Shi chan baluami
cidi famen (Elucidation of the Graduated Approach of
the Perfection of Dhyana; T 46, no. 1916). More popularly known as the Cidi
chanmen this text derives from a set of lectures delivered in 571,
early in Zhiyi's career, taken down by Fashen (n.d.), and later edited by
Guanding. The variable approach is briefly outlined in his Liumiaa famen
(Six Wondrous Teachings; T46, no. 1917). The Mahe zhiguan,
delivered towards the end of Zhiyi's career, summarizes the perfect and sudden
calming and contemplation.
22. T 25, no. 1509. The Dazhidu lun is a commentary on the Mahe banrua baluami
jing (Paiicavif!lsati-silhasrikil-prajiiilpilramitil-sutra or Per-
fection of Wisdom Siitra in 25,000 lines; T 8, no. 223), traditionally attributed to
Nagiirjuna as translated into Chinese by KumarajIva in 405. Modern scholarship,
however, is in general agreement that it was actually composed in China by its
reputed translator. Richard H. ROBINSON summarizes the reasons for this
assessment in his Early Mildhyamika in India and China (Madison: University
JIABS 23.2 256
the relationship between the monks in this line was anything more than
one whereby an orthodox type of meditation based on Nagar-juna's
writings was taught one to the other.
The brief rehearsal of the eastern line ends by quoting Zhiyi, who
credits the system of the three meditative practices of calming and
contemplation to Nagarjuna and calls the latter his "teacher." Guanding
makes the connection between the two genealogies explicit by designat-
ing the Indian exegete both as "thirteenth teacher" of the western line
and the "high ancestor" of Tiantai, the designation "zu" in the sense of
founding ancestor being reserved for Nagarjuna alone. Finally, to
substantiate the claim that Zhiyi's threefold teaching of calming and
contemplation follows Nagarjuna, Guanding cites the famous verse in
Nagarjuna's Zhonglun (Middle Treatise), a key textual inspiration for
the formulation of the Tiantai three truth system of emptiness, provi-
sionality, and the middle. "All things which arise through causes and
conditions, I declare to be identical to emptiness. It is also a provisional
designation. It is also the meaning of the middle way."23
The inclusion of the first or western line in Guanding's introduction to
the Mohe zhiguan only now begins to make sense when viewed in rela-
tion to the eastern line. It seems that the main purpose of the first line
was to introduce Nagarjuna and secure his place in the second. Simply
put, this rendition of the religious history of Zhiyi rests upon Huiwen's
of Wisconsin Press 1967), pp. 34-39. The frrst thirty-one fascicles of this treatise
have been translated into French by Etienne LAMOTTE: Le Traite de La Grande
Vertu de Sagesse, frve volumes (Louvain: Universite de Louvain 1944-1981). As
is evident here, traditional Buddhism placed great stock in the belief that it was a
work by Nagfujuna.
23. Mohe zhiguan, T 46, no. 1911: Ib28-cl. cf. Zhonglun, T30, no. 1564: 33bll-12,
where wu (nonexistence) is used instead of kong !!f (emptiness). For a dis-
cussion of the development of the Tiantai three truths theory see SWANSON,
Foundations of T'ien-t'ai Philosophy. The formulation of the idea of empty,
provisional, and the middle is found on pp. 1-17, where SWANSON also
compares the Kumarajlva Chinese translation of this Zhonglun verse with the
Sanskrit original (found in the Miilamadhyamaka-kiirikii, chapter twenty-four,
verse eighteen), which is generally interpreted as a two truth theory of provisional
and ultimate. Zhiyi's three truth system introduces the middle truth as a third,
absolute truth that transcends and unifies the two truths of classical Indian
Madhyamaka. See NAKAMURA Hajime i:f:tt.t7i:, "ChUdo to kITken: Santaige no
kaishakunikanrenshitei:f:tmt!!f.m 0 G-C,"inYiiki
kyoju shOju kinen: Bukkyo shisoshi ronshii 0
(Tokyo: DaizD shuppansha 1964), pp. 139-180.
PENKOWER 257
mastery of the Dazhidu [un, from which - with::lUt the aid of a mentor -
he fashioned the design for his meditation technique(s). This trans-
historical accomplishment alone guarantees Huiwen's status as a direct
spiritual descendant of Nagarjuna. The line of [Nagarjuna -+] Huiwen -+
Huisi -+ Zhiyi moreover is synonymous with the history of the making
of the Mohe zhiguan and implies a second special spiritual relationship
(in addition to the one between Huiwen and Nagarjuna) between Zhiyi
and Nagatjuna as well.
24
So it would appear that, at this initial phase of
the construction of lineage, Guanding understood the relationship
implied by the dharma transmission of the monks on the first list as
something quite different from the diverse relationships that existed
between and among Huiwen, Huisi, Zhiyi (and himself).
In this [treatise] on calming and contemplation, Tiantai Zhizhe Wmi [Zhiyi]
explains the approach to the teachings (famen) that he practiced within his
own mind. When Zhiyi was born, light filled the room and double pupils
appeared in his eyes.
25
[Later] he performed the repentance [based upon] the
Lotus Sutra, [as a result of which] he manifested dharaqI.26 Then, taking the
24. Zhanran was the first to later nominalize and designate the western line as the
"ancestral/patriarchal transmission of the present teachers" UiJlIH.ll.1iK (jinshi
zucheng) in his Fuxing, T 46, no. 1912: 145a23, 147b19, and 147c5-6, where he
refers to it side-by-side with the transmission of the golden mouth, thus making
the connection between this two lines explicit. He goes on to state: "Had the great
sage-tathagata [Slikyamuni] not been listed fIrst, there would be no reason to list
the twenty-three patriarchs t.ll. (zu). Without the twenty-three patriarchs, intro-
ducing the thirteenth teacher ejjj (shi) [NagiiIjuna] would not make sense. Unless
the thirteenth teacher is introduced, believing in [Huiwen], Huisi,and Zhiyi is
unwarranted." Fuxing, T 46, no. 1912: 143: 10-13; cf. his Souyao ji (full
title: Mohe zhiguanjuxing souyao ji or Record of Selected
Essentials of the Fuxing), XZJ 99: 223a7-9. For more on Zhanran's development
and use of this and other early Tiantai creation stories discussed in this article in
his bid to forge a mid-Tang Tiantai vision see my "Making and Remaking
Tradition: Chan-jan's Strategies toward a T'ang T'ien-t'ai Agenda," in Tendai
daishi kenkyu: Tendai daishi sen yonhyaku-nen gokanki kinen shuppan
0 edited by Tendai daishi kenkyU
henshii i-inkai Kyoto: Tendai gakkai 1997,
pp. 1338-1289 (from back).
25. The legend that such an auspicious sign announced the birth of a sage-king finds
its roots in the Shiji (Book of History), where the sage-ruler Shun is
reported to have had double pupils.
26. The repentance based upon the Lotus refers to one of two forms of the lotus
samadhi or fahua sanmei $':::I!;jc, which together with the vaipulya (fang-
deng) repentance constitutes the half walking-half sitting samadhi
(banxing banzuo sanmei one of four kinds of samadhic prac-
nABS 23.2 258
place of the teacher [Huisi] from whom he received the teachings, he lectured on
the gold-lettered Prajiia[paramita Satra].2? The Chen (557-589) and Sui states
honored him and gave him the title of "imperial teacher." He died in the medita-
tion posture, having attained the stage of the five classes [of disciplehood]. .. 28
... Zhiyi served $ (shi) [Huisi] of Nanyue as his teacher. The latter's
religious practice was inconceivably profound. For ten years he did nothing but
recite scripture. For seven years he practiced the vaipulya [repentance].29 For
ninety days he sat constantly in meditation, and in a single instant attained perfect
realization. The dharma-gates of both the Mahayana and HIuayana radiantly burst
forth [for him].30
tices as described by Zhiyi. This form of the lotus samMhi is practiced over a
twenty-one day period and is based on the twenty-eighth chapter of Kumiirajlva's
Chinese translation of the Lotus Satra, entitled "Exhortations of the Bodhisattva
Samantabhadra" (T9, no. 262: 6la-62a), and on the Foshuo Guan puxian pusa
xingfa jing (Satra on the Practice of the Visualization
the Bodhisattva Samantabhadra; T 9, no. 277). For Zhiyi's synopsis of the lotus
samadhi in the Mohe zhiguan (T46, no. 1911: 14a5-b25) see DONNER and
STEVENSON, The Great Calming and Contemplation, pp. 261-269. STEVENSON
discusses this practice in his "The Four Kinds of Samadhi in Early T'ien-t'ai
Buddhism," in Traditions of Meditation in Chinese Buddhism, edited by Peter N.
GREGORY, Kuroda Institute, Studies in East Asian Buddhism, no. 4 (Honolulu:
University of Hawai'i Press 1986), pp. 67-72.
27. Zhiyi assisted in copying this scripture and the Lotus in gold letter during his stay
with Huisi on Mount Dasu j;:Jirlr, between the Chen and Northern Qi borders,
between 560 and 568.
28. The five preliminary levels of a disciple .li&bffi.:rlJi: (wupin dizi wei) are stages
of attainment in the Tiantai scheme through which one advances to buddhahood.
Ranked just below the fifty-two stages of the perfect path, they correspond to the
stage of identity in meditative practice W!i'T.!lP (guanxing ji), the third of six
identities, a system which also marks progress on the perfect path.
29. The vaipulya orfangdeng repentance is based upon the Dafangdeng tuoluonijing
(T21, no. 1339), translated into Chinese by Fazhong
(n.d.) during the Northern Liang. This repentance is one of the practices that
comprise the half walking-half sitting samadhi and emphasizes ritual purity and
procedures. For Zhiyi's synopsis of this repentance in the Mohe zhiguan (T 46,
no, 1911: 13a24-14a5) see DONNER and STEVENSON, The Great Calming and
Contemplation, pp. 248-261. STEVENSON discusses this repentance in his "The
Four Kinds of Samadhi in Early T'ien-t'ai Buddhism," pp. 61-67.
30. This is said to have come towards the end of a three-year tutelage under Huiwen,
on the last day of the three-month (ninety-day) summer retreat. cf. Guanding's
Sui Tiantai Zhizhe dashi biezhuan (A Separate Biography
of Zhizhe [Zhiyi], the Great Teacher of Tiantai during the Sui [Dynasty D, T 50,
no. 2050: 191c16-17, and Zhanran, Fuxing, T 46, no. 1912: 149a22-23, where he
states: "From that moment on, [Huisi's] eloquence and awakening were so
extensive that when he encountered scriptures he had never heard before he had
PENKOWER 259
Nanyue [Huisi] served (shi) the teacher of meditation (chanshi) Huiwen.
During the reign of Gaozu (r. 550-560) of the Northern Qi, [Huiwen]
wandered alone through [the region between] the Yellow and Huai rivers, his
approach tothe teachings unknown to his age.
3
! Indeed, people [daily] tread the
earth and ga:?e at the heavens
32
yet do not fathom the depths [of the former] or
the heights [of the latter].
Huiwen's mental discipline was entirely based upon the [Dazhidu] lun expound-
ed by Nagfujuna, the thirteenth teacher .=gjji (dishisan shi) in the line of
those who transmitted the treasury of the dharma (fu Jazang). [For this reason]
Zhiyi says in his Guanxin lun WH" (Treatise on the Contemplation of Mind):
"I entrust myself to the teacher Nagarjuna."33 By this we can verify that
Nagfujuna is the high ancestral teacher (gaozushi) [of our line]. .. 34
no doubts whatsoever about their contents and understood them perfectly without
the aid of explanations by others."
31. The term dubu 111 * literally means to wander alone, leaving the impression that
Huiwen was a lone and obscure practitioner, a depiction that certainly jibes with
the fact that virtually nothing is known about this monk. Even Zhanran,
commenting on this passage in his Fuxing (T 46, no. 1912: 149a18-19), feels
compelled to point out that "no biography exists relating the virtues and deeds of
master [Hui]wen." cf. his Souyao ji, XZJ 99: 227a 8-9. Although we cannot be
certain how Guanding intended the term, it seems also to have taken on the
meaning of unparalleled. Zhanran continues: "Because no one could compete
with him, [Guanding in his introduction to the Mohe zhiguan] says that 'he
walked alone.' [This means that] he was without equal in the region between the
Yellow and Huai rivers. Because his understanding was so profound, there was
no one who could fathom it." Souyao ji, XZJ 99: 227all-ll; paraphrased from
the Fuxing, T 46, no. 1912: 149a22-24.
32. From the Zuozhuan tc'Jli:, fifteenth year of the Duke of Xi See also
BURTON Watson, tr., Hsiin Tzu: Basic Writings (New York: Columbia
University Press 1963), p. 15: "If you do not climb a high mountain, you will not
comprehend the highness of the heavens; if you do not look down into a deep
valley, you will not know the depth of the earth ... "
33. T 46, no. 1920: 585c20, paraphrased. The Guanxin lun is Zhiyi's final oral
testament to his students delivered shortly before his death in 597.
34. Mohe zhiguan, T 46, no. 1911: IbI3-16, 20-27; DONNER and STEVENSON, The
Great Calming and Contemplation, pp. 104-107, with minor variations. The term
gaozu is a Chinese familial term meaning great-great-grandfather, an important
position genealogically as the custom in the Wei, Jin, Northern and Southern
Dynasties, Sui (and early Tang) was for founding rulers to establish four temples
for ancestors within the mourning circle (and an additional temple for the first
ancestor, a term that seems not to have been well understood). See Howard J.
WECHSLER, Offerings oj Jade and Silk: Ritual and Symbol in the Legitimation oj
the T'ang Dynasty (New Haven: Yale University Press 1985), p. 126. But, as
Zhanran points out, if Zhiyi is taken as the de Jacto head of the line then
nABS 23.2 260
Overarching tropes
What does this twofold genealogy tell us about the vision Guanding was
here attempting to forge of models of and for religious practice? Unlike
secular biography, the sacred biography under which this statement on
lineage loosely falls shapes and defines its subject in such a way that s/he
comes to embody a paradigmatic vision that stands for the tradition
itself, accomplished by highlighting and conyextualizing events in the
life of the subject. In the case of lineage construction, the facts and
legends that surround each individual is also incorporated and organized
into a grander picture in such a way that each member is seen as a
timeless model of the founding vision or aspects thereof in real time.
That is to say, individual spiritual attainments is homologized with the
more universal pattern of the religious truth or path. Guanding adds still
another dimension to this creative process by juxtaposing two distinct
yet complementary lines of descent. This lends itself to a variety of
powerful mythic and historical archetypal pairings that when taken
together (and synthesized) broadly suggest two inter-related themes,
namely, the dual motif of discontinuity and revival, and the balance
between doctrinal learning ~ (jiao) and meditative praxis a (guan).
Fundamental to Guanding's construct is thus the creation of a vivid
image of a transhistorical bridge between the second century exegete,
Nagarjuna, the thirteenth (or fourteenth) teacher in the western line of
twenty-three (or twenty-four) teachers and the "high ancestor" of the
eastern line, and a humble Chinese meditator by the name of Huiwen,
whose northern origins remain in shadow. For despite that Huiwen is
said to have been Huisi's teacher and is thus only two generations
removed from Zhiyi, who must have heard about Huiwen from his own
teacher, Guanding is unable or unwilling to tell us virtually anything
about this man. Rather than detracting from the narrative, the mythic
Guanding has factored in one too many generations in designating Naglirjuna
great-great-grandfather. Zhanran comments: "According to the Erya mmt 'The
father of the grandfather is the great-grandfather, worthy of veneration. The father
of the great-grandfather is the great-great-grandfather, who is accorded the
highest respect. In this sense Guanding designates NagiiIjuna as the great-great-
grandfather and the term should be used in that way. During the Han, Qi, and
Sui, for example, the first ancestor Ml:tl. (shizu) was considered the high ancestor
~ : t l . (gaozu), a title of unparalleled virtue. OUf tradition also considers
Nagarjuna as our first [ancestor], therefore Zhiyi refers to him as our high
ancestor" (Fuxing, T 46, no.1912: 149b12-21).
PENKOWER 261
dimension of Huiwen's hidden charisma and bucolic obscurity -
especially when juxtaposed to the erudite and urbane profile of
NagaIjuna - only serves to add drama to his singular and inspired medi-
tative feat.
35
At the other end of this drama stands Zhiyi, on whom the focus of the
Chinese line ultimately comes to rest. On the one hand, he is temporally
represented as descending through Huisi and so forth through to the line
of Nagarjuna. Yet like Huiwen, Zhiyi's authority is not solely or ulti-
mately grounded in the received tradition but rather in his direct
personal experience. "[Later Zhiyi] performed the repentance [based
upon] the Lotus Satra, [as a result of which] he manifested dhara."
This milestone in the religious life of Zhiyi alludes to another piece of
mythic-historical lore that was simultaneously growing up around Zhiyi
and offers a completely different and transhistorical trope. Here Zhiyi's
authority is said to derive not from his teachers but from having been
present during the Buddha's recitation of the Lotus Satra on Vulture
Peak. This story, which later came to be known as the direct transmis-
sion 1itsitt11c (zhishou xiangcheng), draws its strength by positing a
physical link between Siikyamuni Buddha and Huisi and Zhiyi, ignoring
Huiwen, and gives pride of place to the Lotus Satra rather than to the
Dazhidu lun.
36
Its appearance in the literature can be traced to Guan-
35. The predilection for the quiet of mountain retreats resonates throughout the
biographies of early Tiantai adherents, as it does throughout the biographies of
Chinese Buddhist adherents in general. This attraction to the solitude of
mountains in fact predates Buddhism, giving this mountain-urban trope cultural
underpinnings. Like Huiwen, Huisi is portrayed as having avoided the major
population and cultural centers and having limited his associations in favor of a
contemplative life dedicated to purity and austerity. While Zhiyi took on more
civic responsibilities than did his predecessors, he too expressed longing for the
periods he spent in the mountains (including his stay on Mount Dasu and his ten
years on Mount Tiantai *.g- u.n, identified himself first and foremost as a
contemplative, and at the end of his life lamented that his own spiritual attainment
had been undermined by his socio-political undertakings. See Zhiyi's last known
correspondence, Guoqing bailu ~ m s ~ (Record of One Hundred [Items Con-
cerning the] Guoqing[si]), T 46, no. 1934: 809c-81Oc, translated in HURVITZ,
"Chih-i," p. 166.
36. An eventual shift in focus away from dependence upon an Indian exegetical text
(Dazhidu lun) and onto a scriptural work (Lotus Satra), which would become the
hallmark of Tiantai, is an example of the now familiar assimilative process said to
be shared by all of the so-called new schools of Chinese Buddhism that
developed during the Sui-Tang, wherein Chinese confidence in its ability to
interpret scripture resulted in a move away from reliance on commentarial
literature and exegetes in favor of an unfiltered look at scripture.
nABS 23.2 262
ding's Sui Tiantai Zhizhe dashi biezhuan, completed in 605, just prior to
the appearance of a lineage statement in the Mohe zhiguan. The story is
founded upon a statement made by Huisi upon first meeting Zhiyi on
Mount Dasu (in Guangzhou, modem Honan). That statement in tum is
based upon a passage in the Lotus SiUra that reveals Sakyamuni as the
transcendent Buddha who had become enlightened in the infinitely long
ago pasP? and who was constantly preaching the ultimate message of the
sutra in his reward body (baoshen; Sanskrit: sa1'[lbhoga-kiiya) in
the seven-jewelled pagoda on Vulture Peak or Mount Grdhraku!a.38
Upon seeing Zhiyi, Huisi reputedly said, "Long ago we listened
to the Lotus Satra on Vulture Peak. Your coming here now is the result
of this association in our past lives."39 Thereupon Huisi instructed Zhiyi
in the lotus samadhi. Fourteen days into it, upon reaching the chapter on
the "Deeds of the Bodhisattva Medicine King of the
Lotus Satra, Zhiyi had his initial enlightenment experience, whereby he
manifested dharm;ii. 40
Thus Guanding prefaces his lineage account with a powerful series of
quotations that claims for Zhiyi what the Buddha has claimed for
himself, namely, that his authority is sufficient unto itself and does not
rely on what he has learned from a teacher.
The Da[zhidu] lun says,"1 [Buddha] practiced without a teacher."41 A siitra says,
"I [Sakyamuni] received the prophecy of buddhahood from the Buddha
A [secular] text says, "Those who are born with knowledge are
37. Lotus Sutra, T9, no. 262: 42b-43a.
38. T 9, no. 262: 32b-34c.
39. Whether Huisi meant this in the literal or figurative sense is difficult to tell; the
implication at least was that their meeting was somehow fated by past circum-
stances. TAIRA Ryosho1jZTWil feels that this story was probably understood in a
literal sense as early as shortly after the death of Zhiyi; see his "Ryozan docha ni
tsuite ilwlOU!tt\ fc::.0v
y
t:," Tendai gakuho 14 (1971): 3-5.
40. Sui Tiantai Zhizhe dashi biezhuan, T50, no. 2050: 191c21-192al, and Zhiyi's
biography in the XGSZ, T 50, no. 2060: 564bI8-22. Zhiyi's XGSZ biography
also recounts a silnilar event in which, during a fourteen-day meditative expe-
rience, Zhiyi suddenly had a vision of himself with Huisi listening to Sakyamuni
preach the Lotus Sutra in the seven-jewelled Pure Land of Vulture Peak.
41. T 25, no. 1509: 65a1.
42. Paraphrased from Foshuotaizi ruiying benqi jing (Siitra
on the Auspicious Appearances and Origins of the Prince [Siddhmha]), T 3,
no. 185: 472c-473a.
PENKOWER 263
the highest. Next come those who attain knowledge through study."43 The
teachings of the Buddha ate a vast and sublime truth. Do they shine of
themselves with the truth of heaven [or do they need to be dyed] like the blue of
the indigo plant?,,44 .
These are indeed statements intended specifically to lend authority to
Zhiyi's formulation of the practice of calming and contemplation, the
subject matter of the text that Guanding is here prefacing. But they are
also more than that. Daniel STEVENSON chooses to translate the last line
in the above quote as, "Do [the Buddhist teachings] shine of themselves
with the heavenly light of truth or is their blue derived from the indigo
plant?" - where he understands the heavenly truth to mean transhistori-
cal insight and the indigo plant to symbolize text or teacher.
45
Bringing
this down a notch, these pair of opposite and rhetorical questions also
ask whether the teachings of the Buddha can be self-realized or need a
teacher-interpreter, such as Zhiyi, to bring them forth.
These ideas serve to highlight the basic tension between received and
inspired truth that runs throughout Guanding's recitation. This is a
profoundly religious pronouncement and Guanding's twofold genealogy
with its textually-oriented western line and its self-awakened line of
eastern contemplatives is, at its heart, a metaphor for the dynamic
interplay between received and inspired tradition that is subsumed under
the two-pronged agenda established by Zhiyi of doctrinal learning and
meditative praxis. The balancing of the two, for Zhiyi, is critical both to
one's own spiritual quest and to the well-being of the Buddhist
community at large. Failure to respect that balance results in just the
kind of discontinuity and need for realignment or revival that make up
the second or corollary message in this story.
As historians of Buddhism are fond of pointing out, Guanding's
twofold genealogy resonates with significant parallels to Chinese history.
43. Paraphrased from the Lunyu ~ f f i n m (Analects) 16:9. See D. C. LAU, tr.,
Confucius: The Analects (London and New York: Penguin Books, 1979), p. 140.
44. The imagery of the indigo plant is taken from the Xunzi 1!fj:y. where the knowl-
edge of a student is favorably compated to that of his teacher. The implication is
that something that is dyed blue from the indigo plant is more vibrant in color
than the original plant itself. See WATSON, tr., Basic Writings of Hsiin Tzu, p. 15.
Zhanran resists such a lofty compatison of Zhiyi to the Buddha, and emphasizes
instead the act of dying (Fuxing, T 46, no. 1912: 143c2-144a2). I thank an anony-
mous reader for this reference.
45. DONNER and STEVENSON, The Great Calming and Contemplation, p. 100.
JIABS 23.2 264
There is certainly historical meritto the notion that Zhiyi's vision of a
unified Buddhism brought together the sophisticated and theoretical
tendencies that flourished in the southern circles of Buddhism under the
aristocratic rule of educated Chinese and the pragmatic Buddhism of the
north that reflected the temperament and interests of foreign rulers
during the Northern and Southern Dynasties period (317-589). Extend-
ing the analogy further, Zhiyi's religious integration is then compared to
the political synthesis accomplished by his benefactor, the first emperor
of the Sui dynasty, who unified the country after a some three hundred-
fifty year geopolitical division. Yet as a statement primarily concerned
with religious identity, that is, as a statement that validates and
substantiates a religious vision and course of action, Guanding's twofold
genealogy is better understood first on a more personal and urgent level
- from the perspective of those who are invited to enter into, witness,
and re-enact the vision. This then is not primarily an abstract union of
north and south, west and east, but a blueprint for the Buddhist path as
laid out by Zhiyi. It is "the approach to the teachings that Zhiyi
practiced in his own mind."
Thus, although Guanding is by no means immune to practical and
political considerations, a call for a balance between doctrinal learning
and meditative praxis, first and foremost, is directed inward (i.e, to
Zhiyi's followers). Casting around for a juxtaposed pair in need of
synthesis, we find Zhiyi's venom directed at a pair of quasi-historical
personalities described as the meditation teacher of enigmatic under-
standing (anzheng chanshi) and the dharma teacher [bond to]
text (wenziJashi). Zhiyi compares the obscured dhyana prac-
titioner, who rejects all doctrinal understanding, to someone who burns
himself by grasping a torch (the mind in concentration) without
knowing how to handle it. The pedantic master of doctrine, whose dis-
cursive study is not tempered by contemplation, is likened to one who
also causes self-inflicted wounds by picking up a sharp knife (the mind)
without knowing how to use it.
46
These unnamed yet tangible transgres-
. sors, each in his own way, is guilty of excess or abuse of the received or
inspired tradition. For Zhiyi, this is not simply a personal exercise in
futility. To fail to comply with the Zhiyi benchmark of integrated
46. Fahua xuanyi, T 33, no. 1716: 686a, and Mohe zhiguan, T 46, no. 1911: 132a. cf.
Bernard FAURE: The Will to Orthodoxy: A Critical Genealogy of North em Chan
. Buddhism, translated by Phyllis BROOKS (Stanford: Stanford University Press
1997), p. 49.
PENKOWER 265
learning and practice, which in turn is a concrete expression of the
inclusive and harmonious design of the Buddha's view of the world, is
to literally create a fractured world, the consequences of which are a
contentious anctjeopardized self and society.47
This idea is dramatically driven home in the biography for Huisi who
lived through the devastation wrought on the Buddhist community by
Emperor Wu :litWf of the Northern Zhou (r. 560-578) during his purges
of the church beginning in 574.
48
Huisi wrote passionately and urgently
about corruption within the monasteries, and who was repeatedly
threatened with bodily harm by contemporary religious rivals. Indeed,
Huisi's change of focus from venerations and recitation (what Guanding
alludes to as the first seven years of this religious life) to meditative
practice is said to have been directly influenced by the reading of the
apocryphal Miaasheng ding jing (Scripture on the Wondrous
and Excellent Absorption), a text composed in China during the late-
fifth or early-sixth century. The Miaasheng ding jing makes a decidedly
Chinese argument in favor of meditation as the best and most efficacious
form of practice leading to salvation, situating it in historical terms as
the practice most appropriate for the age of the decline of the dharma or
mafa *$, itself an apocryphal term created in China with no identi-
fiable Indian equivalent.
49
The scripture prophesies that during this
looming age of decline, the monastic community will become fractured
and contentious; the clergy will spiral into moral decline through too
closely held associations with rulers and benefactors; and the church will
lose all credibility and respect in the eyes of the laity. This will lead to a
collapse of society at large, with both state and the family doomed to
war, famine, and indigence. In this age of decline of the dharma, the
scripture goes on to insist, salvation demands dedication to a life with-
47. See also STEVENSON's excellent discussion on the tension between received and
inspired truth and its implications for Zhiyi, The Great Calming and Contempla-
tion, pp. 33-39.
48. For a discussion of the Northern Zhou persecution of Buddhism see TSUKA-
MOTO Zenryli "Hokushii no haibutsu ni tsuite
two parts, ToM gakuM 16 (September, 1948): 29-101 and 18 (Febru-
ary, 1950): 78-111. CR'EN presents an English summary in Buddhism in China,
pp.184-194.
49. For the history of the Buddhist concept of the decline of the dharma see
NATTIER, Once Upon a Future Time; the development of the idea of moja in
China is discussed on pp. 98-118.
JIABS 23.2 266
drawn from society, of repentance and meditative cultivation. It is this
prescription that led Huisi to seek out Huiwen.
50
These then are the images against which Zhiyi' s religious .ancestry and
the individual descriptions of its various members are pitted. And the
antidote for such moral and spiritual decay is nothing other than the path
as described in the Mahe zhiguan. For this reason, Zhiyi proclaims, the
balance between doctrinal learning and meditative praxis - in Zhiyi's
descriptive terminology, the "two wings of a bird" or the "two wheels of
a cart" - is also a balance between self benefit and benefit to others.
Understanding [i.e., learning] purifies meditative practice; practice advances
understanding. illuminating, enriching, guiding, and penetrating, they mutually
adorn and embellish one another. They are like the two hands of a single body,
which together keep [the body] clean. It is not just a matter of advancing along
one's own personal path towards enlightenment by unlocking impediments and
overcoming obstacles. One must also be well-versed in the scriptures and
commentaries so that one can tum outward and teach and reveal [to others] what
they have not heard before. When one combines one's own training with the
training of others, benefit is complete. If one such as this is not the teacher of all
humankind and the jewel of the nation, who is?51
It is easy to see how such a conception of the dangers inherent in not
striking a balance between teaching and practice, received and inspired
truth, benefit to self and society, and so forth produced the coroIlary
mythic trope of discontinuity and revival also embedded in Guanding' s
rehearsal of Zhiyi' s religious pedigree. Indeed, at the heart of his outline
lies the claim that the dharma, which had been irrevocably lost to India
50. The Dunhuang manuscript of the Miaosheng ding jing is reprinted and discussed
in SEKIGUCm Shindai Tendai shikan no kenkyu 72ii'
(T9kyo: Iwanami shoten 1969), pp. 379-402. See also MAGNIN, La vie et
l' ceuvre de Huisi, p. 31. A relatively well-developed theory on mofa is found in
the Nanyue Si dachanshi li shiyuan wen (Text of Vows
made by the Great Meditation Teacher [Hui]si of Nanyue), T 46, no. 1933, a
work traditionally ascribed to Huisi, but now considered suspect. ETANI Ryiikai
!!:frIliM is a leading voice arguing against such an attribution, placing the com-
position or parts thereof of this work at a later date. See his "Nangaku Eshi no
Rissheiganmon wa gisaku ka :fz:'ifml::X:.f.i!J{'F:6>," Indogaku Bukkyo-
gaku kenkyu 6.2 (March, 1958): 213-216. MAGNIN coun-
ters, concluding that the bulk of the text was written by Huisi (pp.104-116).
51. Mohe zhiguan, T46, no. 1911: 46a; cf. DONNER and STEVENSON, The Great
Calming and Contemplation, p. 37. This passage is also the locus classicus for
the founder of Japanese Tendai (767-822) idea of Tendai monk as
'Jewel of the nation." See Paul GRONER, The Establishment of the Japanese
Tendai School (Seoul: Berkeley Buddhist Studies Series 1984), pp. 143-144.
PENKOWER 267
due to the decidedly negative interventionist policies of the state, had
been rediscovered in China through the individual efforts and inspired
insight into text of a lone practitioner and the first personality in the
Chinese line. Here again, rather than taking a defensive posture with
regard to the obvious lack of a direct link between the western and
eastern lines, Guanding trumpets the idea of discontinuity as the greatest
strength of his succession theory.52
This is all the more accentuated by Guanding presenting the eastern
line in ascending or reverse order (Zhiyi ---+ Huisi ---+ Huiwen ---+
[Nagarjuna]), thereby scripting the discontinuous motif as the climax of
his narrative. To a fledgling tradition that at the time enjoyed success at
court but which nonetheless was barely more than one generation
removed from the memories of the persecutions of Buddhism suffered
under the Northern Zhou, this was hardly an exercise in idle
theorization. Indeed, as later Tiantai with its own tumultuous history and
vicissitudes would bear witness, the mystical ability to weave together
the ideas of discontinuous inheritance and inspired insight into text first
proposed by Guanding in his introduction to the Mahe zhiguan would
continue to resonate in later generations, where it ultimately came to
define the true Tiantai luminary. Inspired insight into the mind of Zhiyi
(i.e., into the Mahe zhiguan) would be credited with resurrecting the
tradition on more than one occasion. Moreover, although Guanding
nowhere suggests this, his conjoining of the dual motif of discontinuity
and revival into the history of the genesis of the Mahe zhiguan, would
later be embellished upon so that the triumph that is the origins of
52. By contrast, the Chan tradition, which also adopts the lineage account found in
the Fu fazang yinyuan zhuan, would set out to prove that this line was not cut off
with SiIp.ha. Rather, it insists that prior to his death, Sirpha transmitted the
teachings to a disciple through whom the doctrine would eventually reach China
via the semilegendary Bodhidharma. Bodhidharma becomes known as the
twenty-eighth (or, for those Chan lines that accept Madhyantika, twenty-ninth)
western patriarch and the first Chinese patriarch of Chan. See Y AMPOLSKY, The
Platform Sutra of the Sixth Patriarch, p. 6 note 8 and p. 8, and Wendi Leigh
ADAMEK, Issues in Chinese Buddhist Transmission as seen through the Lidai
Fabao Ii (Record of the Dharma-jewel through the Ages)," Ph.D. dissertation
(Stanford University 1997), pp. 145-157. This scenario, so antithetical to the one
begun by Guanding, which highlights the idea of discontinuity and, by extension,
reliance on inspired insight into text, would eventually result in a Chan self-
definition characterized as "a special mind-to-mind transmission outside of
scripture" ~ : 9 i - J3U l' (jiaowaibiechuan)
nABS 23.2 268
Tiantai came to be equated to the resurrection of the truth of Buddhism
not only in its adopted home (China) but also in its birth place (India).53
Contextualizing the Appearance of the Statement on Lineage
Two inter-related questions immediately present themselves. First, what
prompted Guanding to introduce a statement on lineage at the head of
his introduction to the Mohe zhiguan at this particular juncture in
history? This brief yet substantial rehearsal after all is the first known
attempt of any kind to define a Chinese teacher (or text) in terms of a
succession theory. .
In attempting to answer this question and the one raised in the next
section (see p.278), we would do well to begin by looking at the
construction process of the Mohe zhiguan, and especially the Guanding
introduction, which is our concern here, for it allows us to situate the
appearance of the statement on lineage within a historical context. As
previously mentioned, the text of the Mohe zhiguan proper is the edited
result of a series of lectures by Zhiyi recorded at the Yuquansi over the
course of the summer retreat of 594, just three years prior to the death
of Zhiyi. Beginning shortly before Zhiyi's death in 597, and on two
separate occasions thereafter, Guanding edited these lecture notes and
reworked his introduction, the result being three editions of the text, of
which only the last survives. Guanding's introduction that stands at the
head of the current edition begins with (a) a brief notation on the
53. The estoeric monk Hanguang 13% (n.d.) is reputed to have said to Zhanran:
"When I was traveling with the master Bukong (Amoghavajra;
705-774) in India, we met an Indian monk who asked me: 'I have heard that the
teachings of Tiantai are circulating in the Great Tang. They are best for distin-
guishing the heterodox from the orthodox, the one-sided from the perfect. Can
you have these texts translated [into Sanskrit] and brought to this country?'"
Zhanran responded: "Isn't this like saying, 'The teachings have been lost to
China and must now be sought in the surrounding regions?'" Zhanran's Wenjuji
X{J]'fc\ (full title: Fahua wenji ji or Notes on The Words and
Phrases of the Lotus Satra), T34, no. 1719: 359c14-18. cf. FZTJ, T49, no.
2035:189a9-13, translated in my "T'ien-t'ai during the T'ang," p. 99. ZhanraIl's
final lament is an adaptation of a passage in the Zuozhuan in which Confucius
laments, "The Son of Heaven (Zhou king) has lost the rites and must seek after
them from in the four directions." Thus, just as the rites, which originated in but
had been lost to the Zhou, had to be sought from afar, the teachings of the
Buddha, which had originated in but had been lost to India, were now being
sought in the land where Buddhism had spread. (Tiantai would also of course
spread this truth east to Japan, where the Tendai school would be established.)
PENKOWER 269
location and date of Zhiyi's lectures followed by the statement on
lineage, (b) a summation of the three kinds of calming and contempla-
tion, and (c) a section on the scriptural verification of these practices.
Now, the function of prefatory remarks is of course to legitimate that
which follows in the main body of the text. In the case before us, the
task that fell to Guanding was the defense of the perfect and sudden
practice of calming and contemplation, about which Guanding boldly
tells us in the now famous line that opens his introduction: "Calming and
contemplation as luminosity and tranquility l1::afYjl!- (zhiguan ming-
jing): [this teaching] has not yet been heard of in former generations
when Zhiyi ... expounded upon it."54 Chinese Buddhists, following in
the tradition of both Indian Buddhist and Chinese secular writers, had
long approached the task of bringing authoritative weight to bear on
seemingly innovative ideas, as is the claim here, by means of a twofold
justification scheme, namely, a recitation of the innate soteriological
value of the doctrinal ideas themselves, supported by select and frequent
references citing scriptural precedence for these ideas. Guanding's intro-
duction follows this formula. In the current edition of the text, these two
elements are represented by a summation of the three kinds of calming
and contemplation and a section on the scriptural verification of these
practices, that is, (b) and (c) above.
55
Guanding does not end there.
Rather, what makes his preface distinctive is the introduction of a histor-
ical justification in the form of a lineage statement. Not only is this the
fIrst recorded case of such a strategy to come down to us, but Guanding
ultimately gives pride of place to that verification by positioning the
lineage statement at the head of the introduction before either the
theoretical or scriptural sanctions for the ideas set forth in the text. This
unprecedented formulation is a momentous event in the history of
Chinese Buddhism. For the fIrst time, a Chinese Buddhist (Zhiyi) and
54. Mohe zhiguan, T 46, no. 1911: 1a7, as translated by DONNER and STEVENSON,
The Great Calming and Contemplation, p. 99. This famous opening phrase of
the Mohe zhiguan can be interpreted in several ways. SWANSON translates; "The
luminous quiescence of cessation-and-contemplation was unknown in former
ages. The wise one [Chih-i] elucidated it..." See The Great Cessation-and
Contemplation, p. 2 and note 3.
55. Four sutra are used to prove all three kinds of practice, one sutra for the gradual
alone, and six sutra for the sudden alone. The Huayan jing . ~ ~ (full title:
Dafang guang fo huayan jing j,:JJ .19Il. j j ( ~ ; AvataTflsaka or Flower Orna-
ment Sutra; T9, no. 278, and T 10, no. 279) ties the section together. See
DONNER and STEVENSON, The Great Calming and Contemplation, pp. 114-127.
nABS 23.2 270
his work (theMohe zhiguan) are grounded in a particular line, traceable
in time through past Chinese teachers, who are then connected to the
Buddha through a particular Indian line (Nagarjuna). By ,insisting that
the text represents nothing less than "the approach to the teachings that
Zhiyi followed in his own mind," it moreover affmns forward projec-
tion or implied continuation, whereby future generations, either by
means of direct instruction or the text of the Mohe zhiguan alone, can
come to know the practice of calming and contemplation and thus the
means to salvation. The genesis of the text, for Guanding, thus is
authenticated by ancestral pedigree and the ancestral line validated by
the text. 56 .
When, how, and why did this fledgling sense of community come
about? Thanks to SATO Tetsuei's meticulous study of the
different redactions of the Mohe zhiguan, we can now speak with some
confidence about the dating and structure of the Guanding preface as it
appeared in each of the three editions to the text.
57
According to SATO,
the first edition was probably completed by 597, just prior to Zhiyi's
death, and therefore known to Zhiyi. What is significant about this first
56. This innovative strategy of adding (and highlighting) a historical justification in
the form of a lineage statement to the more traditional theoretical and scriptural
justifications of text is so germane to the history of ideas that it is pointed to as a
defining reason for calling Tiantai the first of the so-called new schools of
Chinese Buddhism to appear during the Sui-Tang. See YOKI Reimon
"Zui-To jidai ni okeru Chugoku-teki Bukkyo seiritsu no jijo ni tsuite no kosatsu
Nippon Bukkyo
gakkai nenpo B 19 (1953): 79-96.
57. With regard to the text proper, the most important distinction between the first
nyp editions and the third is the title change. The first two editions were given the
title Yuandun zhiguan (The Perfect and Sudden Calming and Contem-
plation), which was changed to the name by which we know this text today with
the last edition. The first two editions were nonetheless distinguishable in that the
first had twice the number of fascicles (twenty) as the second (ten). The extant
text is also divided into ten fascicles, although some editorial changes seem to
have been implemented between the second and final editions. All three editions
were extant at least through the mid-eighth century, as is known from the Fuxing,
Zhanran's (711-782) commentary on the Mohe zhiguan. The second edition was
probably among the works brought to Japan by Hanzhen (Japanese: Ganjin;
688-763) in 754. One of the two non-extant editions seems to have survived into
the twelfth century in Japan, as is evidenced by an old-edition citation in
Shoshin's Shikan bugyo shiki This, and the following
information, is based on SATO Tetsuei, Tendai daishi no kenkyii, pp. 370-382
and 396-400.
PENKOWER 271
edition for our purposes is that it contained no statement on lineage.
Rather, of the three sections that comprise the current Guanding intro-
duction, the first edition supplied only a summation of the three kinds of
calming and contemplation omitting both statements on lineage and on
scripture. Those two sections not found in the first edition were intro-
duced into the Mohe zhiguan in Guanding's second edition. The order in
which these sections were laid out in that edition, however, differed
from that of the final edition. In the second edition, a synopsized state-
ment on scriptural verification came first and was followed by a state-
ment on lineage. 58 These new sections were not yet delineated from the
introductory remarks of Zhiyi, which were first separated out and placed
behind the Guanding introduction in the extant edition. The summation
of calming and contemplation was positioned last in the second edition.
There is no way to tell whether or to what extent the lineage account in
the second edition was re-worked prior to its inclusion in the final
edition. Dating the second edition also remains problematic. Zhanran's
Fuxing, the authoritative mid-eighth century Chinese commentary on the
Mohe zhiguan, which had all three editions of the text at its disposal,
points to the existence of the second edition just prior to 605. ShOshin's
Shikan bugyo shiki, an important twelfth century Japanese sub-commen-
tary on the same, which knew of (at least) one of the two no longer
extant editions, moves the appearance of the second edition to just the
other side of that year. The final edition redistributed the sections of the
introduction and separated them out under Guanding's name. While the
date of completion of the Mohe zhiguan is also elusive, SATO leans
toward one close to Guanding's death in 632, but settles on a time frame
between 607 and 632.
Summarizing the construction process of Guanding's preface then,
some kind of statement on lineage made its appearance around 605,
when it was sandwiched between a brief statement on scriptural verifi-
cation and a statement on the three forms of practice. The lineage state-
ment that survives today in its position as first among three strategies to
legitimate the text of the Mohe zhiguan may have appeared as early as
607 but was probably completed not long before Guanding's death in
632. Guanding's genealogical rehearsal thus was not in circulation, at
least in written form, during the lifetime of Zhiyi. Rather, the above
58. This was either deleted in the final edition or incorporated into the longer
statement on same that now comprises the last section of the current text.
JIABS 23.2 272
chronology susgests that, despite that the continued religious and institu-
tional well-being of the communities established by Zhiyi remained
foremost in his thoughts even as death drew near, at the. time of his
demise, there was scant apparent interest in constructing a written
narrative for Zhiyi in the form of a genealogical account through
successive generations. Rather, this idea developed gradually during the
initial years of the seventh century, and was finally given a place of
prominence at the head of Guanding's introduction to the Mohe zhiguan
sometime between the last decade of the Sui and the fIrst decade or so of
the Tang. .
This of course does not preclude the possibility that a conception of
lineage or several conceptions of lineage existed in some embryonic
form prior to the death of Zhiyi. At least by the last quarter of the sixth
century, changes had taken place in the way in which temple property
was organized and inherited within the Buddhist community in general,
by Zhiyi with his strong support by both the Chen and Sui courts in
particular, that may have stimulated interest in creating lineage accounts
to protect that property. At the very least it must have had a homoge-
nizing effect on the notion of discipleship. Around the time of Zhiyi and
Guanding, in contradistinction to the earlier practice of inviting monks
to head monasteries irrespective of their theoretical orientation, the late
Northern and Southern dynasties practice of donating estates or tax bases
to particular monasteries and/ or specific teachers of particular doctrines
or texts, which were then inherited by the heir or chief disciple of the
master to whom the institution had been donated, became particularly
popular with the Sui emperors and nobility.59 Zhiyi and his early com-
munity were certainly recipients of such largess. The three major temple
complexes associated with early Tiantai and created or conceived by
Zhiyi were all established and/or supported under this system.
Some have suggested that interest in constructing a line of descent may
have started as early as with the establishment of the Xiuchansi { ~ W ~
(Monastery for the Practice of Meditation), founded by Zhiyi on Mount
Tiantai in southeast Zhejiang province in 575.
60
This monastery was
granted imperial support two years later when Emperor Xuan '"W of
59. John JORGENSEN: "The 'Imperial' Lineage of Ch'an Buddhism: The Role of
Confucian Ritual and Ancestor Worship in Ch'an's Search for Legitimation in
the Mid-Tang Dynasty," Papers on Far Eastern History 35 (1987): 99.
60. JORGENSEN, Ibid.: 99.
PENKOWER 273
the Chen (r. 569-582) ordered that part of the taxes from nearby Shifeng
county be used to assist the fledgling community, and that two families
from the county be released from other civic responsibilities to supply
water and firewood to it. The name of the monastery was bestowed by
the emperor the following year (578).61 Some fifteen years later,
towards the end of his career, Zhiyi again established the Yiyinsi
(One Sound Monastery) at the southeast foot of Mount Yuquan .:li5JHi!
in Hubei in 592. This monastery, which subsequently served as the plat-
form for the series of lectures by Zhiyi that became the Fahua xuanyi
and the Mahe zhiguan (594), was granted an imperial plaque and had its
name officially changed to the Yuquansi (Jade Spring Monastery) by
Emperor Wen )Cfl of the Sui (r. 581-604) in 593.
62
One need look no further than to the familial language borrowed from
secular genealogical accounts and employed by Zhiyi in his final testa-
ment to his disciples to appreciate the sense of distinct communities that
were evolving around particular teachers or father figures in conjunction
with developments such as these.
63
Yet the evidence suggests that the
idea of constructing lineage statements through successive generations
began to germinate slowly only after the demise of Zhiyi. Indeed, at the
time of Zhiyi's death, the management of the three major temple
complexes associated with him - the above-mentioned Xiuchansi and
Yuquansi and the Guoqingsi (Monasteryfor the Purification of
the Nation) discussed immediately below - simply passed to three
61. Guoqing bailu, T 46, no. 1934: 799a24-b1. The Xiuchansi took second place to
the Guoqingsi (Monastery for the Purification of the Nation), established
at the foot of the mountain in 601.
62. Guoqing bailu, T46, no. 1934, S06c12-1S. During the Tang, the Yuquansi
gradually became more syncretic in nature, becoming home to monks of the
Northern Chan, Vinaya, and esoteric traditions in addition to Tiantai. The Tiantai
community at the Yuquansi became increasingly independent of its counterpart
on Mount Tiantai, and remained a vibrant center for Tiantai study and practice
until the Buddhist persecutions of the mid-ninth century.
63. In a deathbed scenario reminiscent of that of the Buddha, when asked to whom
the community should look after he is gone, Zhiyi insists: "The rules of discipline
is your teacher. I have always told you, take the four kinds of
samadhi as your clear guide ... Only these great teachers can be the bases of your
support. Through the teaching we have met; through the teaching we are kin ;m
(qin). If you transmit the lamp of the Buddha, then you are my family :jjfg
(juanshu). If you cannot, then you are no follower 1JE (tu) of mine." Sui Tiantai
Zhizhe dashi biezhuan, T 50, no. 2050: 196b15-22. Here as elsewhere, Zhiyi
shows little interest in separating out his particular brand of Buddhism.
JIABS 23.2 274
different Zhiyi disciples.
64
Rather, the creation story enterprise was
closely associated with efforts to sculpt a defining biography for the
deceased teacher. This in turn was integrally connected, to Zhiyi's
identification with Mount Tiantai and the construction and institutional
and political recognition of the Guoqingsi. This temple complex, located
at the foot of the mountain, was conceived by Zhiyi but was .not
completed until 601, four years after his death. It was granted an official
plaque and so named in 605 by Yang Guang Zhiyi's most power-
ful benefactor, who by then had ascended the throne as the second Sui
emperor Yang mwr (r. 604-617).65 The last and most prestigious C?f the
three major monastic compounds associated with Zhiyi, from its ranks
came the most concerted of the early bids to create and control a
particular post-Zhiyi vision. While the management of the Guoqingsi
was placed in the hands of a more senior Zhiyi disciple named Zhiyue
(543-616), the task of orchestrating a biographical history of Zhiyi fell
firmly to Guanding.
Thus to return to the "why now" question posed at the start of this
section, it is no coincidence that the appearance of a lineage statement in
the second edition of Guanding's preface to the Mahe zhiguan roughly
coincides with the date of the dedication of the Guoqingsi (605). The
construction of the lineage was one of three literary projects, overseen
by Guanding and begun on the heals of the completion of the monastery
in 601, designed - through artful representation - to retroactively
enhance or instill religious meaning into events that highlighted the life
and surrounded the death of Zhiyi. In this regard, the lineage statement
in Guanding's introduction to the Mahe zhiguan, which represents the
culmination and legitimation of Zhiyi's religious career and which
comes to rest on the idea of discontinuity and trans historical (read:
miraculous) insight, cannot be fully appreciated or understood without
64. The management of the Xiuchansi went to Zhixi (n.d.; biography in the
XGSZ, T 50, no. 2060: 582a-583a), that of the Yuquansi to Daoyue (n.d.;
XGSZ, T 50, no. 2060: 661c-662b), and that of the Guoqingsi to Zhiyue
(543-616; XGSZ, T 50, no. 2060: 570c-571a).
65. Guoqing bailu, T 46, no. 1934: 812b21-c8, and 816a19-b19. In 839, during his
travels in China, the Japanese pilgrim Ennin IEC (793-864) reported one
hundred-fifty monks in permanent residence there and more than three hundred
monks as participants in its summer retreat. See Edwin O. REISCHAUER, tr.,
Ennin's Diary: The Record of a Pilgrimage to China in Search of the Law (New
York: The Ronald Press Co. 1955), p. 79.
PENKOWER 275
also placing it within the context of the equally miraculous process by
which the Guoqingsi came to epitomize the raison d'etre of Zhiyi's
institutional and political careers.
In the years following Zhiyi's death leading up to 605, in addition to
his duties as an emissary between the community on Mount Tiantai
and the Sui court and as editor of the Mohe zhiguan, Guanding was
primarily occupied with the writing and compilation of two works
documenting the life of Zhiyi. The first was Zhiyi's official biography,
which became known as the Sui Tiantai Zhizhe dashi biezhuan
(A Separate Biography of Zhizhe [Zhiyi], the Great
Teacher of Tiantai during the Sui [Dynasty]), begun in 601 and com-
pleted in 605.
66
The second was the Guoqing bailu (Record
of One Hundred [Items Concerning the] Guoqing[si]), a select collection
of correspondences and documents relating to Zhiyi, the early religious
community, and other religious and secular leaders. This work was
begun around 601 under the direction of an otherwise unknown monk
by the name of Zhiji (n.d.), but was completed in or shortly after
607 by Guanding, who took over the project upon the death of its initial
compiler.
67
Unlike secular biographies but in line with all good sacred
biographies, the primary concern of these two works was neither breadth
nor impartiality. Rather, their meaning was derived through the melding
of paradigmatic and narrative thought. While Zhiyi's religious integrity
and soteriological fortitude are well-represented in both of these
documents, what is of particular interest to us here is the process by
which events surrounding the life and death of Zhiyi were ultimately
connected in an integrated fashion to Mount Tiantai and to the sub-
sequent plans for the construction and promotion of the Guoqingsi.68
The period immediately following the passing of Zhiyi was one of
66. T 50, no. 2050. On dating see Koichi SHINOHARA, "Guanding's Biography of
Zhiyi, the Fourth Patriarch of the Tiantai Tradition," in Speaking of Monks:
Religious Biography in India and China, edited by Phyllis GRANOFF and Koichi
SHINOHARA (Oakville, New York, London: Mosaic Press 1992), pp. 107-108
and 115-116, note 22.
67. T 46, no. 1934. On dating see IKEDA Rosan Kokusei hyakuroku no
kenkyu (Tokyo: Daizo shuppansha, 1982), pp. 13-15.
68. The following follows the excellent and detailed study on this phenomenon
whereby Zhiyi's biography is constructed to identify him with Mount Tiantai and
the Guoqingsi by Koichi SHINOHARA, "Guanding's Biography of Zhiyi," pp. 97-
232.
JIABS 23.2 276
great uncertainty his followers. In anticipation of that transition, just
three days prior to his death, Zhiyi wrote to Yang Guang, his most
powerful benefactor, informing him of his intentions to h!lve a major
monastic complex built at the foot of Mount Tiantai. According to the
preserved correspondence, at that time, Zhiyi requested that the prince
ultimately have an official plaque made for it, transfer ten monks from
the Yuquansi to administer it, and assign fields to the new temple to
ensure its economic surviva1.
69
The following year (598) Yang Guang
vowed to take over the actual building of the complex and ensure its
support.10 With that, Zhiyi's rather straightforward request in life was
gradually transformed into a series of ever more potent prognostications
(combined. with posthumous miraculous sightings of Zhiyi), which
ultimately seamlessly melded the biography of Zhiyi, the story of the
Guoqingsi, and the fortunes of Yang GuanglEmperor Yang and the Sui.
In the process, Zhiyi's biography becomes one in which his true
religious life is played out on Mount Tiantai, demarcated by his first
visit and enlightenment experience on Huading Peak in 575 and his
second visit and death on the mountain in 597, an event depicted as
having been somehow fated.
Briefly chronicling the miraculous predictions posthumously credited
to Zhiyi, we find: (a) on the heals of Yang Guang's promise of support
for the Guoqingsi a story appears that Zhiyi had predicted that a power-
ful person would eventually step forward to take on the building of the
monastery,7l and (b) by the time Yang Guang became Emperor Yang
and began deliberating on the choice of name for the new monastery in
605 that story had evolved into a dream in which Zhiyi prophesied that,
only after the unification of the nation, a great benefactor would build a
headquarters for Tiantai which, in tum, would serve to purify the nation
The name ofthe Guoqingsi was taken from this power-
ful prophesyJ2 To these miracle tales were eventually retroactively
added prophesies and dreams beginning even prior to Zhiyi's leaving
home and suggesting a karmic connection between Zhiyi and Mount
Tiantai. Zhiyi's biography thus reports several early visions of the yet-
69. Guoqing bailu, T46, no. 1934: 809b-81Oe.
70. Guoqing bailu, T 46, no. 1934: 810e-812apassim.
71. Guoqing bailu, T46, no 1934: 812a17-19, and Sui Tiantai Zhizhe dashi
biezhuan, T 50, no. 2050: 195e28-196a18.
72. Guoqing bailu, T 46, no. 1934: 816a4-13.
PENKOWER 277
unvisited environs of the mountain, of Zhiyi's residence there with
imperial support, and finally of his pre-ordained death on the
mountain.?3 This biography would be copied and circulated throughout
the empire and an official inscription based on it erected on the
mountain by imperial decree in 605.7
4
In this way, by the very early
years of the seventh century, the biography of Zhiyi, in large part,
becomes the story of the establishment of the Guoqingsi, and the
Guoqingsi, at least from the perspective of the Zhiyi's followers
associated with it and the Sui court, synonymous with Tiantai.
This same technique is at work in Guanding's introduction to the
Mohe zhiguan where the purpose is to legitimate a particular religious
vision and a particular text. By identifying the Mohe zhiguan with a
particular historical narrative, the text comes to stand for the tradition
and the tradition for the text. Moreover, in much the same way that
Guanding fortifies the institutional and political aspects of Zhiyi's
biography by interweaving transhistorical and real time events, that is,
miracle stories and court connections, in the Mohe zhiguan Guanding
sanctions Zhiyi's soteriological revelations and practices by melding
together a transhistorical trope based on inspiration into text and a
tangible line of descent to which he could lay claim. As much as these
gestures can be said to be motivated by a desire to honor the late Zhiyi,
they are nonetheless also a fledgling yet orchestrated attempt by the
Guoqingsi-based Tiantai community, and Guanding's group in particu-
lar, to control the religious and political discourse and secure continued
support for its newly created monastic center. The introduction of
genealogical account in the Mohe zhiguan is significant as the first
instance of use of such a device to justify a particular soteriological
course of study and praxis. Yet it cannot be separated from the equally
important effort by the same group to try to ensure that the Guoqingsi
came to institutionally represent Zhiyi and the tradition.
73. See, for example, Sui Tiantai Zhizhe dashi biezhuan, T50: 191b14-28, where
Zhiyi has an early vision, including his death on the mountain, and 192c23-
193b19, where the unification of the Sui and the name of the Guoqingsi are
already predicted at the time Zhiyi entered Mount Tiantai for the first time in 575.
As part of the description of the events leading up to the death of Zhiyi,
Guanding also recounts a story of how a deity, known to Zhiyi from an earlier
encounter, visited him on the mountain and informed him of his impending death;
T 50: 195c5-27.
74. Guoqing bailu, T46: 817al-3. The text of the inscription, composed by Liu
Guyan 1PP,IDi-g (d.u.), is preserved on 817a-819b.
JIABS 23.2 278
The Niigiirjuna Connection
The question remains of why Guanding singles out Nagarjuna among the
exegetical luminaries of the eastern line as the bridge between the
teachings of the Buddha in India and the origins of the Mohe zhiguan.
For while it is true that in general Zhiyi relies heavily on the language
of the Dazhidu lun and Zhonglun - along with the apocryphal Renwang
jing and Yingluo jing - for his formulation of the Tiantai
three truths and three discernments or contemplations .::::.fi (sanguan)
theory,75 Zhiyi's integrated systemization of the totality of the received
tradition in fact is distinguished for its catholicity and comprehensive-
ness. For Zhiyi the perfect teaching (yuanjiao) that informs the
Mohe zhiguan is thus not the property of anyone scripture or tradition
but can be found in a great variety of texts, including the many that are
featured therein.
Indeed, in circulation were a diversity of contemporary prototypes
about how texts and teachings got their start that offered very different
tropes than the statement on lineage introducing the Mohe zhiguan,
which insists upon an intimate relationship between Huiwen and a
Nagarjuna text. For example, the previously introduced direct transmis-
sion omits Huiwen altogether and forges a transhistorical link between
the Buddha and Huisi and Zhiyi, giving pride of place to the Lotus
(scripture) rather than to the Dazhidu lun (treatise). As previously noted,
this trope is nonetheless interwoven in embryo form in the M 0 h e
zhiguan rendition of the lineage, thereby acknowledging the centrality of
that scripture to Zhiyi.76 A second example specifically related to the
genesis of meditative techniques is a slightly later line that came to be
known as the transmission of the nine teachers fLSffii'* (jiushi xiang-
cheng). This genealogical account made its appearance in a now non-
75. For a discussion on the construction of these texts and their impact on Zhiyi see
SWANSON, Foundations ofT'ien-t'ai Philosophy, pp. 38-56. For the Dazhidu
lun in the Mohe zhiguan, see T 46, no. 1911: 4a18-11a14, translated in DONNER
and S1EVENSON, The Great Calming and Contemplation, pp. 140-218.
76. The direct transmission forms the basis for Saich6's Tendai Ry6zen mill lineage;
Kechimyakufu lfrl)l))HI [full title NaishO buppo soja kechimyakufu
or Diagrammatic Description of the Secretly Certified Blood-
lineages of the Buddha-dharma], DZ 1: 215-230. By the ninth century in China,
this legend was popular among Lotus devotees centered around Chang'an
:f.it3!i:. See Fahua zhuanji T51, no. 2068 56c-57a, and Hongzanfahua
zhuan T 51, no. 2067: 22c.
PENKOWER 279
extant Guoqing guang baiZu (an apparent supplement to the
Guoqing bailu). This narrative responds to the failure in the M ohe
zhiguan to provide a genealogical history of Huiwen's Chinese religious
training, yet no effort to connect Tiantai to its Indian roots or its
meditative techniques to doctrine. Instead it makes a rather allusive and
unsuccessful bid at reconstructing six of Huiwen's early Chinese medita-
tion teachers (thus in the end adding nothing to the prestige of the
shadowy Huiwen) and ends with Huiwen, Huisi, and Zhiyi for a total of
nine Chinese personalities'??
Early conceptions of lineage, including the one found in the Mohe
zhiguan, were indeed posthumous and somewhat arbitrary constructs,
each designed to depict the origins of text or teachings meditative praxis
77. My only known referencce to this line is found in Zhanran (Souyao ji, XZJ 99:
227all-12). The dating of this line is discussed in my forthcoming Tientai
Buddhism and the construction of Lineage during the Tang, chapter three. The
first six meditation teachers are presented by Zhanran as follows (Fuxing, T 46:
149a24-b7):
(1) Ming lj)j prescribed to the seven [preliminary] expedients [of meditation]
(qifangbian).
(2) Zui:ll'z employed the method of interfusing the mind 1M!{,\ (rongxin).
(3) Song used [seeing into] original mind :z!s:{,\ (benxin).
(4) Jiu:\'; adhered to the method of silencing the mind ;JlH,\ (jixin).
(5) Jian used the method of enlightening the mind 7 {,\ (liaoxin).
(6) Hui used the method of treading on or planting [one's feet in] mind
(taxin).
Huisi's biography in the XGSZ lists Jian and Zui (among unnamed others) as
teachers Huisi met during his search for a dhyana master prior to his meeting
Huiwen (T 50: 563a13-14); in discussing the same event in Zhiyi's biography,
the same text names Jiu and Zui (T 50: 564b14). Of the first six, only the last has
been identified by Saicha (Kechimyakufu, DZ 1: 224) as Fu Xi (497-569)
who is said to have advocated a meditation technique called guarding the one
without wagering (shouyi buyi). For a description of Fu Xi's medita-
tion technique according to Daoxin see David W. CHAPPELL, ''The Teachings of
the Fourth Ch'an Patriarch Tao-hsin (580-651)," in Early Ch'an in China and
Tibet, edited by Whalen LAI and Lewis R. LANCASTER, Berkeley Buddhist
Studies Series, vol. 5, (Berkeley Asian Humanities Press 1983), pp. 114-117. As
part of his criticism and rejection of this line as a whole for its reliance on
meditation technique alone and its failure to provide a "string of pearls" line of
descent, Zhanran (Fuxing, T 46: 149b7-8) insists that all six personalities prior to
Huiwen were "contemporaries of Huisi whom the latter met during his search for
dhyana masters before he joined Huiwen and received the Dazhidu lun," thus
concluding that Tiantai "was not transmitted [to Huiwen] by the first [six
teachers]. "
JIABS 23.2 280
or religious experience, institution or other material concerns. Some-
times working in tandem, sometimes not, these narratives reflected a
variety of regional and cultic visions of Zhiyi. The development of the
various uses of those narratives was part of a process which involved
trial and error and the pulling together, taking apart, and rearranging of
the various bits of lore that had been growing up among Zhiyi's
followers. As we have seen, the adoption of one construct did not and
does not necessarily preclude the equal acceptance of another in another
context. The experimental nature of the Huiwen/Nagarjuna connection
moreover is perhaps best exposed when we consider still a n o t h e ~ early
prototype, also advanced by Guanding, that attempts to illustrate the
delicate balance between doctrinal learning and meditative praxis and the
idea of discontinuity and renewal, both so important to his soteriological
vision as expressed in his introduction to the Mohe zhiguan. This brief
alternative account of the origins of Tiantai is introduced into the
literature as recently as in the Sui Tiantai Zhizhe dashi biezhuan, the
Zhiyi biography completed by Guanding in 605 just prior to the
appearance of the second edition of the Mohe zhiguan in which the now
famous lineage statement of Nagarjuna -+ Huiwen -+ Huisi -+ Zhiyi was
introduced in the Guanding preface. Yet the rendition of Zhiyi's
ancestors found in the biography features not Nagarjuna and Huiwen but
rather the fifth century west-east pair of Buddhabhadra (359-429) and
Xuan'gao ~ ~ (402-444) as the architects of a balanced Buddhism.7
8
78. Buddhabhadra was a disciple of the famous Kashmiri Sarvastivadin dhyana
master Buddhasena. Arriving in Chang' an around 408 at a time when
Kumarajlva held great sway in the capital, Buddhabhadra was known for his
attention to meditative practice and the rules of discipline. Although, like Kumara-
jlva, Buddhabhadra would become known as the translator of a great variety of
texts, he was also known to have challenged both Kumarajlva's brand of
Buddhism and the failure of the latter's community to adhere to strict discipline.
Following the death of their mentor, Kumarajlva's followers managed to bring
trumped up charges stemming from disputes over discipline and doctrine against
Buddhabhadra to the attention of the court and he was banished from the capital
in 410. Even the intervention of the powerful Huiyuan, who befriended Buddha-
bhadra on Mount Lu between 410 and 412, could not persuade the northern court
of Yao Xing to reconsider its decision against Buddhabhadra and admit him back
into the capital. Buddhabhadra left Huiyuan's community on Mount Lu in 412
for the southern capital of Jiankang, where he remained until his death in 429.
This, of course, throws into suspicion whether or to what extent the Chang' an
native Xuan' gao ever studied under the famous Sarvastivadin monk as the former
was only eight years old at the time Buddhabhadra was exiled from the capital.
PENKOWER 281
This balance was later corrupted, and remedied and saved from
extinction by Huisi rather than Huiwen, who is omitted entirely (and
who thus seems to have been ultimately drafted to serve as a convenient
lirik to the northern meditative tradition), and Zhiyi.
Long ago Buddha[bhadra] and Xuan'gao developed meditation (ding) and
wisdom (hui) in tandem. Later, their [teachings] deteriorated and became like a
one-wheeled [cart] and a single-winged [bird. So the situation remained] until it
was righted and revived by [Huisi of] Nanyue, and reached its prosperity here
[with Zhiyi].79
Although we have no way of knowing with certainty why Guanding
ultimately rejects the above pairing in favor of Nagarjuna/Huiwen, that
this prototype exists at all suggests that we need to look beyond (or in
addition to) the purely soteriological in understanding Guanding's state-
ment on lineage advanced in the Mohe zhiguan. In the last section, we
looked at the Guoqingsi agenda that informed the decision to create a
lineage statement. Here, we extend that discussion outward to include
the relationship between Mount Tiantai and the broader religious world
in which it operated, as well as personalize it by looking, as best we can,
into what this lineage account accomplished for Guanding himself.
Issues of self-identity and self-definition were of course not develop-
ing on Mount Tiantai in a vacuum. One equally influential group inter-
ested in charting its own genealogical history was the Sanlun = ~ tradi-
tion, a group, like Tiantai, that during the Chen was headquartered in
the southern mountains and flourished in and around the southern
capital. Unlike the Guoqingsi community on Mount Tiantai, however,
this groupe ultimately managed a smoother transition out of its base
on Mount She jlilJ (in modem-day Jiangsu) to Chang'an once the polit-
ical power base shifted north with the consolidation of the Sui around
Be that as it may, like Buddhabhadra, Xuan' gao not only came to be remembered
for his attention to meditation and adherence to the precepts, but also would
suffer at the hands of the state. Accused of plotting rebellion against the Western
Qin (385-431), he was eventually exiled from Chang' an to Mount Yangtang in
Hobei. Although later pardoned and brought back to the capital, Xuan' gao was
ultimately claimed as a young victim during the atrocities leading up to the
persecution of Buddhism in 446 by the Northern Wei. Buddhabhadra has a
biography in the GSZ, T 50, no. 2059: 334b-335c, and the Chu sanzang jiji,
T 55, no. 2145: 103b 104a, summarized by HURVITZ, "Chih-i," p. 115 note 1.
Xuan'gao's biography is located in the GSZ, T 50, no. 2059: 397a-398b, and is
summarized by HURVITZ, "Chih-i," p. 115 note 2.
79. Sui Tiantai Zhizhe dashi biezhuan, T 50, no. 2050: 192c21-23.
nABS 23.2 282
the tum of the seventh century and, later, during the early years of the
Tang. Scholars such as ANDO Toshio for instance, suggest
that during this critical juncture in history it was necessary for Tiantai to
husband its resources, one option being to broaden its base of appeal
within the religious community. Under the circumstances, ANDO pro-
poses, the natural affinity between Tiantai and the Madhyamaka-oriented
ideas of the Sanlun and Silun traditions, with their concept of two
levels of truth and the doctrine of the middle way between existence and
non-existence, may have combined with political realities to influence
Guanding to designate Nagarjuna, the founder of the
school in India, to the position of "high ancestor" of Tiantai in the first
decades of the seventh century.80 Other scholars such as SHIOIRI Ryodo
take a more caustic posture and argue that the gradual rise of
a Tiantai self-awareness in south China may have been due in part to
competition and conflict between and among the Sanlun group on
Mount She, Chengshi and Tiantai.
81
This all must be reviewed in light of recent scholarship spearheaded by
HIRAI Shun'ei and others that reveals a far more complex
relationship than had been previously realized between Zhiyi and
Guanding and, by extension, between Guanding and his older contempo-
rary and spokesperson for the Sanlun tradition, Jizang (549-623).82
Jizang, Parthian on his father's side but born and educated in the
southern capital of Jinling (aka Jiankang during the
Northern and Southern Dynasties period, had been a student of Palang
80. ANno Toshio: Tendaigaku: konpon shiso to sono tenkai 0
-{-O) Jll*-i (Kyoto: Heirakuji shoten, 1968), pp. 8, 10-16. SIllMAn Dait6 11if
(Tendai kyogakushi [Tokyo: Meiji shoin 1929], pp. 116-118)
agrees with this assessment. He sees Tiantai as the consolidation and
systematization of three (in addition to itself) of thirteen pre-Tang traditions:
Sanlun, Silun, and Niepan or the so-called Nirvfu:!a school. VI Hakuju '-#
isfi (Shina Bukky6shi [Tokyo: Iwanami shoten 1946], p.116)
adds the Chengshi mG. (Satyasiddhi or Tattvasiddhi) tradition as a fourth group
that collapsed into Tiantai.
81. SHIOIRI Ry6d6: "Shoki Tendai-san no kyOdan-teki seikaku," Nippon Bukkyo
gakkai nenpo B 39 (1973): 136.
82. Jizang's biography, located in the XGSZ, T50, no. 2060: 513c-515a. The most
exhaustive biography and study on Jizang is included in HIRAI Shun'ei's
Chiigoku hannya shisoshi kenkyi1: Kichizo to Sanron gakuha
liJf)j; 0 (Tokyo: Shunjusha 1976), pp. 345-642. See also Ming-
Wood LID, Madhyamaka Thought in China (Leiden: E.J. Brill 1994), pp. 82-187.
PENKOWER 283
(507-581), an older contemporary of Zhiyi who resided at the
Xinghuangsi in Jinling by imperial edict between 558 and his
death in 581, that is, during the first of two extended periods (567-575
arid 585-589) that Zhiyi spent in the Chen capital.
83
Falang himself is
usually credited with the revitalization of Sarilun, which had begun in
north China with the translation and propagation of the Sanlun texts by
Kumarajlva but which had been overshadowed in the late-fifth century
by the rise in interest in Chengshi and Nirval.1a literature.
84
Falang in
turn was a student of Sengquan (n.d.) about whom little is known
except that an imperial order by Emperor Wu 1EC;* of the Liang (r. 464-
549) brought him to study under Senglang 1'/!1'M (Korean: Siingnang;
n.d.), a Korean monk who had come south to become abbot of the
Qixiasi on Mount She at the turn of the sixth century.85 Much
like Huiwen and Huisi, both Senglang and Sengquan are depicted as
quiet meditators who steered clear of the capital, preferring instead "the
seclusion of hidden forests and a taste for meditation," leaving Falang -
and Jizang in the next generation - to advance the Sanlun cause.
86
83. Falang's biography is located in the XGSZ, T 50, no. 2060: 672a-b. No known
writings survive.
84. The Chengshi tun RlGlrtnii (Satyasiddhi-sastra, T 32, no. 1646, by Harivarman as
translated by Kumarajlva) was originally popularized from within the ranks of
Kumarajlva's group but became, by the early-sixth century, so popular in its own
right as to eclipse Sanlun. See FUKuHARA Ryogon Jojitsuron no ken-
kya: Bukkyo shoha no gakusetsu hihan RtJftniiO)li7fJi: 0 m*u
(Kyoto: Nagata bunsh6do 1969), pp. 103-108. Jizang and his group would even-
tually criticize it as being a HInayana work (Santunxuanyi or The Pro-
found Meaning of the Three Treatises), T 45, no. 1852: 3c-4c, and Fahua xuan-
tun i**1{tnii or Profound Treatise on the Lotus), T34, no. 1720: 364a20-b2).
Interest in the Nirvlil)a Satra (T 12, no. 374 and 375) was so great during the
fifth and sixth centuries that scholars now talk of a so-called Nirvana School. See
FUSE Kogaku ;ffiM!!l'i!i-ffi, Nehansha no kenkya two volumes
(Tokyo: Kokusho kankokai 1942). Although that scripture falls under the rubric
of tathiigatagarbha literature and so stands in opposition to the Madhyamaka
tradition from which Sanlun derives, most Sanlun adherents, including Jizang,
embraced this text as the final sermon of the Buddha.
85. No biography survives for Sengquan. HIRAI (ChUgoku hannya shisoshi kenkya,
pp. 269-275) discusses what little is available. Senglang died in his seventies at
the end of the Daye era (605-617). His biography is located in the XGSZ, T 50,
no. 2060: 507c-508a.
86. XGSZ, T50, no. 2050: 477c5-7. Despite these characterizations, Zhanran notes
in his Fahua shiqian (full title: Fahua xuanyi shiqian
or Commentary on The Profound Meaning of the Lotus Satra, T33, no. 1717:
JIABS 23.2 284
Every indication is that Jizang, in particular, took that mission
seriously. By the age of nineteen (in 568) he was appointed assistant
homilist in charge of guests at the Xinghuangsi by Fahmg. In that
capacity and despite his youth it is conceivable that Jizang was among
several unnamed advanced disciples sent by Falang to participate in
debates on meditation (among other possible topics) against Zhiyi s.oon
after the latter entered the capital, an event that reputedly lasted for
several tens of days and is said to have cost Falang several students who
chose to remain with Zhiyi.87 Jizang andZhiyi were again together in
the capital during the four years prior to the overthrow of the Chen
court by the Sui in 589, Jizang by then a force in his own right (Palang
having died in 581) and Guanding having had recently joined Zhiyi as a
disciple. With the change of dynasty, Zhiyi left for his native Jingzhou
(in modern-day Hunan) and eventually ended up on Mount Tiantai for
the last time, while Jizang came to reside for the next ten years or so at
the Jiaxiangsi in Guiji (in modern-day Zhejiang), the sobriquet
by which he is ultimately known.
Sometime around the turn of the century, that is, around the time that
Yang Guang had committed his support to the building of the Guoqingsi
after the death of Zhiyi, the future emperor invited Jizang to reside in
the Huiri B in his area command of Jiangdu in Yangzhou. This
temple was one of four (two Buddhist, two Daoist) centers (dao-
chang) /""3l:.ffB. (xuantan) established to house prestigious monks of the
951a21-23) that Senglang "rebuked the Chengshi masters, [leaving them] tongue-
tied and speechless."
87. This according to the Sui Tiantai Zhizhe dashi biezhuan, T 50, no. 2050:
19258-11 and 192b7-14. Zhiyi criticized Sengquan's form of meditation focusing
on the perception of impermanence as superficial and dubious. Among the
Sanlun adherents with whom Zhiyi is known to have associated is Zhibian
(n.d.), a co-disciple of Falang under Sengquan, who invited Zhiyi to take up
residence in the Songxisi (XGSZ, T 50, no. 2060: 564c3-4). Guanding's
Sui Tiantai Zhizhe dashi biezhuan (T 50, no. 2050: 192b21-22) associates a
certain Huibian with this invitation and the XGSZ (T 50, no. 2060: 564c
22-23) lists a monk by that name as a disciple who followed Zhiyi to Mount
Tiantai when he left the capital in 575. SEKIGUCm Shindai (Tendai
shilwn no .LI::WiO)lVfJi; [Tokyo: Iwanami shoten 1969], p. 131) thinks
that these two monks as well as a certain Jingbian also listed as a Zhiyi
disciple, are the same person. SATO (Tendai daishi no kenkyu, pp. 241-265)
thinks that Jingbian composed the Xiao zhiguan IJ\.LI:: Wi (full title: Xiuxi zhiguan
zuochan Jayao or Essentials for Sitting in Meditation and
Cultivating Calming and Contemplation, T 46, no. 1915) attributed to Zhiyi.
PENKOWER 285
realm. Sometime prior to Yang Guang ascending the throne as Emperor
Yang in 604, he built a similar but more prestigious complex in
Chang' an called the Riyansi B I&=lf. 88 Jizang moved to this state-run
center in the capital also by invitation of the prince, where his skills as
an orator gained him increasing recognition. Jizang's close association
with the Sui court appears not to have worked against him once the
Tang replaced the Sui; he spent the last years of his life as one of the
Ten Monks of Great Virtue (shidade), a collective system of
leadership established by the first Tang emperor Gaozu jilijlEI. (r. 618-
626) as part of a general set of policies aimed at weakening the
autonomy of the Buddhist church.
89
Jizang's erudition and breadth of study are well-documented. Among
this best known works are the Dacheng xuanlun **3(IDfB (Treatise on
the Profound [Meaning of the] Mahayana) and Erdi yi (Meaning
of the Two Truths), comprised of expositions on a number of topics
central to Buddhism, and several commentaries on the Sanlun texts,
which highlight the polemical nature of his Buddhism, the most notable
being the Sanlun xuanyi (Profound Meaning of the Three Treatises), a
work that outlines and contrasts the ideas and doctrinal positions of the
three quintessential Sanlun treatises to those of non-Buddhist and non-
Sanlun literature. More to the point of this discussion, Jizang was also a
great proponent of the Lotus and Nirvii;lJa siltras, two scriptures which of
course have special meaning for Zhiyi.
9o
Jizang reputedly lectured on
the Lotus Satra some three hundred times over the course of his career
as compared to some one hundred lectures he gave on the Sanlun.
91
Jizang himself comments on his eventual shift of focus away from
88. See YAMAZAKI Hiroshi LllOIiiT$;:, Zui-To Bukkyoshi no kenkyu
(Kyoto: H6z6kan 1967), pp. 85-115, where the flrst four temples are iden-
tilled as the Buddhist Huiri and Fayun and the Daoist Yuqing .3S.l'f
"S:l!l and Jintong :3Z:Wll"S:lll, and the Riyansi is discussed.
89. For the Ten Monks of Great Virtue see Stanley WEINS1EIN, Buddhism under the
T'ang (Cambridge: Cambridge University Press 1987), pp. 9-11; 0DA Yoshihisa
"T6sho no jiidaitoku ni tsuite mWO)+:k:@vC:-JvY"c," Oryo shigaku
5 (1979): 51-79;and YAMAZAKI Hiroshi LllOIiiT$;:, Shina chasei Bukkyo
no tenkai (Kyoto: H6z6kan 1971), pp. 602-607.
90. Dacheng xuanlun, T 45, no. 1853, and Erdi yi, T 45, no. 1854. For a listing of
extant works attributed to Jizang see HIRAI, Chagoku hannya shisoshi kenkyu,
pp. 355-356.
91. XGSZ, T50: 514c27-28.
JIABS 23.2 286
studying the Sanlun treatises and onto the Lotus.
92
Of the twenty-six
extant works that are attributed to him, five are on that scripture -
including the important Fahua xuanlun (Profound Treatise on the Lotus)
in which he glorifies the Lotus Sutra as "perfect in its teaching and
complete in its principle" (jiaoyuan liman) - and together
comprise some thirty percent of the total volume of his extant literary
corpus.
93
Indeed, in his works on the Lotus, Jizang calls on a classifi-
cation of the teachings scheme (panjiao) called the three dharma
wheels (sanzhong falun) to defend the less than doctrinally
rigorous sutra against those who criticized it as being expedient and
incomplete. According to this scheme, the teaching career of the Buddha
is divided into three successive phases. The original dharma wheel fJt:zjs:
t*iMfl (genben falun) was turned when immediately after his enlighten
ment the Buddha preached the one vehicle -* (yicheng) of the Huayan
jing for the benefit of bodhi-sattva alone; this was followed by a forty-
year career of turning the ancillary dharma wheel (zhimo
falun) during which time the Buddha preached the three vehicles ==*
(sancheng) for the benefit of those with lesser capabilities; finally, when
all were ready to receive it, the Buddha turned the dharma wheel
intergrating the ancillary into the original (shemo guiben
falun) and preached the Lotus, explaining the expedience (fang-
bian) of the three vehicles and revealing the one vehicle. Thus, Jizang
insists, the content of the Lotus Sutra, in its position as third dharma
wheel and as accessible to all, embodies the same supreme one vehicle
teaching and is as pure and perfect as the Huayan jing, which occupies
the position of first dharma wheel but is confined to bodhisattva alone.
94
Indeed, as KAN'NO Hiroshi points out, both Zhiyi and Jizang share a
generational interest in criticizing the earlier position (especially as
92. Fahuajing tonglue XZJ 43: la4-5.
93. T 34, no. 1720: 364bS. See also his Fahua youyi (Outline of the
Lotus), T34, no. 1722: 647c1S-19, where the Lotus is called the "true essence of
all scriptures" (zhongjing zhi shitl) and the "secret treasury of all
Buddhas" (zhufo zhi bimizang). On Jizang's attempts to harmo-
nize the Lotus with the Sanlun treatises see SUEMITSU Yasumasa
"Kichizo no Hokekyo kaishaku ni tsuite 1<:0vY'C,"
Indogaku Bukkyogaku kenkyii 32 (19S3): 239-242.
94. Fahua youyi, T 34, no. 1722: 634c16-23 and 635a5-S. See also Lm, Madhya-
maka Thought in China, pp. 125-135, and HIRAI, Chiigoku hannya shisoshi
kenkyii, pp. 506-510.
PENKOWER 287
espoused by the Lotus commentator Fayun that the Lotus was
inferior to the Nirvfir;a Satra because it lacked the idea of the eternal
Buddha.
95
Traditional Tiantai literature makes much of Jizang's appre-
ciation of the Lotus, especially his admiration for Tiantai interpretations
thereof. Indeed, two seemingly spurious letters allegedly written by
Jizang to Zhiyi and included in the Guoqing bailu are traditionally
considered to be the source of the idea that Jizang denounced Sanlun to
become a disciple of Zhiyi on the strength of the latter's insights on the
Lotus.
96
In contrast to other snippets of correspondence that have been
preserved between the two monks and which are neutral in tone
(including other letters in the Guoqing bailu),97 in the letters in question
Jizang pleads with Zhiyi in exceedingly reverential terms to provide
instruction on the Lotus, identifies Zhiyi with such luminaries as
Maitreya, Confucius, and even Nagarjuna, and vows to become Zhiyi's
disciple.
98
Modern scholarship has long all but unanimously rejected a
conversion of Jizang under Zhiyi.99 Indeed, nowhere in the Guanding-
95. See KAN'NO Hiroshi "A comparison of Zhiyi's and Jizang's
Views of the Lotus satra: Did Zhiyi, after all, Advocate a "Lotus Abso-
lutism?", S6ka daigaku kokusai Bukky6gaku k6t6 kenkya nenp6
3 (1999):
96. Guoqing bailu, T46, no. 1934: 821c26-822a2 and 822a13-26.
97. Letters written by Jizang to Zhiyi are found in the Guoqing bailu, T46, no.
1934: 821a14-21. See also a note from Zhiyi to Jizang preserved in Zhanran's
Fuxing (T 46, no. 1912: 821c21-822b2): "When there is understanding but no
practice, one will not be able to subdue the [ten thousand] things; when there is
practice that is not supported by understanding, one cannot convert others."
98. In Jizang's alleged second letter, dated 597, that is eight years prior to the
completion of the Sui Tiantai Zhizhe dashi biezhuan and the appearance of a
lineage statement in Guanding's introduction to the Mohe zhiguan, are also
anachronistically found references to (a) the direct transmission, referring to
Huisi and Zhiyi as "two venerables, one succeeding the other"
(liangzun shaoxi), and (b) the prophecy that Zhiyi (i.e., Tiantai) would spread the
dharma in India. For a discussion on the correspondence between Jizang and
Zhiyi and the spurious nature of a Jizang discipleship under Zhiyi, including
translations of the pertinent materials, see CHEN, Making and Remaking History,
pp. 6-39, and his "Stories from the Life of Chi-tsang and Their Use in T'ien-t'ai
Sectarian Historiography," Asia Major third series 11.1 (1998): 53-97.
99. See, for example, HIRAI Shun'ei's Chagoku hannya shis6shi kenkya, "Kichizo
to Chigi: kyoden chuso 0 meguru sho-mondai 0
T6y6 gakujutsu kenkya 20.1 (1981): 101-116, which
compares the various accounts of the relationship between Jizang and both Zhiyi
and Guanding as found in the Guoqing bailu, XGSZ, and the FZTJ, and his
nABS 23.2 288
composed official biography. of Zhiyi nor in the biographies of either
Zhiyi or Jizang in the Xu gaoseng zhuan (Supplement to the Biographies
of Eminent Monks), compiled by Daoxuan (596-667) and c.ompleted in
664, is there any mention of Jizang as a disciple of Zhiyi or a Jizang
invitation beseeching instruction on the Lotus from Zhiyi, events that
should not have gone unrecorded.lOo
To the contrary and regardless of the nature of the professional
relationship between Zhiyi and Jizang, the latter specifically and
repeatedly cites "the time-honored ideas of Guannei"
(Guannei jiuyi), "the transmission [of Senglang of the Qixiasi] on
Mount She" :JiU.!t;f1W (Sheling xiangchuan), and "the transmission of
the Mount [She] gate" (Shanmen xiangcheng) as the source
of his authority, thus linking himself through these Chinese monks to
Kumarajiva.
10I
Indeed, declarations of this sort are found in such
relatively early works as Jizang's Fahua xuanyi, composed sometime
during the last decade of the sixth century while still in south China
residing at the Jiaxiangsi,102 that is, around the time of his alleged
conversion to Tiantai and upwards of a decade before Guanding's
ground-breaking Hokke mongu no seiritsu ni kansuru kenkyu RlGiz:
(Tokyo: Shunjiisha 1985).
100. Zhanran, who accepts a Tiantai conversion by Jizang under Zhiyi, makes
reference to the Guoqing bailu entries in support of the claim that Jizang was
converted by Zhiyi (Fahua wenjuji, completed around 775; T34, no. 1719:
213a28-b17). By the early-ninth century onwards, various components of the
conversion story are also cited in Japan. See, for example, Saich6's Ehyo
Tendaishu (full title: Dai To Shinragi shoshii gishO ehyo Tendai
or Dependence on Tendai Doctrine
by Scholars from the Other Schools of China and Korea; DZ 3: 362-363),
compiled in 813. For a complete summary of this literature see CHEN, Making
and Remaking History, pp. 107-152.
101. See, for example, Fahua xuanlun, T34, no. 1720: 440c18, Weimo jing yisu
(Commentary on the Vimala Sutra; T38, no. 1781: 916c15, and
Erdi yi, T 45, no. 1854: 103b27. Guannei refers to the area in and around Chang
'an in which Kumiirajiva introduced the texts of Nagarjuna. Jizang's partisan
spirit is also revealed in his attacks against anyone who criticized his teacher or
who lacked a lineage. He said of one monk: "He lacks a transmission from
teacher to disciple fllUffiTff$ (wu shizi xiangchuan). Surely, one's scholarship
depends upon what cine learns from one's teacher" (Fahua xuanyi, T45,
no. 1852: 36c).
102. Dacheng xuanlun, T 44, no. 1853: 49b, and Fahua tonglue, XZJ 43. cf. SAT<),
Tendai daishi no kenkyu, p. 321.
PENKOWER 289
second editing of the Mohe zhiguan in which a statement on lineage is
introduced, as well as in texts that date from after the tum of the century
(and thus postdate Zhiyi), such as his Vimalakfrti commentary composed
when Jizang had already established himself as a leading voice in
Chang'an.!03
That such self-descriptions should be understood primarily as efforts
to chronicle one's religious history and not as evidence of the existence
of autonomous entities called, for example, Sanlun (or Tiantai and so
forth) during this time is highlighted by HIRAI, who insists that Sanlun
be better thought of as a group or faction (ha) and not as a full-
blown tradition or school * (sha).104 This caveat is particularly impor-
tant when we tum to the relationship between Jizang and Guanding.
Recent scholarship is beginning to reveal the extent to which Guan-
ding's editorializing of the lecture notes of Zhiyi was influenced by the
writings of the Sanlun adept. HIRAI Shun'ei's ground-breaking study on
Zhiyi's Fahua wenju (Words and Phrases of the Lotus satra), for
instance, shows that large sections of that text, which began as Zhiyi
lectures at the Guangzhesi in Jinling in 587 and were not brought
together for the last time by Guanding until forty years later,105 were
taken from or influenced by the Lotus Satra commentaries of Jizang, by
then recently deceased.l
06
Portions of the Fahua xuanyi were also
revised to include ideas and language borrowed from Jizang's works.1
07
Nor was Jizang's influence over Guanding limited to interpretations of
the Lotus. HIRAI makes a case that Zhiyi's Vimalakfrti commentary,
completed prior to Jizang's commentary on the same, was later edited by
Guanding and shows extensive borrowings from the Jizang's work.
108
SAKAMOTO K6baku moreover suggests that the link between
103. LID, Madhyamaka Thought in China, pp. 86-87. On the dates of composition of
Jizang's works see also HIRAI, Chugoku hannya shisiJshi kenkyu, pp. 358-381.
104. See HIRAI's Chagoku hannya shisiJshi kenkyu. See also YOSHIZU Yoshihide
"The Relation between Chinese Buddhist History and Soteriology,"
translated and edited by Paul GRONER, in Paths to Liberation: The Marga and
its Transformations in Buddhist Thought, edited by Robert E. BUSWELL, Jr. and
Robert M. GIMELLO, Kuroda Institute, Studies in East Asian Buddhism, no.7
(Honolulu: University of Hawai'i) 1992, pp. 309-338.
105. T 34, no. 1718: 1b19-20.
106. See HIRAI Shun'ei, Hokke mongu no seiritsu ni kansuru kenkyu.
107. See also SATO, Tendai daishi no kenkyu, pp. 340-363.
108. HIRAI, Hokke no seiritsu ni kansuru kenkyu, pp. 64-71
JIABS 23.2 290
Zhiyi and Zhanran with regard to the concept of buddha-nature of the
insentient, first promoted within Tiantai in the latter's lin' gangbei :tit: IJijJU
(Diamond Scalpel),109 is in fact Guanding through Ji;zang. Guan-
ding's commentaries on the NirviiY}a, the scripture par excellence
through which buddha-nature is discussed in China, was written some
twenty years after the death of Zhiyi. Although insentient buddha-nature
is explicitly denied in one instance,110 this work is influenced by such
texts as the Dacheng xuanlun where Jizang had earlier worked out his
own justification for the idea of insentient buddha-nature on the basis of
such Sanlun ideas as "buddha-nature as the first principle of emptiness"
and the "middle way."1ll .
109. T 46, no. 1932.
110. Daban niepanjing su, T 38, no. 1767: 184c.
111. Niepanjing youyi (T 38, no. 1768: 232b), a text written after Jizang moved north
around 600, and Dacheng xuanlun (T 45, no. 1853: 35c), respectively. See
SAKAMOTO Kobaku "Keikei daishi no mujo busshO-setsu
Tendai gakuhi5 10 (1967): 61-63, and his "Nehan-
gyosho seiritsu ni kansuru isshiten: Doro, Horo 0 megutte
0 mM'
19.2 (March, 1971): 301-303. For Jizang's ideas on insentient buddha-
nature see Dacheng xuanlun , T 45, no. 1853: 38c-40c; also KAMATA Shigeo:
ilfflil;1i:t, Chagoku Bukkyo shisoshi kenkya (Tokyo:
Shunjiisha, 1967), pp. 30-50; HIRAI: Chugoku hannya shisoshi kenkya, pp. 617-
40; Aaron K. KOSEKI: "PrajiHiparamita and the Buddhahood of the Non-sentient
World: The San-Iun Assimilation of the Buddha-Nature and the Middle Path
Doctrine," nABS 3.1 (1980): 16-33; and LID: Madhyamika Thought in China,
pp. 160-187. For Zhanran's position see my "Buddhism during the T'ang," pp.
382-555.
;SATO Tetsuei, without crediting developments to Jizang's influence, also
concludes that the Guanyinxuanyi (T 34, no. 1726), a work tradition-
ally attributed to Zhiyi as recorded by Guanding and containing the clearest
expression of the Tiantai doctrine on evil, should be ascribed to Guanding. The
Japanese Jodoshii 1Jl-* monk, Fujiyaku (1707-1781), was the first to question
the authorship of this text, primarily on the basis of the inclusion of this
controversial doctrine (Bukkyo daijiten 1: 771a-772b). More recently, SATO
Tetsuei, while acknowledging that the idea of evil nature is a logical extension of
the doctrine of xingju (nature inclusion) found in Zhiyi's later works,
nonetheless concludes that the commentary, and hence the explicit idea of evil
nature, should be assigned to Guanding (Tendai daishi no kenkyu, pp. 475-496,
and its Zoku, pp. 411-427). ANno Toshio, who finds the idea of evil nature
implicit in other works definitively assigned to Zhiyi, and who notes Guanding's
failure to discuss the idea in such works as his Daban niepan jing xuanyi,
challenges SATO's conclusions (Tendaigaku, pp. 387-414). For a discussion of
PENKOWER 291
These borrowings have led some to cry foul. Indeed, there is recent
speculation that the appearance of a Jizang discipleship under Guanding
by dint of the latter's expertise on the Lotus in Guanding's biographical
entry in the rid-seventh century Xu gaoseng zhuan was initiated by
followers of Guanding for the principal purpose of warding off or
otherwise deflecting any potential criticism or allegations of.piracy.1I2
Yet, even without carefully delving into the complex issue of the
conventions of textual and commentarial practice during the late-sixth
and early-seventh centuries, militating against the idea that Guanding's
was a case of plagiarism in the contemporary sense of the term is
KAN'NO Hiroshi's recent discovery that Jizang too incorporated large
sections of Zhiyi's Weimo jing su into his commentary on the Vimala-
kfrti without crediting his source.!13 KAN'NO further goes on to argue
elsewhere that Zhiyi and Jizang shared the same organic view of
scripture. Briefly stated, both principally employed doctrinal classifica-
tion systems in such a way as to emphasize not a hierarchy of value
between and among scriptures but rather the "salvific capacity" of the
practitioner. In other words, both understood all Mahayana texts to be
equal in their expression of ultimacy and different in relation to sentient
beings proper, with the Lotus being the most universal or, for Zhiyi,
solely perfect.
114
In Peter GREGORY's terminology, this is represen-
tative of the hermeneutical function of panjiao schemes and not their
more polemical or sectarian use that came to the fore in the next
century.115 KAN'NO calls that a shift away from the perfect teaching
this theory as characteristic of Tiantai thought see Neal DONNER, "Chih-i's
Meditation on Evil," in David W. CHAPPELL (ed.), Buddhist and Taoist Practice
in Medieval Chinese Society, Buddhist and Taoist Studies II, Asian Studies at
Hawaii, no. 34 (Honolulu: University of Hawai'i Press 1987), pp. 49-64, and
Brook ZIPORYN, What's So Good About Evil: Value and Anti-value in Tiantai
Thought and its Antecedents, Ph.D. dissertation (University of Michigan), 1996.
112. CHEN: Making and Remaking History, pp. 73-82. Jizang's conversion under
Guanding is found in XGSZ, T 50, no. 2060: 584b13-18 and is translated by
CHEN on p. 80. The account is almost universally accepted as spurious by
modern scholarship due to the younger status of Guanding, and especially the
historical inaccuracy of the story, for which there is no corroborating evidence.
113. I thank an anonymous reader for this reference.
114. KAN'NO: "A Comparison of Zhiyi's and Jiang's Views of the Lotus Siltra."
115. GREGORY: Tsung-mi and the Sinification of Buddhism (Princeton: Princeton
University Press 1991), pp. 115-116.
JIABS 23.2 292
absolutism ofZhiyi and towards a Lotus absolutism characteristic of
Zhanran.
116
This is not to argue that the pull and tug, appropriation and expro-
priation between Guanding and Jizang in the first decades of the seventh
century were not without arguments on soteriological, institutional, or
other grounds of the type that accompany a fledging sense of self-iden-
tity.ll? In this regard, recent scholarship has begun to take note of the
discrepancies between Guanding's autobiographical statement attached to
the end of his Daban niepan jing xuanyi (Profound Meaning of the
Mahiiparinirvii1}a-sutra), completed around 619,118 as well as
documents in the Guoqing bailu, compiled by Guanding and completed
around 607, and his biographical entry in the mid-seventh century Xu
gaoseng zhuan. This revised account, followed by later Tiantai but
which nonetheless finds little of note to record beyond 619, is taken in
large part from a memorial stele composed and erected for Guanding by
his followers at the Guoqingsi shortly after his death in 632.1
19
Favoring the earlier sources over the latter suggests a less illustrious
personal career for Guanding and a less enviable relationship between
the post-Zhiyi community at the Guoqingsi and the late Sui court than
had previously been assumed.l
20
For despite Yang Guang's personal
116. Zhanran's position s discussed in my ''Making and Remaking Tradition."
117. Indeed, throughout the works of Zhiyi/Guanding and Jizang are found both criti-
cisms and admiration one for the other, sometimes identified by name, some-
times simply as "someone" (youren A or huo KAWAMURA Kosho 1'1J;ft
"Kanjo-sen Nehangyo gengi ni okeru 'aruhito' to wa dare osasu ka
:f;I
kenkyii 34.1 (December, 1985): 218-225, and KAN'NO
Ifuoshi: Chiigoku hokke shiso no (Tokyo: Shun-
jiisha 1994), pp. 788-791.
118. T 38, no. 1765: 14b20-15a8. Guanding's Dahan niepanjing xuanyi was written
as a compendium of his much larger Dahan niepan jing su (T 38, no. 1767), a
work begun in 614 and completed five years later (14c14-15).
119. Guanding's biography is located in XGSZ, T 50, no. 2060: 584a-585b. For the
reference to the stele see 585b8-11.
120. See CHEN's comparison of the three texts in his Making and Remaking History,
pp. 40-82; TAMURA Tokkai "Shoan daishi no Nehangyo gensho ni
tsuite Sange gakuhO n.s. 1.4
(1931): 35-44; SASAKI Shokaku "Kanjo-den ni tsuite no ichi ko-
satsu IndogakuBukkyogaku kenkyii
49 (1976): 18-119; FUJIMOTO Ken'ichi "Kanjo-sen Daihatsu
nehangyo gensho ni tsuite," Tendai gakuhO 17 (1974), pp. 113-116; KIYOTA
PENKOWER 293
devotion to Zhiyi and his support of the his followers through the early
years of the seventh century, represented primarily by his commitment
to the construction of the Guoqingsi, once the Sui had shored up its
southern bases, it shifted its attention to the northern capital of Chang' an
and its constituents there. Yet neither Guanding, who spent eight years
as an emissary to the court (597-605), nor any member of the Guoqing-
si, including its head abbot Zhiyue, appears to have been singled out as a
personal favorite of the court, or in recognition of ritual expertise or
scholarly erudition, or to reside in the prestigious Riyansi (as had Jizang,
for instance).121
Curious is the embellishment in Guanding's biography of an event that
took place in 602. The biographical entry begins with a personal letter
of invitation to the capital from Yang Guang to Guanding, in which the
latter is lauded as an "advanced disciple" (gaozu) of Zhiyi and
results in a particularly successful three-month summer lecture tour by
Guanding on the Vimalakrrti-nirdda-sutra attended and praised by the
future emperor.
122
Yet the Guoqing bailu reports simply that during a
debate conducted by several well-known Lotus experts in the palace,
Yang Guang found himself wanting to confer with Zhiyi's Lotus com-
mentary and sent an edict to the Guoqingsi community at large asking
that a monk with familiarity with the text bring it to the capital. The
document goes on to say that Guanding was selected as courier, that
upon his arrival in the capital he was instructed to hand over the com-
mentary to the office in charge of copying, that he then waited around to
proofread the prepared galleys, which were then sent to the palace for
recitation.
123
In other words, debate on the Lotus, including Zhiyi's
comments on it, was conducted without the physical presence or invited
participation of Guanding (or, it appears, any Zhiyi follower). While the
Xu gaoseng zhuan version of this event serves mainly to enhance the
Jakuten mffBlRk "Sh6an Kanj6 kenkyU josetsu
Tendaigaku ronsha I 1 (May, 1984): 65-113; and HIRAI: Hokke
mongu no seiritsu ni kansuru kenkya.
121. Zhiyue, in his capacity as head of the Guoqingsi, has only one recorded
invitation to the Sui court to participate in a vegetarian feast in the memory of the
newly deceased empress of Emperor Wen who died in 602 (SGSZ, T 50, no.
2060: 570c15-16).
122. XGSZ, T 2060: 584c8-l2.
123. Guoqing bailu, T 46, no. 1934: 814clO-20; cf., CHEN, Making and Remaking
History, pp. 47-50.
JIABS 23.2 294
personal prestige of Guanding, the subtle shift in political favors it
anticipates nonetheless may have had a homogenizing effect on the way
temple complexes perceived and presented themselves, or OI} Guanding
himself as keeper of "the approach to the teachings Zhiyi practiced in his
own mind," making the idea of employing a statement on lineage (i.e., a
reminder of the Guoqingsi Zhiyi pedigree) to introduce the M ohe
zhiguan just three years later all the more poignant.
CHEN Jinhua however points to another event recorded in Guanding's
biography and omitted (read: replaced) in the Xu gaoseng zhuan that
gives pause to the long-held consensus spearheaded by TSUKAMOTO
Zenryu that Zhiyi's followers continued to enjoy a close
beneficiary relationship with the court through to the end of the Sui - so
much so that Tiantai was shunned by the Tang for its close association
with the preceding dynasty.124 Namely, Guanding reports being called to
the capital to participate in the Rinyansi controversy of 611 (the subject
matter of which is lost to history), only to be brought up on charges of
sorcery and subsequently arrested and sent north.125 Would such
treatment have been possible had some level of withdrawal of imperial
support for the group at the Guoqingsi not already taken place?126
During the Riyan[si] controversy, I was summoned to the capital by the emperor.
On route, we encountered a flood in Taolin (present-day Huayin county, Honan),
and I was separated from my group at night. Later, I was falsely accused of
sorcery, arrested, and escorted to Yu and Ii (in modern-day Hobei). While
124. TSUKAMOTO Zenryii, "Zui no Kanan seifuku to Bukky6
in TSUKAMOTO Zenryii chosakushii vol. 3 (Tokyo: DaitO shup-
pansha, 1975), pp. 172-173.
125. According to the Xu gaoseng zhuan entry for the same year, Yang Guang, now
long ensconced as Emperor Yang, was at his field headquarters in Zhuoye (in
the present-day northern province of Hobei) touring the area in preparation for
conducting one of the three ill-fated expeditions he would wage against
Koguryo. Moved by a memory of Zhiyi, the emperor summoned Guanding to
reminisce about Zhiyi in the manner of "two co-disciples enjoying each other's
company," a reference to the emperor's (then prince) having taken lay bodhi-
sattva precepts under Zhiyi.
126. That Guanding did not spend much if any time detained in the north is suggested
in that his autobiography states that during the upheaval that resulted in the
downfall of the Sui in 616 he moved no fewer than five times, spending the bulk
of that time in Anzhou.
PENKOWER 295
crossing an ice-covered river in the north, my horse fell in, but I survived.
Surrounded by danger and walking on thin ice, it was like treading among the
dead. How to describe the sheer dread and trepidation? 127
While CHEN speculates that Jizang might have headed the monks at the
Riyansi who bested Guanding and opened the way for his arrest, there is
no substantiating evidence to support that claim. Nevertheless,the event
seems to have marked Guanding's exit from things political. Rather than
ever venturing back into religious or secular politics (even after the
Tang replaced the Sui), Guanding choose instead a quiet retreat on
Mount Tiantai, devoting himself to the writing of his commentaries of
the Nirvo;(la - the scripture that had initially caught his attention under
his first teacher Huizheng ~ f 2 E ; (n.d.) and had later brought him to Zhiyi
- and to the final editing of the Zhiyi texts, which seriously engaged
him in Jizang's commentaries (among others), as previously noted.
Thus contrary to traditional claims of the absolute centrality of the
Lotus, the decades immediately following the death of Zhiyi can be
characterized as a highly fluid and uncertain state with regard to textual
classification, doctrine, and practice. Guanding, in the process of formu-
lating a historical justification for the Mohe zhiguan and in part influ-
enced by competition from Jizang and Sanlun, began the process of
bringing together the disparate strategies that reflected competing claims
of the Dazhidu [un and the Lotus. For in the end Guanding, or at least
his immediate followers at the Guoqingsi in the fourth decade of the
seventh century, came to understand his legacy in terms of his close
association with mastery of practice and texts, that is, as keeper of the
"approach to the teachings that Zhiyi practiced in his own mind" and
recorder of the history of the mountain complex. Guanding's official
identification (re: justification) of the perfect calming and contemplation
with Nagarjuna, to whom both Zhiyi and Jizang were aligned philo-
sophically, and with the Dazhidu [un, which was not one of the Sanlun
trilogy and served as a symbol of the practical orientation insisted upon
by Zhiyi, thus had the added appeal of lending inspired insight into
Guanding's claim to the Lotus. We thus find a a mid-seventh century
prototype, which was probably initiated by Guanding's disciples upon
127. T 38, no. 1765: 14c7-1O; cf., CHEN, Making and Remaking History, p. 51. For
the dating and discussion of this event see SAKAMOTO, "Kanj6-den [nil
kan[suru] ichi, ni no mondai."
JIABS 23.2 296
his death in 632, for defin.ing the true heir to Zhiyi. Guanding comes. to
be named in the third generation after Huisi and Zhiyi (omitting
Huiwen) on the basis of scripture.
[Guan]ding received the dharma from Tiantai [Zhiyi], inheriting the way of Huisi.
In the third generation t!t (sanshi) after [Hui]si and [Zhi]yi, there is no differ-
ence in their principles * (zong). In discernment and in preaching he constantly
depended upon the Lotus. He also lectured on the Nirvi'ilJa, linguangining
Jingming {'is, and so forth, and spoke on such ideas as round an
sudden, calming and contemplation, the four mindfulnesses, and so forth. His
breadth was considerable. Furthermore, Zhizhe [Zhiyi's] eloquence, flowing like
clouds and pouring like rain, was like the heavenly net or a necklace of previous
stones. Only Guan[ding] was able to uphold and comprehend [wHat was
preached by him].
PETER SKILLING
Vasubandhu and the Vyakhyayukti Literature
Justly famed for his Abhidharmakosa, for his Vijfiaptimiitratiisiddhi, and
other works, Vasubandhu looms large in the history of Indian
Buddhism. But despite his fame one of his most important works is
scarcely known to modern scholarship. This is the Vyiikhyiiyukti, or
"Principles of Exegesis", preserved in Tibetan translation. The work was
enormously influential, both in India and Tibet. Its importance in India
may be seen, for example, in the Nibandhana on the Arthaviniscayasutra
composed by Vlryasddatta at Nalanda during the reign of Dharmapala
(later half of the eighth century), or in the work of Vlryasddatta's
contemporary Haribhadra. At the same time the Vyiikhyiiyukti played a
key role in the formulation of the theoretical principles adopted for the
great task of translating Indian texts from Sanskrit into Tibetan, as
attested by the sGra sbyor bam po gfiis pa (Madhyavyutpatti). The
earliest manuscript witnesses of the Vyiikhyiiyukti are in the collections
of Dunhuang (ca. 9th-11th century) and Tabo (ca. 11th-13th century).
In Tibet its continued influence may be seen in the Entrance Gate for
the Wise (mKhas pa 'jug pa'i sgo) composed by Sa-skya PaI).gita (1182-
125112), in Bu-ston's History of Buddhism (Chos 'bywi, composed in
1321), and in written and oral teachings up to the present day.
I. The Problem of Authorship: The Works ofVasubandhu the Koakiira
Numerous works are attributed to a Vasubandhu in the colophons of
Sanskrit manuscripts, in Tibetan and Chinese translation, or by historical
traditions.! Debate continues as to whether Vasubandhu the Kosakara -
References to the Pali canon are to the editions of the Pali Text Society.
References to the Pili co=entaries are to NaIanda or Mahamakuta editions, as
specified. P = Daisetz T. Suzuki (ed.), The Tibetan Tripi!aka, Peking Edition,
Tokyo-Kyoto ("Otani Reprint"). BST = Buddhist Sanskrit Text Series, Mithila
Institute, Darbhanga. Sigla and references for the Vyakhyayukti texts are given in
Appendix 8. I am grateful to Cristina Scherrer-Schaub and Christoph Clippers
for their valuable co=ents.
1. See Hajime NAKAMURA, Indian Buddhism: A Survey with Bibliographical
Notes, Tokyo 1980, pp. 268-73 for a list, with detailed bibliography, of works
attributed to Vasubandhu.
Journal of the International Association of Buddhist Studies
Volume 23. Number 2.2000
JIABS 23.2 298
the author of th,e Abhidharmakosa- later "converted" to the Mahayana,
and composed the Vijfiaptivadin and Mahayana works ascribed to him
by tradition, or whether the latter were composed by another, earlier
Vasubandhu.
2
I do not intend here to rehearse the theses and counter-
theses that have been proposed,. or to discuss all of the works attributed
to Vasubandhu, or to wrestle with the problem of his date. Rather, I.will
only deal with a group of twelve works for which I believe there to be
sufficient internal or external evidence to assert that they were composed
by the Kosakara. I use two main criteria. The first is cross-references in
the works of Vasubandhu himself or those of his commentators. These
establish that the works are related: that they were known to and
accepted by Vasubandhu, or held by representatives of his lineage to be
his own compositions. The second criterion is style. Vasubandhu's prose
style is distinctive: it is confident and learned, replete with citations and
allusions to canonical and other literature, and to the opinions of
different teachers or schools. It often employs debate: an "opponent"
raises an objection, which is resolved through recourse to reasoning or
scripture. Vasubandhu's prose cannot be mistaken for that of, say,
Asanga or Candraldrti. His verse is terse: compact, concise, mnemomic,
2. See E. FRAUWALLNER, On the Date of the Buddhist Master of the Law Vasu-
bandhu, Rome 1951, which gives an extensive bibliography of earlier studies;
Padmanabh S. JA1NI: "On the Theory of the Two Vasubandhus," Bulletin of the
School of Oriental and African Studies XXI.1 (London 1958), pp.48-53; Alex
VIlA YMAN: Analysis of the Sriivakabhami Manuscript, University of California
Press, Berkeley and Los Angeles 1961, pp.19-24; N. AIYASWAMI Sastrin:
Vi1]1satikii Vijiiaptimiitratiisiddhilt, Gangtok 1964, pp. ii-vi; A.K. WARDER:
Indian Buddhism, Delhi 1970, pp.444-47; Akira HIRAKAWA in collaboration
with Shunei HIRAI, So TAKAHASHI, Noriaki HAKAMAYA, Giei YOSHIZU: Index
to the (P. Pradhan Edition), Part One, Sanskrit-Tibetan-
Chinese, Tokyo 1973, Introduction; Jean DANTINNE: Le Traiti des Cinq
Agrigats, Brussels 1980, pp.xiii-xviii; NAKAMURA,loc. cit.; Thomas A.
KOCHUMUTTOM: A Buddhist Doctrine of Experience, Delhi 1982, pp.xi-xiv;
Stefan ANACKER: Seven Works ofVasubandhu, Delhi 1984, pp. 1-24; Roger J.
CORLESS: "On the Continuity ofVasubandhu's Thought: A Suggestion from the
Continuity of Wittgenstein's Thought," in N.H. SAMTANI (ed.), Amalii Prajiiii:
Aspects of Buddhist Studies (professor P.V. Bapat Felicitation Volume), Delhi
1989, pp.455-62; Bhikkhu PAsADIKA: "Once Again on the Hypothesis of Two
Vasubandhus," in V.N. JRA (ed.), Kalyii1Ja-mitta: Professor Hajime Nakamura
Felicitation Volume, Delhi 1991, pp.15-21; Marek MEJOR: Vasubandhu's
Abhidhannakosa and the Commentaries preserved in the Tanjur, Stuttgart 1991
(Alt- und Neu-Indische Studien), pp.l-l3.
SKILLING 299
it could not be mistaken for that of Nagarjuna or Santideva. Applying
these criteria I classify the works into two groups:
(1) Works shown to be by the Kosakara on the evidence of internal
cross-references.or references by Vasubandhu's commentators, and con-
firmed by style, sources used, methodology, and development of ideas:
the Vyakhyayukti, the Karmasiddhiprakarara, the Pratftyasamutpadadi-
vibhaflganirdea, and the Pancaskandhaprakarara;
(2) Works that may be accepted as by the Kosakara on the evidence of
style, sources used, methodology, and development of ideas: the Gatha-
saf(Lgraha texts (and their excerpt, the the Vif(Lsatikii,
the Trif(Lsika, and the Trisvabhavanirdesa.
3
(1) Vyakhyayukti and (2) VyakhyayuktisiltrakhaI).t;iasata
The Vyakhyayukti and the Vyakhyayuktisiitrakharasata are interdepen-
dent. Without reference to the Vyakhyayukti, the Vyakhyayuktisiitra-
kharasata is a random collection of brief passages from the scriptures.
Without reference to the Vyakhyayuktisiitrakharasata, the Vyakhyayukti
is incomplete, since the latter gives the extracts in abbreviation. Given
this interdependence, there is no reason to doubt that the two works are
indeed by the same author - Vasubandhu, according to the colophons.
External evidence - the evidence of authors within the commentarial
tradition of Vasubandhu's works - supports the claim that the Vasu-
bandhu of the Abhidharmakosa is that of the Vyakhyayukti. Yasomitra,
3. That I do not mention other works ascribed to Vasubandhu, such as the com-
mentaries on the Madhyiintavibhiiga and the MahiiyiinasiltriilaT[lkiira, etc., does
not mean that I attribute them to a "second Vasubandhu", but simply that I have
not had the leisure to examine them thoroughly. For a proposal that the author of
the commentary on the latter was not Vasubandhu but AsaIi.ga, see Alex
WAYMAN, "Doctrinal Affiliation of the Buddhist Master AsaIi.ga," in Amalii
Prajiiii (see n. 2), pp. 202-03. For a careful assessment of the authorship of the
see Jens BRAARVIG, Vol. II,
The Tradition of Imperishability in Buddhist Thought, Oslo 1993, pp. cxvii-cxxx.
For the *Sukhiivatfvyilhopade1a, a treatise on the Pure Land attributed to
Vasubandhu and preserved only in Chinese, see Minoru KrYOTA, "Buddhist
Devotional Meditation: A Study of the SukhiivatfvyilhOpade1a," in Minoru
KrYOTA (ed.), Mahayiina Buddhist Meditation: Theory and Practice, Honolulu
1978, pp. 249-96; Jean ERACLE: Aux sources du Bouddhisme Mahayana: Trois
soutras et un traite sur la Terre Pure, Geneva 1984, pp.293-323; Hisao
INAGAKI: Ojoronchil: T'an-luan's Commentary on Vasubandhu's Discourse on
the Pure Land: A Study and Translation, Kyoto 1998.
nABS 23.2 300
in his Vyiikhyii on the Abhidharmakosa, discusses the case taken by the
verb namas-kr, and states anenaiva iiciirye1}a vyiikhyiiyuktau "namas-
krtya muni1[L murdhnii" iti.4 The citation is from line d of tl;te first of the
opening verses of the Vyiikhyiiyukti (32al), thub la spyi bos phyag 'tshal
te. Anenaiva iiciirye1}a, "this same master", refers to Vasubandhu as
author of the Abhidharmakosa, and credits him with the authorship of
the Vyiikhyiiyukti.
Beyond this, Vasubandhu himself refers to the Vyiikhyayukti in his
Karmasiddhiprakara1}a:
5
mam par Mad pa'i rigs pa las kyan I den san mda sde thams cad ni mi ~ n a n ies
bsgrubs teo
It has been proven in the Vyiikhyiiyukti that "the complete [corpus] of siitras is
today no longer extant".
The reference is to Vyiikhyiiyukti 114bl foIl., which deals with "the lost
portions of the canon". While it establishes that V asubandhu knew the
Vyiikhyiiyukti and held it to be authoritative in at least this case, the
reference does not absolutely establish common authorship. However,
the style of the reference is unusual in Vasubandhu's works, and to me
suggests a cross-reference to one of his own works.6 In the Vyakhyiiyukti
itself (l24b4), Vasubandhu refers back to the same subject with similar
phrasing:
sans rgyas kyi gsun mtha' dag ni den san mi snan no ies de skad kyan
bstan zin to. "
That the complete teachings of the Buddha are today no longer extant has already
been explained.
4. Abhidharmakosavyiikhyii I 9,12. References to the Abhidharmakosabhiirya are
to the editions of P.PRADHAN -Abhidharmakosabhii.Jyam of Vasubandhu,
Revised second edition, K.P. Jayaswal Research Institute, Patna 1975 (Tibetan
Sanskrit Works Series VIII) - and Swami DWARIKADAS Shastri, Abhidharma-
kasa & Bhiirya of Acharya Vasubandhu with Sphu!iirthii Commentary of Aciirya
Yasamitra, Bauddha Bharati, 4 vols., Varanasi 1970-73 (Bauddha Bharati Series
5-7, 9). References to the Abhidharmakasavyiikhyii are to Dw ARIKADAS Shastri.
5. Etienne LAMOTTE: "Le Traite de l'Acte de Vasubandhu", Melanges chinois et
bauddhiques, Vol. 4, Brussels, July 1936, p. 201,15 (text), p. 252 (translation).
6. Traditional sources ascribe a certain vanity to Vasubandhu, and it strikes me as
rather rare for him to refer favourably to another siistra - outside of the canon -
as having established anything.
SKILLING 301
Another of Vasubandhu's works, the PratftyasamutpadadivibhaJiga_
nirdesa (23b2), contains a reference to the Abhidharmakosa:
de dag ji ltar rdzas su yod pa ma yin pa ni chos mrion pa'i mdzod las ies par
bya'o.
How [the viprayuktasal'{Lskaras] are not substantial entities (dravya) may be
known from the Abhidharmakoia.
Asariga makes cross-references to sections of his Yogacarabhumi,7 as
does Buddhaghosa to his Visuddhimagga in his A!!hakatha.
8
It is natural
that Vasubandhu - like any prolific writer - should do the same.
Explaining the phrase "listen, and bear well and carefully in mind"
(sP}U sadhu ca s u ~ ! h u ca manasi kuru) - an injunction with which the
Buddha frequently opens a discourse - the Vyakhyayukti (135a3-7) gives
three similes for the "faults of a listener":
(1) an overturned pot: even though it rains, it will not collect water;
(2) a dirty pot water added to it will become polluted;
(3) a leaky pot: it will not hold the water.
The same similes (the first of which occurs at Anguttaranikaya I 130,5
foIl.) are given in the Pratftyasamutpadadivibhanganirdesa, both in its
Tibetan version and in a surviving Sanskrit fragment,9 and in the
Gatharthasa'!!grahasastra.
1O
In his Abhisamayala,!!kartiloka, Hari-
bhadra, commenting on the same phrase, gives the similes, concluding
with eva'!! hi sravar:za,!! saphalam ity acarya vasubandhul:z: "thus is
listening [to the Dharma] fruitful: so says Aciirya Vasubandhu".ll The
7. WAYMAN: "Doctrinal AffIliation," pp.202, 205, and notes 10,24,25.
8. For these references, cf. George D. BOND, The Word of the Buddha, the Tipi!aka
and its Interpretation in Theravada Buddhism, Colombo 1982, pp. 117-18; K.R.
NORMAN: Pali Literature, Wiesbaden 1983, p. 122; Bhikkhu NAt:<AMOLI: The
Path of Purification, Kandy 1975, pp. xx-xxi.
9. P5496, Vol. 104, mdo tshogs 'grel pa, chi, 3b5-8; Sanskrit in TUCCI (see n. 17
below), p. 614,1-6.
10. Gatharthasal'{Lgrahaiastra, P5604, Vo1.119, mnon pa'i bstan bcos, nu,263b6-
264al. For the first simile with two different similes in the Theravadin tradition
see Bruce EVANS (tr.), "Tayopuggala Sutta," in Fragile Palm Leaves: for the
preservation of Buddhist literature 4 (September 2541/1998), pp. 8-9.
11. U. WOGIHARA (ed.), Abhisamayalal'{Lkaraioka Prajiiaparamitavvyakhya the
Work ofHaribhadra, The Toyo Bunko, Tokyo, [1932] 1973, pp.137,26-138,6.
It is not clear from which of Vasubandhu's works Haribhadra has taken the
similes - if he is not paraphrasing - since his citation does not agree exactly with
any of the three works of Vasubandhu that give them.
nABS 23.2 302
similes are given in the Arthav.iniScayasutranibandhana and the Artha-
viniscaya!rka,12 In Tibet they have enjoyed an enduring popularity, from
the time of sGam po pa and Bu ston, both of whom cite them, up to the
present, having become part of the standard instruction on how to listen
to the Dharma.
13
On the basis of the evidence presented above - that of Vasubandhu
himself, of Yasomitra, and of style and thought - I conclude that the
Vyakhyayukti was composed by the Kosakfu'a, and that it preceded his
composition of the Karmasiddhiprakara[La ..
(3) KarmasiddhiprakaraI).a
14
The common authorship of the Karmasiddhiprakara[La and the Abhi-
dharmakosa was accepted by LAMOTTE, who edited the Tibetan and
translated it into French,15 Since the Karmasiddhiprakara[La is posterior
to the Vyakhyayukti, which contains a brilliant defence of the Mahayana,
LAMOTTE's conclusion that the former is not a Mahayana work
(pp. 175-79) cannot be sustained. Internal evidence, of which LAMOTTE'
was aware but chose to interpret differently, also supports its affiliation
with the Mahayana: Vasubandhu's citation of the SaTJ'ldhinirmocanasutra
12. N.H. SAM TAN I (ed.): The Arthaviniscaya-satra & its Commentary
(Nibandhana), K.P. Jayaswal Research Institute, Patna 1971 (Tibetan Sanskrit
Works Series XIII); 83,6-13; Arthaviniscayapkii, P5852, Vol. 145, 10a4-1Ob4.
13. E. OBERMILLER (tr.): History of Buddhism (Chos-hbyung) by Bu-ston, 1. Part,
The Jewelry of Scripture, Heidelberg 1931 (Suzuki Research Foundation Reprint
S e g ~ s 5) (henceforth referred to as Bu ston [OBERMILLER]), p.79; Herbert V.
GUENTHER (tr.): The Jewel Ornament of Liberation, London 1959, p. 36 (cf. n.
25, pp. 38-40).
14. See bibliographical data in NAKAMURA, pp.268-69. French translation by
Etienne LAMOTTE: "Le Traite de l'Acte de Vasubandhu" (see n.5 above);
English translation in ANACKER, op. cit., pp. 83-156; Tibetan edited with Tlkiiby
Ryoshun KAJIHAMA: Karmasiddhiprakara!la by Vasubandhu, Karmasiddhi
Tlkii by Sumatisfla, Sarnath 1988. For the Tibetan edited by MUROn (not seen)
see reference in Marek MEJOR, "On V asubandhu' s Pratftyasamutpiidavyiikhyii,"
Studia Indologiczne 4 (1997) (Aspects of Buddhism: Proceedings of the
International Seminar on Buddhist Studies, Liw, 25 June 1994), Oriental
Institute, Warsaw University, p. 152: "Muroji ". made a comparative edition of
Vasubandhu's Karmasiddhiprakara!la and showed its close relation to the
P[ratItya]s[amutpiida]vy[akhya]".
15. LAMOTTE: "Le Traite de l' Acte", pp. 180-81.
SKILLING 303
as "spoken by the Blessed One",16 his presentation of the Yogacara
classification of the "store consciousness" (iilayavijfiiina, p. 197,8 foIl.),
and (perhaps) his closing verse (p. 204,33) with the dedication of merit
in order that the world may realize Buddhahood. I conclude that the
KarmasiddhiprakaralJa was composed by Vasubandhu as an advocate of
the Mahayana, but addressed primarily to adherents of the Sravakayana.
(4) Prafilyasamutpadadivibhariganirdesa
The Pratftyasamutpiidiidivibhaliganirdda is preserved in Tibetan trans-
lation (P5496, Vol. 104) and in a few Sanskrit fragments from Nepal,
published by Tucci who accepted its common authorship with the
Kosa.J7 Like the Vyiikhyiiyukti and Giithi'irthasaf[lgrahasiistra it depends
on the sutra literature, in this case on the Pratftyasamutpiidiidivibhaliga-
sutra, upon which it is a commentary. Mainly in prose, it also contains
saf[lgrahasloka (8a2) and summarizing verses introduced as giithi'i (17 a2,
18a3, 20a5, etc.). It resembles the other works of Vasubandhu in style
and methodology; it cites or refers to many of the same sUtras, and
contains many of the same ideas or arguments. Like the Vyiikhyiiyukti
and the KarmasiddhiprakaralJa, it refers to the Saf[ldhinirmocanasutra
(25b5); like the KarmasiddhiprakaralJa and the PaficaskandhaprakaralJa
(see following) it accepts the iilayavijfiiina (24b2 foIl.). Its defence of
that concept is couched in similar terms to that of the Karmasiddhi-
prakaralJa. It also refers to the Yogiiciirabhumi (5b2, 13a1, 25b4), in
one place citing it at length on the definition of ignorance (avidyii).18
16. LAMOTTE: "Le Traite de l'Acte", p.198,14, beam Idan 'das kyis ... gsuris so =
uktal'(! bhagavatii.
17. Giuseppe TUCCI: "A Fragment from the PratItyasamutpada-vyakhya of Vasu-
bandhu," Journal of the Royal Asiatic Society, London 1930, pp. 611-23. Two
chapter colophons preserved in Sanskrit (pp.619,5 and 620,24) give as title
Pratftyasamutpiidavyiikhyii. MEJOR: "On Vasubandhu's Pratltyasamutpilda-
vyakhya," gives a comprehensive list of studies, translations, and editions
(pp. 151-52), including Yoshihito G. MUROn, Vasubandhu's Interpretation des
Pratftyasamutpiida: eine kritische Bearbeitung der Pratftyasamutpiidavyiikhyii
(Sal'(!skiira- und Vijfiiinavibhmiga), Stuttgart 1993 (Alt- und Neu-Indische
Studien 43).
18. mal 'byor spyod pa'i sa las: llb3-12b4 = Vidushekhara BHATTACHARYA: The
Yogiiciirabhilmi of Aciirya Asanga, Calcutta 1957, pp.204,2-206,9. For the
passage - and other definitions of avidyii - see Marek MEJOR, "On the
Formulation of the Pratftyasamutpiida: Some Observations from Vasubandhu's
Pratftyasamutpiidavyiikhyii," Studia Indologiczne 4 (1997), pp. 145-46.
nABS 23.2 304
Again like theVyiikhyiiyukti, it was commented upon by GUl).amati (see
below). I have not found any indications of its chronology relative to the
Vyiikhyiiyukti and the KarmasiddhiprakaralJa.
(5) Paiicaskandhaprakaral).a
19
The short prose Treatise on the Five Aggregates (P5560, Vol. 113; =
PSP) deals with the iilayavijfiiina when discussing the vijfiiinaskandha.
Yasomitra cites the treatise four times in his Kosavyiikhyii:
(1) Kosavyiikhyii 1:33ab, p. 89,23-25 (P5593, Vol. 116, 68b2-4) = PSP
16a3-4: tatM hy anena paiicaskandhake uktam: vitarkalJ katamalJ ... ;
(2) Kosavyiikhyii 2:24, p. 187,9-11 (P5593, Vol. 116, 131a3-4) = PSP
14a3-4: tatM hy anenaiviiciiryelJa pancaskandhake likhitam: chandalJ
katamalJ .. ;
(3) Kosavyiikhyii 3:32b, p. 477,25-478,14 (P5593, VoL 116, 346a7-
346b3) = PSP 14a3-5: yatM pancaskandhake likhitarrt ... chandalJ
katamalJ '" ity evam iidilJ pancaskandhakagrantho
(4) Kosavyiikhyii 3:32b, p. 478,14-16 (P5593, Vol. 116, 346b3-4) =
PSP 14a1: tatra hy uktam: pafica sarvatragii . ...
In the fIrst two references, Yasomitra attributes the work to the Kosa-
kara (anena ... uktam; anenaiviiciiryelJa likhitam).
Three commentaries on the PancaskandhaprakaralJa - by Sthiramati
(P5567, Vol. 114), GUl).aprabha (P5568, Vol. 114), and PrthivIbandhu
(P5569, Vol. 114) - are preserved in Tibetan translation. At the begin-
ning of his commentary (sems tsam, hi, 67b2), GUl).aprabha states that
one should open a treatise (Siistra) by paying respect (mchod par brjod
pa) to one's teacher [the Buddha]. There is, however, no such statement
in the PancaskandhaprakaralJa. GUl).aprabha explains the omission:
mchod par brjod pa ni mdzod kyi bstan bcos (corr. from bstan chos) la sogs pa
las smos pa de fiid 'dir blta bar bya ste, bsam pa des na slob dpon gyis 'dir
mchod par brjod pa ma byas so.
[Since there is no such opening in the Paficaskandhaprakara"(la,] the statement of
homage given in the Kosasiistra (mdzod kyi bstan bcos), etc., should be supplied
here: with this in mind, the Acarya [Vasubandhu] did not compose [a new]
statement of homage here [in the Paficaskandhaprakara"(la].
19. See bibliographical data in NAKAMURA, p. 270. Tibetan and Chinese versions
with French translation in D ANTINNE, op. cit.; English translation in ANACKER,
op. cit., pp. 51-82.
SKILLING 305
GUlJaprabha cites ktirikiis from the Kosa as authoritative (67b7 = ktirikii
1:3ab; 93b8 = ktirikti 1:17ab), cites sutras as cited in the Kosa, and para-
phrases the
Pp:hivIbandhu also deals with the subject of homage at the beginning
of his commentary (sems tsam hi, 101b2):
de la gtsug lag 'di yan mnon pa'i chos mdzod la 'jug pa'i sgo Ita hur gyur pa
dan de'i yan lag tu gtogs la, bstod pa'i cho ga ni mnon pa'i chos mdzod kyi
mgor glens zin pas, der 'dus pa'i phyir 'dir logs iig tu ma Mad do.
This treatise [the PaiicaskandhaprakaralJa] is like a gateway to the Abhidharma-
koia and is counted as a supplement to or part of (yan lag = anga) that work.
The rite of homage has been spoken at the beginning of the Abhidharmakoia;
since the former [the PaiicaskandhaprakaralJa] is included in the latter [the
Abhidharmakoia], it is not given separately.
P[1:hivThandhu states that the Kosa is addressed to one who understands a
developed exposition (vipaficitajfia), the PaficaskandhaprakaraJ}a to one
who understands a condensed exposition (udghatitajfia) (102a3):
de la spros pas ies pa ni giun fib tu Mad na don rtogs par 'gyur ba ste; de dag
gi don du ni mnon pa'i chos mdzod la sogs pa yan dgos par 'gyur roo glens pas
ies pa ni mdo tsam du bstan na don man du khon du chud par 'gyur ba ste; de'i
phyir gtsug lag 'di brtsams pa don med par ma yin teo
Thus both GUlJaprabha and f>rthivIbandhu, like Yasomitra, attribute the
Kosa and the PaficaskandhaprakaraJ}a to the same Vasubandhu, and
treat the latter as a natural continuation of the former. LA V ALLEE
POUSSIN's notes show other places in which alternate definitions pro-
posed by the Kosakara reappear in the Paficaskandhaka.
20
DANTINNE
(op. cit., pp. xxiii-xxiv) has noted the relationship between the Pafica-
skandhaprakaraJ}a and the Abhidharmakosa, Asariga's Abhidharma-
samuccaya, and the Trirrzsikti-ktiriktis.
(6) GathasaI1J.grahasastra, (7) GatharthasaI1J.grahasastra, (8) Ekagatha-

What we may call the "Gtithtisarrzgraha literature" is similar to the
"Vytikhytiyukti literature" in that it consists (primarily) of two inter-
20. See references at Louis DE LA V ALLEE POUSSIN, L 'Abhidharmakosa de
Vasubandhu, Vol. VI, reprint: Brussels 1971, p.121.
21. The correct Sanskrit form of the first two titles is uncertain: see Peter SKILLING,
"The Synonyms of Nirval].a according to Prajfiavarman, Vasubandhu, and
Asanga," Buddhist Studies Review 11.1 (1994), p. 34, n. 19.
nABS 23.2 306
dependent works: the GlithiisaTflgrahaslistra (P5603)22 - a collection of
21 verses drawn from canonical Sravakayana texts such as the Udlina-
varga - and the GlithlirthasaTJ1-grahaslistra (P5604), a prose commentary
on the verses.
23
The latter cites, without naming them, the Abhidharma-
kosa (Klirikli and cites Siitra passages identical to those of the
"Hundred Extracts", and contains several passages common to the
Vylikhyliyukti and the PratrtyasamutplidlidivibhariganirdeSa. It also
adopts a number of explanations from works of Asa:Iiga such as the
Abhidharmasamuceaya. The definition of Vaipulya in the Glithiirtha-
saTJ1-grahaslistra (246a3), unlike that in the Vylikhyliyukti, does. not
mention the Mahayana:
24
sin tu rgyas pa'i sde ni don rgyas par stan pa'o. gian dag ni gian dag tu brjod
de, 'phags pa dge 'dun phal chen sde pa mams kyi 'don pa kho na Ita bu' a ies so.
The Vaipulya explains a topic at length. Others define it differently as "[scriptures]
recited by the Arya Mahiisiirp.ghikas".
The GlithlirthasaTJ1-grahaslistra does not cite or refer to any Mahayana .
texts or contain any uniquely Mahayana ideas, but it does resort to
Asailga's Abhidharmasamuceaya as a major source. Until a complete
study has been made, I would tentatively describe it as anterior to the
Vylikhyliyukti. At any rate, since it relies heavily on the Abhidharma-
kosa, it cannot be by the "elder V asubandhu" .
The first verse of the GlithiisaTJ1-grahaslistra (240a2-3), along with its
commentary from the GlithiirthaSClTJ1-grahaslistra (241a7-242b4), occurs
as a work in the Tanjur under the title (Tshigs
su bead pa gcig pa'i bsad pa, P5488, Vol. 104), also ascribed to
Vasubandhu. The may therefore be classed with the
22. Tibetan text with German translation by A. SCHlEFNER, in Melanges asiatiques
tires du bulletin de l'Academie Imperiale des Sciences de St.-Petersbourg, Vol.
VIll, April 1878, pp. 559-93; English translation by W. Woodville ROCKHlLL as
an appendix to his Udlinavarga: A Collection of Verses from the Buddhist
Canon, London 1883, pp. 213-16.
23. For identification of the verses see Fumio ENOMOTO, "Sanskrit Fragments
Relating to the SaIpyuktagama" [in Japanese], Bukkyjj Kenkya [Buddhist Studies]
15 (Hamamatsu 1985), pp. 81-93, especially pp. 91 and 93, note 13 and below,
Appendix 7.
24. For Vaipulya see Peter SKILLING, Mahlisatras: Great Discourses of the Buddha,
Vol. n, Parts I & n, Oxford 1997, pp. 31-42.
SKILLING 307
"Giithiisaf[lgraha literature". The verse, from the Virupa-avadiina of the
A vadiinasataka, 25 is also cited in the
A commentary, the Ekagiithiitfkii (P2063, Vol. 46) is ascribed to
Dignaga, who aJso composed a brief commentary on the Abhidharma-
kosa, the Marmapradfpa. Dignaga's authorship of the latter and his
connection with Vasubandhu has been questioned by some scholars; the
Ekagiithiitfkii may be added to evidence in favour of the tradition.
I hope to have demonstrated so far that these six works were composed
by Vasubandhu the Kosakara. From these works we can extract doctrinal
criteria that can be used to determine the authenticity of other works
attributed to Vasubandhu. Three of them expound the theory of the
iilayavijfiiina; three of them cite or refer to the Saf[ldhinirmocanasutra as
authoritative. Since that siitra is the main source for the theory of the
three svabhiivas, we may assume that Vasubandhu accepted that theory.
Works traditionally attributed to Vasubandhu that deal with these
subjects in a style similar to that of the above-mentioned works can be
accepted as by the Kosakara.
(9) ViIp.satikakarika
26
and (10)
Like the Vyiikhyiiyukti, the Pratftyasamutpiidiidivibhmiganirdea, and
the Karmasiddhiprakarar;a, this is a Mahayana work addressed primarily
to adherents of the Sravakayana. The prose auto commentary resembles
25. P.L. VAIDYA (ed.): AvadanasatakaY[!, Darbhanga 1958 (Buddhist Sanskrit Texts
Series 19), p. 251,18-21; citation in AbhidharmakosabhafYa 4:112, (PRADHAN)
267,13-16, (DWARIKADAS) 738,20.
26. See bibliographical data in NAKAMURA, pp. 268-69. Also: Sanskrit and Tibetan
ed. in N. AIYASW AMI Sastrin, op. cit.; Sanskrit with Hindi translation in Mahesh
TIWARY, Vijftaptimatratasiddhi of Vasubandhu, Varanasi 1967; Sanskrit and
Tibetan with Hindi translation in Ramsarikara TRIPATHI and Sempa DORJE:
VijftaptimatratasiddhiJ:t (PrakaraIJa-dvayam) of Acarya Vasubandhu: Viy[!satika
with an Auto-Commentary, TriY[!sika with the Commentary of Sthiramati, Leh
(Ladakh) 1984; Sanskrit with English translation in Thomas E. WOOD: Mind
Only.' A Philosophical and Doctrinal Analysis of the Vijftanavada, University of
Hawaii Press, Honolulu 1991 (Monographs of the Society for Asian and
Comparative Philosophy, no. 9), Chap. 5; English translations in
KOCHUMUTTOM,Op. cit., pp. 260-75; in ANACKER, op. cit., pp.157-79; in
David J. KALUPAHANA, The Principles of Buddhist Psychology, Albany 1987,
pp. 173-92; translated from the Chinese under the tide The Treatise in Twenty
Verses on Consciousness Only, in Francis H. COOK, Three Texts on Conscious-
ness Only, Numata Center for Buddhist Translation and Research, Berkeley 1999
(BDK English Tripi!aka 60-Ill).
JIABS 23.2 308
these works stylistically, and its arguments are based on the doctrines
and lore of the Sravakayana, from the canon of which a number of
siitras are cited. AIY ASW AMI Sastrin (pp. iii-v) has- noted the
relationship to the Kosa, particularly the "debate" on the existence of the
atom.
(11) Trirp.sika-karikas
27
This work deals with the iilayavijfiiina and the three svabhiivas, and
gives some of the classifications of the prose PaficaskandhaprakaralJa in
verse. At times the relationship to that text is so close that the two.could
easily be combined: the verses of the Trirrzsikii followed by the prose of
the PaficaskandhaprakaralJa.
(12) Trisvabhavanirdesa
28
I have noted that Vasubandhu accepted the Sarrzdhinirmocanasutra, the
main source for the "three svabhiiva theory", as authentic, and that the
three svabhiiva are mentioned in the Trirrzsikii-kiirikiis. P.S. JAINI has
shown that the author of the on the Abhidharmadfpa
accuses the Kosakara of accepting and, according to Jaini's interpreta-
tion, writing on this theory.
27. See bibliographical data in NAKAMURA, pp. 269-70. Also: Sanskrit with Hindi
translation in Mahesh TIw AR Y, op. cit.; Sanskrit and Tibetan with Hindi
translation in TRIPATHI and Sempa DORJE, op. cit.; Padmanabh S. JAINI: "The
Sanskrit Fragments of VinItadeva's Trir!!Sika-[fka," Bulletin of the School of
Oriental and African Studies, London 1985, pp.470-92; Sanskrit with English
translation in Thomas E. WOOD, op. cit., Chap. 3; English translations in
op. cit., pp.254-59; in ANACKER, op. cit., pp.181-90; in
KALUPAHANA, op. cit., pp.192-214; translated from the Chinese under the title
The Thirty Verses on Consciousness Only, in Francis H. COOK, op. cit. (= BDK
English Tripitaka 60-II). See also Louis DE LA VALLEE POUSSIN' s monumental
translation of Hsiian-tsang's monumental compendium, Vijfiaptimatratasiddhi, la
Siddhi de Hiuan-tsang, 2 Vols., Paris 1928, 1929; English translation with
Chinese text by Wei TAT: Ch'eng wei-shih [un, The Doctrine of Mere-
consciousness, Hong Kong 1976; English translation from the Chinese under the
title Demonstration of Consciousness Only, in Francis H. COOK, op. cit. (= BDK
English Tripitaka 60-1).
28. See bibliographical data in NAKAMURA, p.271. Sanskrit text with English
translation in Fernando TOLA and Carmen DRAGONETTI: "The Trisvabhava-
klirika of Vasubandhu," Journal of Indian Philosophy 1113 (Sept. 1983),
pp.225-66 and in Thomas E. WOOD, op. cit., Chap. 2; English translations in
KOCHUMUTTOM,op. cit., pp. 247-53; in ANACKER, op. cit., pp. 287-97.
SKILLING 309
Lambert SCHMITHAUSEN reached a similar conclusion about the
authorship of most of the works discussed above: "I use the expression
'(works of) Vasubandhu the Kosakara' when referring to (the author of)
Vyiikhyiiyukti, Karmasiddhi, Pratftyasamut-
piidavyiikhyii, Pancaskandhaka, Vi1J'lsatikii and Tri1J'lsikii, because I find
that there are good reasons for taking the author of these works
to be one and the same person".29 Bhikkhu PAsADIKA remarks that
"Schmithausen does not enter into details regarding the V asubandhu
controversy but refers to publications by Kazunobu Matsuda and Gijin
Muroji through whose observations he finds his own conclusions
corroborated". PASADIKA also refers to similar conclusions reached by
Christian LINDTNER: "Lindtner speaks of V asubandhu as 'the author of
the Vi1J'lsatikii, the Tri1J'lsikii, Karmasiddhi, Abhidharmakosa, and
Vyiikhyiiyukti, etc. "'.30
Accepting a common authorship of the works discussed above, and
accepting the tradition that the Kosa was Vasubandhu's first work, we
may suggest the following tentative chronology, dividing his activity
into two main perspectives:
31
"Sautrantika perspective"
Abhidharmakosa (Kiirikii and BhiirYa)
Giithiisal!lgrahasiistra and Giithiirthasal!lgrahasiistra [plus EkagiithiibhiirYa]
29. Bhikkhu PAsADIKA, op. cit., p.16, citing Lambert SCHMITHAUSEN: Alaya-
vijfliina, On the Origin and the Early Development of a Central Concept of
Yogiiciira Philosophy, Tokyo 1987, p. 262, n. 101.
30. Bhikkhu PAsADIKA, op. cit., pp.16-17, citing Chr. LINDTNER: "Marginalia to
Dharmaklrti's Pramagaviniscaya I-II," Wiener Zeitschrift fur die Kunde
Sudasiens XXVIII (1984), p. 154 (not seen).
31. I must emphasize that this approach is provisional. Vasubandhu' s presentation of
Mahayana ideas by no means entails a rejection of Sautrantika thought, or its
wholesale replacement by some sort of ready-made Mahayana package. It is quite
possible that Vasubandhu's profound study of the Sravaka scriptures led him
directly to the ideas expressed in the Vyiikhyiiyukti. The oft-repeated story of
Vasubandhu's conversion by his brother Asanga is given only in later
biographies. There is no reason why Vasubandhu should not have composed the
Kosa as a Mahayanist, without advocating or mentioning the Mahayana since this
would not be relevant. Indeed, Robert KRITzER, in his Rebirth and Causation in
the Yogiiciira Abhidharma (Vienna 1999, p. viii) writes that Vasubandhu's
"'Sautrantika' opinions ... can generally be traced to the Yvgiiciirabhami,"
leading him to suspect "that Vasubandhu already accepted Y ogacara doctrine
when he wrote the AbhidharmakosabhiirYa".
JIABS 23.2 310
"Mahayana perspective"
Vyakhyayukti and Vyakhyayuktisutrakhaftasata
Kannasiddhiprakarafta
Pratftyasamutpadadivibhariganirdeia
Paficaskandhaprakarafta
Virrzsatikli, Karikli and
Trirrzsika-karikas
Trisvabhavanirdeia.
From Yasomitra's reference to the Kosakara as the author of the
Mahayanist Vyiikhyiiyukti and of the Pancaskandhaprakararza, and from
Vasubandhu's cross-reference to the Vyiikhyiiyukti in his Karmasiddhi-
prakararza we may suggest that Vasubandhu the Kosakara became an
energetic proponent of the Mahayana. This suggestion is corroborated by
the unknown author of the Abhidharmadfpa, who severely chides the
Kosakara for his sympathy with the Mahayana. Furthermore, the
Vyiikhyiiyukti, the Karmasiddhiprakararza, and the Pancaskandhapra-
kararza show the influence of the Y ogacara school and of Asanga, as
does the Giithiirthasaf[Lgrahasiistra, while the Vif[Lsatikii, the Trif[Lsikii,
and the Trisvabhiivanirdesa are devoted to the exposition of specific
Vijfiaptivadin doctrines.
We may now attempt to sketch a broad outline of the Kosakara's
career.
32
Vasubandhu mastered the scriptures and doctrines, the
Tripi!aka and Abhidharma, of the (MUla)Sarvastivadin school, specifi-
cally according to the tradition of the of Kashmir, and also
studied the Tripitaka and Abhidharma of other schools. The result of
these studies was his Abhidharmakosa, which summarized the tenets of
the of Kashmir in its kiirikiis, but subjected them to a
critique in its prose In this work he shows sympathy for the
thought of the Sautrantikas, but also maintains the intellectual indepen-
dence which would be a hallmark of his work. At an uncertain point in
his career - before or after the composition of the Kosa? - he was
deeply influenced by Yogacara thought as exemplified by Asanga and
the Saf[Ldhinirmocanasutra. He retained, however, his independent and
critical attitude, and retained or refined many of the ideas and arguments
already presented in his Abhidharmakosa. He was prolific, and com-
posed works on various subjects.
32. This is provisional, since it does not take into accout some of the important works
attributed to Vasubandhu.
SKILLING 311
This career, based entirely upon (originally) Indian literary sources,
either those of the Kosakara himself or those of his commentators,
agrees in outline with that given in Paramartha's biography of Vasu-
baridhu. I would like to stress that these commentators worked within
the same traditi'on and lineage of thought, and were connected with
Vasubandhu if not directly then by succession. We find the same com-
mentators, connected with each other by tradition, associated with texts
attributed to Vasubandhu. Their testimony, which goes back to an Indian
tradition in the direct lineage of succession of disciples of Vasubandhu,
cannot be lightly dismissed.
I will not attempt a detailed refutation of FRAUWALLNER's conclu-
sions, which are based primarily on Chinese sources, especially the life
of Vasubandhu by Paramartha, whom he describes as the authority "who
carries the greater weight by far", "the chief witness" (both on p. 11),
and "our oldest and best authority (p. 13)". My own evidence is based
rather on the works of Vasubandhu himself, and those of the commenta-
tors within his tradition.
33
FRAUW ALLNER's "revised" biography of the
second, younger or later Vasubandhu contains only one salient literary
event: his composition of the Abhidharmakosa. Since the evidence
presented above proves that the Kosakara did indeed advocate the Maha-
yana, it undermines the greater part of FRAUWALLNER's arguments.
The career outlined above does the reverse of what FRAUWALLNER set
out to do: it identifies Vasubandhu the Kosakara with Vasubandhu the
Mahayanist, and thereby robs FRAUWALLNER's earlier Vasubandhu of a
biography.
This said, the problems of chronology remain. FRAUWALLNER's
treatment of the dates of the two Vasubandhus is rather shaky, since it is
based on his own conclusions about the date of the Buddha employed by
33. I fmd it extraordinary that FRAUWALLNER - who appeals to "the rules of sound
criticism" (op. cit., p. 30) and "scientific logic" (p. 31) - should reach such firm
conclusions without taking into account many of the major works attributed to
Vasubandhu, or giving any serious consideration to his thought. He creates more
confusion than that he claims to have been created by his sources, which he treats
in a cavalier manner. He repeatedly describes Paramartha as trustworthy, but, it
turns out, only for the date of Vasubandhu: otherwise he chops Paramartha's
biography up into parts, and ascribes anything that goes against his thesis to
Paramartha's pupils (pp. 18-20). I do not think FRAUWALLNER's monograph is a
good starting point for the study of Vasubandhu, and hope that the Chinese
evidence will be assessed anew, in the light of the thought ofVasubandhu based
on a thorough examination of the works attributed to him.
JIABS 23.2 312
his Chinese sources. I confess here to a pessismitic conclusion that the
dates given in the Chinese sources utilized by FRAUw ALLNER are either
hopelessly confused, or, even if accurate, they cannot be resolved
because of uncertainty about the date of the NirvaI).a upon which they
are variously based.
34
Beyond this, there are Yasomitra's references to an earlier Vasu-
bandhu, which I cannot dismiss as lightly as has Stefan ANACKER.3
5
In
addition to these well-known references, there is one in Sthiramati's
commentary on the Kosa:
36
slob dpon dbyig gfien sna rna = purviiciirya
Vasubandhu, parallel to the vrddhiiciirya Vasubandhu at Yasornitra
1:13d, p. 45, bottom.
37
This reference is not a simple borrowing from
Yasomitra, since Sthiramati gives a direct citation in several lines of the
view of this "earlier V asubandhu", not given by Yasomitra. The
references deal with points of interpretation of the Abhidharma. All that
can be said is that there was indeed an earlier Vasubandhu, who worked
within the (MUla)Sarvastivadin or tradition. Considering the
lack of references to him elsewhere, he seems to have been a relatively
minor figure in the history of Indian Buddhist thought. There is no
reason whatsoever to suppose that he turned later to the Mahayana, as
did (according to the biographies) the Kosakara.
The Kosakara was clearly influenced by the Y ogacara and by AsaIiga.
I do not care to conclude whether or not he was the latter's brother, or
half-brother, but simply note that both Chinese and Tibetan tradition,
34. There seems to me to be a serious flaw in FRAUWALLNER's conclusion (p.lO)
that the earlier date belongs to the earlier Vasubandhu, brother of Asanga and a
person other than the Kosakara. The "elder Vasubandhu" mentioned in the
commentaries on the Kasa (and, it seems, nowhere else) does not appear to have
been sufficiently well-known for his date or biography to have enjoyed a "wide
diffusion". The points attributed to this "elder Vasubandhu" are fme points within
the Abhidharma tradition of, presumably, the V and have nothing to
do with the Mahayana. Thus there is no evidence to connect this elder
Vasubandhu with the Mahayana or with the story of Asanga's brother who
"converted" to the Mahayana.
35. Op. cit., pp. xii-xiii.
36. P5875, Vol. 146, lia mtshar bstan
beas, ta, 7Ial.
37. The Tibetan ofYasomitra (P5593, Vol. 116, mdzad 'grel, eu, 37a5, also has slia
ma, as at 39a7. I have not been able to trace parallels to Yasomitra's other
references in Sthiramati. For the references see AIYASWAMI Sastrin, ap. cit.,
p. vi.
SKILLING 313
based on different sources from different periods, agree that the two
great scholars were related. A tradition current in Tibet at the time of
Taranatha,38 but rejected by him, was that the two were "spiritual
brothers". Given the lack of evidence it seems to me more fruitful to
attempt to determine the relationship between the known works of
Asanga and the known works of the Kosakara than to worry about their
genealogy.
A note on GU1}amati
LA V ALLEE POUSSIN notes that Yasomitra refers to GUl:).amati four
times in his Vyiikhyii on the Abhidharmakosa.
39
The first reference
occurs in Yasomitra's opening verses, wherein he states that GUl:).amati
and Vasumitra had composed commentaries on the Kosa. In the
remaining three references Yasomitra registers his disagreement with
GUl:).amati's interpretations; in two of them he refers to a Vasumitra as a
student ( s i ~ y a ) of GUl:).amati.
4o
Taranatha describes GUl:).amati as a
disciple of GUl:).aprabha and "a master of the Abhidharma".41 Tibetan
tradition describes GUl:).aprabha as a direct disciple of Vasubandhu.42
This suggests a tentative succession:
Vasubandhu - GUIfaprabha - GUIfamati - Vasumitra - (Yasomitra).
GUl:).amati's commentary on the Koa has not been preserved. His
commentary on another of Vasubandhu's works, the Pratftyasamutpiidii-
divibhanganirdeia (P5496), is preserved ill Tibetan translation under the
title Pratftyasamutpiidiidivibhanganirdeia!fkii (P5497). He is also said to
have composed a commentary, no longer extant, on Nagarjuna's
MuZamadhyamakakiirikiis.
43
Hajime NAKAMURA suggests a date of
38. Antonius SCHIEFNER: Taranathae de Doctrinae Buddhicae in India
Propagatione, St. Petersburg 1868, p.92,19; translation in Debiprasad
CHATTOPADHYAYA (ed.): Taranatha's History of Buddhism in India, Calcutta
1980, p. 167.
39. L'Abhidharmakosa de Vasubandhu, Vol. 1, repr. Brussels 1971, p. xix.
40. The references are at DWARIKADAS, Vol. 1, p. 3,18; 9,9, and VoL 2, p. 406,18.
41. SCHIEFNER,op. cit., p. 122,18: yon tan 'od kyi slob ma mlion pa pa yon tan blo
gros; CHA'ITOPADHYAYA, op. cit., p. 210.
42. Bu ston (OBERMILLER) II 160; Tiiraniitha in SCHIEFNER, op. cit., p. 100,10,
CHA'ITOPADHYAYA, op. cit., p. 179.
43. David SEYFORT RUEGG: The Literature of the Madhyamaka School of
Philosophy in India, Wiesbaden 1981, pp.49, 62, 112.
nABS 23.2 314
circa CE 420-500 for GUl).amati, and states that he was a contemporary
of Dignaga and that Sthiramati was his disciple.
44
II. A Survey of the Vyakhyayukti Literature
What I describe as the "Vyakhyayukti literature" consists of three texts,
two by V asubandhu and one by GUl).amati. Originally composed. in
Sanskrit, they survive only in Tibetan translations made about 800 CEo
They are preserved in the "Mind Only" (sems tsam) division of the
Tanjur, the collection of treatises and commentaries translated into
Tibetan:
45
(1) Vyakhyayuktisiitrakhw}rjasata (VyY-sii), rNam par Mad pa'i rigs
pa'i mdo sde'i dum bu brgya: One Hundred Extracts from the
Discourses for the Principles of Exegesis by Vasubandhu (dByig gfien),
translated by Visuddhasirp.ha, Sarvajfiadeva, and
Mafijusrlvarman.
(2) Vyakhyayukti (VyY): rNam par Mad pa'i rigs pa: The Principles of
Exegesis by Vasubandhu (dByig gfien), translated by Visuddhasirp.ha,
Sakyasirp.ha, and
(3) Vyakhyayuktitika (VyY-t): rNam par Mad pa'i rigs pa'i Mad pa:
Commentary on the Principles of Exegesis by GUl).amati (Yon tan blo
gros), translated by Visuddhasirp.ha, Sakyasirp.ha, and
Structure: the Vyakhyayuktisiitrakhal).t;lasata
The Vyakhyayuktisiitrakha1}rjasata is a collection of extracts or partial
citations of siitras, of varying brevity. They are presented without any
introduction, classification, or comment, separated simply by ies bya ba
daft = * iti ca, and without any indication of source or title. As is often
the case in Indian literature, the "Hundred" of the title is a conventional
round number: counting each passage separated by the phrase ies bya ba
daft as one extract, there are one hundred and nine.
46
In the Vyakhya-
yukti, Vasubandhu's initial citation of an extract is usually followed by
44. Indian Buddhism, A Survey with Bibliographical Notes, Osaka 1980, p. 280.
45. Bibliographical notices are given in Appendix 8.
46. This is the figure that I have arrived at. The extract numbers (signified by ) in
the present article are my own: they are not assigned numbers in the text. Others
might enumerate the extracts differently. The exact number of extracts, and
various other points about the three Vyilkhyilyukti texts, can only be determined
when critical editions are prepared.
SKILLING 315
the phrase ies bya ba ni mdo sde'i dum bu'o:.: *iti sutrakhm:u;lalJ, or a
variant thereof.47 The term sutrakha1}t;la does not occur in the Abhi-
dharmakosa, but occurs in another work attributed to Vasubandhu, the
Gatharthasmrzgrahasastra,48 and in GUf.1amati's '[fkii on Vasubandhu's
Pratftyasamutpadadivibhmiganirdesa.
49
The Sanskrit is attested in the
50
It will be seen below that the "Hundred Extracts" serves simply as a
collection of sources for the Vyakhyayukti itself. It is not an independent
work.
Structure: the Vyakhyayukti and its 1lka
The Vyakhyayukti, and, following it, the Fka, is divided into five
untitled chapters (le'ur bcas pa):
Chapter 1: VyY (P) 32a1; (D) 29a2
Chapter 2: VyY (P) 45b4; (D) 40a7
Chapter 3: VyY (P) 98a7; (D) 83b4
Chapter 4: VyY (P) 113a8; (D) 96b7
Chapter 5: VyY (P) 133a4; (D) 114a7
Style: the Vyakhyayuktisutrakhaf.1<;lasata
VyY-t (P) 1a1; (D) 139b1
VYY-t (P) 19b7; (D) 155a6
VyY-t (P) 125b3; (D) 248a1
VYY-t (P) 146b4; (D) 265a3
VyY-t (P) 164a1; (D) 278a7
The "Hundred Citations" are all in "canonical" prose, with the exception
of 29 and 39, which are in verse. Most of the extracts have parallels
in the Pali canon.
Extract 29 (21a7), on the subject of suffering (dulJkha), is close to a
verse cited in the (6:3, Pradhan p. 331,20) and
the Arthaviniscayasutranibandhana (Samtani, p. 127,9): sarrzskaranitya-
tarrz jiiatva, atho vipari1}amatam, and so on, with a variant in line a (' du
byed sdug bsnal iiid dan ni = sarrzskaradulJkhatarrz). Extract 39 (22a2),
a verse of six lines listing conditions connected with concentration
47. Several extracts from the VyY-sii are not, however, called siUrakhm;u;la in the
VyY: 19, VyY 55b5; 29, VyY 57b8; 93, VyY 88b2. I doubt whether this
affects their status, or the total number of extracts: 1 (32a3), for example, is not
called a sutrakhalJ{ia when first introduced, but is later referred to as "the first
sutrakhalJ{ia" (36a3). At any rate, more than one hundred are described as
sutrakhalJ{ia in the Vy Y.
48. P5604, Vol. 119, mflOn pa'i bstan bcos, nu, 255a5.
49. P5496, Vol. 104, mdo tshogs 'grel pa, chi, 106b4.
50. Nathrnal TATIA (ed.): Kashi Prasad Jayaswal
Research Institute, Patna 1976 (Tibetan Sanskrit Works Series 17), p. 142,26.
JIABS 23.2 316
(samiidhi) and calming (samatha), is described at Vyiikhyiiyukti 61b4 as
a "verse summary" (sdom gyi tshigs su bead pa = uddiinagiithii). No
accompanying prose is given in either the "Hundred Extracts" or the
Vyiikhyiiyukti. Another four line uddiina (sdom) occurs at the end of the
longest citation, 66, on the benefits of "the notion of impermanence"
(anityasa1pjfiii, 25a4): in this case it is to be counted with the preceding
prose. Such "verse summaries" were composed by the editors of the
Tripi!aka, originally during the period of oral transmission, although
there is no guarantee that they were not altered or added to by the
editors of later written recensions. The present examples most pro",?ably
belong to either the Sa7J1yuktiigama or the Ekottariigama of - as will be
seen below - the (MUla)Sarvastivadins. (A single line of another ud-
diina, possibly from the Sa7J1yuktiigama, is cited at Vyiikhyiiyukti 41a7.
A complete uddiina is cited at Abhidharmakosa 3:12d, p. 122,23.)51
Most of the extracts are spoken by the Buddha to the monks in
general, or, in a few cases, to specific monks, or to brahmans and
"wanderers". Two extracts from the Arthavistarasutra ( 67, 25a5; 68,
25a8) are spoken by Sariputra to the monks. A number of extracts are
addressed to the Buddha, such as 4 (19b!), which is most probably
spoken by King Prasenajit, as is a similar passage in the Pa1i
Aligulimiilasutta (see Appendix 1, extract [4]).
Style: the Vyakhyayukti
The Vyiikhyiiyukti is mainly in "siistra prose".52 Apart from citations
from canonical verse texts, such as the Udiinavarga, which is frequently
cited under the name giithii (see Appendix 5.1), the Vyiikhyiiyukti
contains many verses in a non-canonical style. These may be classed
under three headings:
51. Wrongly given in the printed editions as udiinagiitha: both the Tibetan and the
style of the verse confIrm that the correct form is uddiina-. For the source, cf.
Bhikkhu PAsADIKA, Kanonische Zitate im des Vasu-
bandhu, Gottingen 1989, p. 53, 168.
52. That is to say the original would have been in a form of "standard" Sanskrit,
similar in style to treatises (siistra) such as Vasubandhu's Abhidharmakosa-
or VlryasrIdatta's Arthaviniscayasutranibandhana (which, it will be
shown below, is heavily indebted to the Vyiikhyiiyukti and other works of Vasu-
bandhu).
SKILLING 317
(1) Terse "verse summaries" (bsdus pa'i tshigs su bead pa = sa1J2graha-
sloka), found in all chapters.
53
These are listed in Appendix 2.1. In most
cases the sa1J2grahasloka follow the prose treatment of the subject that
they summarize,
(2) Verses introduced by phrases like "furthermore, in this context, it is
said", listed in Appendix 2.2. In some cases (e.g. 33b5, 34al) these
verses follow the prose that they summarize, and thus seem to be identi-
cal in function to the sa1J2grahasloka. In other cases (e.g. 48a8, 50a7,
l04a5) they introduce independent ideas, without any accompanying
prose.
(3) Verses without any introduction, listed in Appendix 2.3. There are
many such verses in Chapter 5.
The authorship of these verses and their relationship to the work as a
whole remains to be determined. Since Vasubandhu is well-known as an
accomplished verse writer, it is possible that he composed at least some
of the verses of the Vyiikhyiiyukti. In addition, verses in a "non-canoni-
cal" style are cited as Agama. These will be discussed below.
Style: the VyakhyayuktipJ.di
GUI).amati's prose Tlkii clarifies the root-text by paraphrasing Vasu-
bandhu's ideas and by expanding his often terse statements. In style it
resembles GUI).amati's '[fkii on Vasubandhu's Pratftyasamutpiidiidi-
vibhanganirdesa.
Contents of the Vyakhyayukti
The Vyiikhyiiyukti opens with two verses (32al-2): the first pays homage
to the Buddha, while the second announces that the work has been
composed for the benefit of "those who wish to explain the siitras" (mdo
rnams 'chad 'dod). The phrase mdo rnams 'chad 'dod of the verse
occurs in prose as mdo rnam par 'chad par 'dod pa at 32a2; the Sanskrit
53. For the rather complex problem of the sal'[lgrahasloka, see Katsumi MIMAKI,
"Sur Ie role de l' antarasloka ou du sal'[lgrahasloka", in Indianisme et
Bouddhisme, Melanges offerts a Mgr Etienne Lamotte, Louvain-Ia-Neuve 1980,
pp. 233-44. The use of the sal'[lgrahasloka is characteristic of Vasubandhu's
works. MIMAKI (p. 233) notes two in the and the
present article shows that they occur in the Vyiikhyiiyukti and the Pratftya-
samutpiidiidivibhaliganirdeia.
JIABS 23.2 318
is probably sutraTJ'l vylikhyatukama, as at Arthaviniscayasutranibandhana
72,1. The phrase mdo mam par bsad pa occurs at 33b4.
Vasubandhu lists the main topics that he will deal with nea:r the
beginning of his work (33b4):
How are the sutras to be explained? By five methods:
(1) according to the purpose of the sutra (mdo sde'i dgos pa = sutriinta-
prayojana), defmed and discussed in Chapter 1, 34b3-36b5;
(2) according to the summarized meaning (bsdus pa'i don = pi!liirtha), defined
and discussed in Chapter 1, 36b5-37a2;
(3) according to the sense of the words (tshig gi don = padiirtha), defined and
discussed Chapter 1, 37a2-45b3 (end), in Chapter 2, throughout (45b3-98a6),
and in Chapter 3, 98a7-98b8;
(4) according to sequence or connection (mtshams sbyar ba = anusal!!dhi),
defined and discussed in Chapter 3, 99a1-100b3;
(5) according to objections and responses [to objections] (brgal ba dan Ian =
codyaparihiira), defined and discussed in Chapter 3, lOOb3 foll.
Vasubandhu notes that the fifth topic is to be counted as one since it
means "the resolution of objections" (brgaZ ba'i Zan gdab pa). He then
discusses the relationship of the five topics to each other and their
subdivisions (33b7-34b3); with (1) sutrantaprayojana (34b3-36b5); with
(2) pi1Jt}artha (36b5-37a:2); and with (3) padartha (37a:2 foil.). He deals
with topics (4) and (5) in Chapter 3. The five topics provide the basic
structure of the Vyakhyayukti, and are dealt with in the first four
chapters. The fifth chapter, as will be seen below, is an "appendix" on
the benefits of listening to the dharma.
A citation of the summarizing verse by Haribhadra in his Abhi-
samayaZaTJ'lkaraZoka gives us the Sanskrit technical terms: 54
(Haribhadra) tatra:
pray;}janal!! sapi!liirthal!! padiirthaf/. siinusal!!dhikaf/..
sacodyaparihiiras ca viicyaf/. sutriirthaviidibhift.
iti paflcabhir iikiiraif/. sutral!! vyiikhyiitavyam iti vyakhyayuktau ni'!lftam.
(Vyiikhyiiyukti 33b5) 'dir smras pa:
mdo don smra ba dag gis dgos pa bsdus pa'i don bcas dan,
tshig don bcas dan mtshams sbyor bcas, brgallan bcas par bsfiad par bya.
54. WOGIHARA 15,23. See below for the relationship between Haribhadra's Aloka
and the Vyiikhyiiyukti. The verse is also cited at Bu ston (OBERMllLER) 171, and
Sa-skya adapts the five topics in his mKhas pa la 'jug pa'i sgo (David P.
JACKSON: The Entrance Gate for the Wise (Section Ill): Sa-skya Pa!ltj.ita on
Indian and Tibetan Traditions of Pramii!la and Philosophical Debate, 2 Vols.,
Vienna 1987: I 195-96).
SKILLING 319
A slightly different list is given in the
(142,11) under the heading vyakhyasarrzgrahamukha: satrasyotpattipra-
yojana, padartha, anusandhi, abhipraya, codya, parihara.
A secondary structure is based on the "Hundred Extracts". The first
two chapters of the Vyakhyayukti deal systematically with the "Hundred
Extracts" in order of occurrence: Chapter 1 opens with the fIrst extract,
while Chapter 2 opens with the second and closes with the last. Although
the extracts are thus relevant to only the first two chapters, the last
extract ( 109) sets the stage for Chapters 3 and 4. Vasubandhu usually
gives abbreviated citations; GUI).amati usually repeats the extract in full.
Vasubandhu classifies the first ten extracts by subject matter; after that,
he introduces them without classifIcation. A concordance of the first ten
extracts and their headings is given as Appendix 1.
Chapter 1
Chapter 1 opens with the statement (32a2) that "from the very start
(adim eva), one who wishes to explain the sutras (satrarrz vyakhyatu-
kama) should have listened to many [teachings] (thos pa man ba =
bahusruta), should be endowed with the basis of listening to [teachings]
(thos pa'i gii can), and should accumulate [teachings] that have been
listened to (thos pa bsag par bya'o)". Vasubandhu then refers to and
explains the fIrst sutra extract.
In dealing with padartha (tshig gi don), Vasubandhu gives thirteen
examples of one word having diverse meanings (VyY 38b3, tshig gcig
la don du ma).55 Each example opens with a verse of up to four lines
listing the several meanings, each of which is then illustrated by a brief
citation from scripture. For example, the treatment of the first term,
rapa, opens with two lines of verse:
gzugs ses bya ba [1] gzugs phun dan,
[2] kha dog [3] bsam gtan [4] mam par 'dren.
"Riipa" refers to [1] the aggregate of matter (riipaskandha);
to [2] colour (varrza), [3] meditation (dhyilna), and [4] aspect (ilkilra).
A list of the thirteen terms is given as Appendix 3.
55. The same term occurs in GUI,lamati's Pratftyasamutpildiidivibhangildinirdeia!fkil,
chi 112a5: sgar gcig la don du mao
nABS 23.2 320
Vasubandhu then deals with three types of meaning (42b3):
(1) so so re re La brjod par bya ba 'i don;
(2) bsdus pa'i don (pil)ljiirtha);
(3) dgos pa'i don (prayojaniirtha).
Under the latter, he discusses the purpose of "numerical teachings"
(grafts gsufts pa, 43b2). The chapter concludes with seven reasons why
the Buddha travelled from place to place; fifteen reasons why the
b h i k ~ u s travelled from place to place; reasons why the Buddha remained
in seclusion and refrained from teaching for periods of a fortnight and
of three months; and fifteen reasons for going into the presence of
another. The verse summaries for the first three topics are cited without
attribution by Haribhadra in his Abhisamayala1!lkaraloka.
56
Chapter 2
Chapter 2, the longest, deals with the remaining extracts. It opens with
"four extracts in praise of the Buddha", which are followed by "three
extracts in praise of the Dharma" and "one extract in praise of the
Saqlgha" (see Appendix 1).
Vasubandhu follows more or less the same technique in his treatment
of the individual extracts. First he gives an abbreviated citation, for
which the reader must refer to the "Hundred Extracts" for the full
passage. He then explains or glosses the passage phrase by phrase in
order of occurence (go rims biin, 32b5, 33bl, etc.). He often supplies
alternate explanations, introduced by gian yaft (46a2, 5, 8) or rnam
grafts gian yaft (35b6, etc.).
The last extract, at the end of the chapter (96b8), defines a monk who
"knows the Dharma" (chos ses pa = dharmajiia) as one who knows the
twelve aftgas, the "classes" or "gemes" of the Buddha's teaching. The
citation is parallel to Aftguttaranikaya IV 113,12-16, which lists the nine
aft gas of the Theravadin tradition. Vasubandhu defines each of the
aftgas in turn; when he comes to the Vaipulya (97a8), he mentions the
Mahayana for the first time:
The Vaipulya anga is the Mahayana: the instruction through which bodhisattvas,
by means of the ten perfections (dasapiiramitii), accomplish the ten stages
56. For an edition, study, and translation of these passages, see PRAPOD
Assavavirulhakarn & Peter SKILLING, "Vasubandhu on Travel and Seclusion,"
Manusya: Journal of Humanities 2.1 (Chulalongkorn University, Bangkok,
March 1999), pp. 13-24.
SKILLING 321
(dasabhumi) and attain Buddhahood, the foundation of the ten powers
(dasabaliiSraya ).57
At 97b5, Vasubandhu explains the terms Vaipulya and Mahayana. The
latter is so called because it is "an extensive vehicle", because it
possesses "seven greatnesses", which Vasubandhu lists.58 Chapter 2 ends
with the statement that the one hundred extracts have been explained in
part; using this part as an example, the wise may investigate the meaning
of other phrases [from other sutras] (98a5).
Chapter 3
Chapter 3 (98a7) opens with:
Furthermore, the meaning of terms (tshig gi don = padartha) is to be understood
in four ways:
(1) through synonyms (rnam graris = paryaya);
(2) through definitions (mtshan fiid =
(3) through derivation (ries pa 'i tshig = nirukti)
(4) and through classification (rab tu dbye ba = prabheda).59
Vasubandhu gives a definition and a canonical example for each, noting
that the Buddha "employs one or the other after taking into account the
specific needs of the [person to be] trained" ('dul ba'i khyad pa, 98b2).
This concludes the third topic, padartha (99b8), which is followed by an
explanation of the fourth topic, anusaf!ldhi (99a1-100b3).
Vasubandhu then deals with the fifth topic, "raising an objection"
(brgal ba = codya), which is of two types: "objection to the wording"
(sgra la brgal ba, l00b4-IOla5) and "objection to the meaning" (don la
brgal ba). The latter is also of two types: sna phyi 'gal bar brgal ba
(IOla5), "raising the objection that there is an inconsistency of 'before
and after' (contextual inconsistency?)" and rigs pa dan 'gal bar brgal ba
(lOlb5), "raising the objection that there is a logical inconsistency".
"Logic" (rigs pa = yukti) is defined as the three pramalJa:
anumiina, and aptagama (102b6 mdor na rigs pa ni 'dir tshad ma rnam
pa gsum po mrLOn sum pa dan rjes su dpag pa dan yid ches pa'i gsun
no).
57. For Vaipulya see Peter SKILLING, Mahasutras: Great Discourses of the Buddha,
Vol. II, Parts I & II, Oxford 1997, pp. 31-42.
58. Cf. Bu ston (OBERMILLER) I 38.
59. The passage is cited at Bu ston (OBERMILLER) I 72.
JIABS 23.2 322
The discussion of the fifth topic takes up the entire chapter, and is
divided into two parts. The fIrst (lOOb3-104a4) raises various objections
by citing passages from sutras that apparently contradict each other.
Only the objections are given, without response. For example, under sria
phyi 'gal bar brgal ba (lOlb2):
In some places "the four truths of the noble" are taught.
6o
In some places "the
three truths of the brahrnans" are taught.
61
In some places it is taught that "there
are two truths: conventional truth (sa1'J'lvrtisatya) and ultimate truth (paramiirtha-
satya)"; in some places it is taught that "There is only one truth: there is no
second".62
The final objection to be raised (103b2) refers back to Vasubandhu's
definition of the Vaipulya ariga as the Mahayana: the Vaipulya ariga
cannot be the Mahayana, which cannot be the teaching of the Buddha,
since it is inconsistent with the teaching of the Buddha well-known to all
schools (sarvanikaya). Passages from Mahayana sutras - "all dharmas
are without inherent nature (niJ:tsvabhava), are unborn (ajata), are
undestroyed (aniruddha); all dharmas are from the beginning calm
(iidisanta), by nature in complete nirviil).a (parinirvJ1a)" - are set against
passages from the canon of the Sravakas - "In dependence upon the eye
and visible form, visual consciousness arises"; "Alas - conditioned
things are impermanent: they have the nature to arise and cease"
(Udanavarga 1:3).
In the second part (from 104a4), Vasubandhu answers the objections
one by one, in order of occurrence. All but the last objection - that the
Vaipulya ariga cannot be the Mahayana, which is taken up in Chapter 4
- are dealt with in the remainder of the chapter.
In several cases Vasubandhu brings in the important concept of dgoris
pa = abhipraya, the intention of a specific teaching, a theme funda-
mental to many of his works. For example, replying to the objections
cited above, he states: "The various enumerations by which truth is
taught to be fourfold, threefold, twofold, or single do not contradict
each other, since they [were spoken with] different intentions (dgoris pa
tha dad pa'i phyir, 109b8)."
60. Cpo Sa1'J'lyuttanikiiya V 425,14, etc.
61. Cpo Ariguttaranikiiya II 176,26, which gives four.
62. Suttanipiita 884a; Nalinaksha Durr (ed.): Bodhisattvabhiimi/:! (Being the XVth
section of Asarigapiida's Yogiiciirabhiimil}), Patna 1978, p.198,15.
SKILLING 323
Chapter 4
Chapter 4 is devoted entirely to a debate on and a defence of the
authenticity of the Mahayana. It opens (113a8) with the last objection
raised in the first' section of Chapter 3, repeating the challenge: "Because
it is inconsistent with the well-known teachings of the Buddha, the
Mahayana cannot be the word of the Buddha. Therefore [your] state-
ment that 'the Vaipulya is the Mahayana' contradicts the scripture
(iigama)." Vasubandhu turns the objection back upon the "opponent":
"If this is so, then what for you is 'the well-known teaching of the
Buddha' is. equally not the word of the Buddha, because it is self-
contradictory". He then cites a number of contradictory sutra passages:
"There are three feelings: pleasant, unpleasant, and neutral";63 "What-
ever feeling there may be is suffering";64 "Pleasant feeling should be
viewed as suffering";65 "To perceive suffering as pleasure is a misappre-
hension (viparyiisa)",66 and so on.
67
Vasubandhu then makes the statement that "in the Sravakayana, the
teaching of the Buddha is not extant in its entirety" (114b2), which he
backs up with citations from and references to the sutras. He maintains
that the canon of the Sravakas is incomplete, since Ananda did not know
all the sutras, and that there is in fact no "teaching of the Buddha well-
known to all schools" since the canon is arranged differently by different
schools, and since some schools reject certain sutras accepted by other
schools. Other topics dealt with include "final" and "provisional"
meaning (nftiirtha, neyiirtha), "teaching with a specific intention"
(abhipriiya), and "literal interpretation" (yatharutiirtha).
The longest citation of a Mahayana sutra, which takes up six folios, is
from Chapter 2 of the Mahayiinaprasiidaprabhavaniisiitra (P8l2) on the
twenty-eight unwholesome views that arise "when a bodhisattva analyses
the Dharma in a literal (yathiiruta) and incorrect (ayonisas) manner". A
similar citation, with some variants, is given by Asailga in his Abhi-
63. SaTflyuttanikiiya IV 204, etc.
64. Majjhimanikiiya ill 208,27; SaTflyuttanikiiya II 53,20; IV 216,17 fo11.
65. SaTflyuttanikiiya IV 207,6.
66. Cpo Aliguttaranikiiya II 52,20.
67. The last three passages are cited together at 6:3,
(PRADHAN) 330, 11-13, (DWARIKADAS) 880,2-4.
JIABS 23.2 324
dharmasamuccaya as from the Mahadharmadarsadharmaparyaya.
68
Otherwise, the siitra and the passage seem little known.
69
The final topic dealt with is whether Sakyamuni Buddl]a was an
emanation (nirmana, 129M foll.). This brings up various points con-
nected with Sakyar'nuni's career as a bodhisattva and a Buddha, and the
subject of his past unwholesome deeds that gave rise to negative effects
in his final life (karmaploti, 131a6). Vasubandhu refers to the Upaya-
kausalyasiltra and the Lokottaraparivarta for the correct interpretation
of the latter.7
0
At the mention of the Upayakausalyasiltra, the "oppo-
nent" states that it is not authoritative, since it allows a bodhisattva to
kill.?l Vasubandhu then explains under what circumstances and .with
what intention it is permissable for a bodhisattva to do so (131b7 foll.).
From this brief summary, it may be seen that the chapter is extremely
interesting in that it contains many arguments not seen elsewhere.
68. Pralhad PRADHAN (ed.), Abhidharma Samuccaya of AsaTiga, Visva-Bharati,
Santiniketan 1950 (Visva-Bharati Studies 12) 84,3-10. The Abhidharma-
(112,18) gives the same title, as does the Tibetan of the
Abhidharmasamuccaya, P5550, Vol. 112, sems tsam, ii, 124a4: chos kyi me loTi
chen po 'i chos kyi rnam graTis. The Peking Kanjur title is ('Phags pa) Theg pa
chen po la dad pa rab tu sgom pa ies bya ba theg pa chen po 'i mdo. At the end
of the sutra itself (37a3-5), the Buddha gives Ananda four alternate, "internal"
titles, according to a traditional formula; all four contain the element mahayana-
prasada-, and none of them dharmadarsa. Kazunobu MATSUDA has written (in
Japanese) about the sutra with reference to the Abhidharmasamuccaya and
Vyakhyayukti in his "On the two unknown Sutras adopted by the Yogacara
School, based on a passage found in the writings of Bu ston and Blo gros rgyan
mtshan," in Zuiho YAMAGUCHI (ed.): Buddhism and Society in Tibet, Tokyo
1986, pp. 269-89.
69. EDGERTON (BHSD 425a), for example, gives only one reference under the
"Kanjur title," to Mvy 1365, and none for Mahadharmadarsa. The opening of
the passage cited by Vasubandhu is given under the "Kanjur title" by Bu ston in
his "Treatise on the Tathagatagarbha": see David SEYFORT RUEGG, Le Traite du
Tathagatagarbha de Bu Ston Rin Chen Grub, Paris 1973, p. 84.
70. For the former see the translation from the Chinese as "On the Paramita of
Ingenuity" in Garma c.c. CHANG (ed.): A Treasury of Mahayana Satras,
Pennsylvania State University Press 1983, pp. 455 foil.; translation from Tibetan,
Mark TATZ: The Skill in Means (Upayakausalya) Satra, Delhi 1994, pp. 74 foIl.
For karmaploti see Sally Mellick CUTLER, "Still Suffering After All These
Aeons: the Continuing Effects of the Buddha's Bad Karma," in Peter Connolly
and Sue HAMILTON (ed.): Indian Insights: Buddhism, Brahmanism and Bhakti:
Papers from the Annual Spalding Symposium on Indian Religions, Luzac
Oriental, London 1997, pp. 63-82.
71. See A Treasury of Mahayana Satras, pp.456-57; TATZ, pp.73-74.
SKILLING 325
Chapter 5
On the whole, Chapter 5 leaves controversy behind. In the main it deals
with how and why one who explains the siitras should generate respect
in his audience .. It contains numerous similes, including a number of
"numerical" similes, some of which are common to the Gathartha-
sarrzgrahaSiistra:
Like water the Dharma bestows five benefits; 72
Like fire it has four functions;73
Like a pleasant beach it is resorted to for five reasons;74
As for three reasons a cloth does not take drawings, so for the "pictures" of
jiiiina on the cloth of the mind;75
Like a monkey the mind is agitated in two ways;76
Like the sun rising in the sky, the Buddhavacana performs five functions;77
As there are five benefits from relying on a competent guide, so for relying on a
competent teacher;78
As leather becomes workable for three reasons;79
As three strengths enable travellers to safely traverse a jungle path. 80
Similes of limestone (l42b6) , a sentinel (142b8), musicians and a blind
man with a lamp variously illustrate the dangers of having listened to
many [teachings] (bahusruta) but knowing only the words and not the
meaning (artha).81 Examples are given of how even a little listening to
the Dharma brings great benefit.
82
Ten distinctions that make good
stronger than evil are listed (146a5).
Vasubandhu lists ten reasons why the Dharmavinaya is well-taught
(legs par gsuris pa = svakhyata, 135a3). He defines the seventh (137b6),
72. Vyiikhyiiyukti 141a4; Giithiirthasa7J7grahasiistra 260b8; Bu ston (OBERMILLER)
110.
73. Vyiikhyiiyukti 141a4; Giithiirthasa7J7grahasiistra 260b5; Buston (OBERMILLER)
111.
74. Vyiikhyiiyukti 141b3; Giithiirthasa7J7grahasiistra 261a3; Bu ston (OBERMILLER)
111.
75. Vyiikhyiiyukti 141b7; Giithiirthasa7J7grahasiistra 261a7.
76. Vyiikhyiiyukti 142a6.
77. Vyiikhyiiyukti 142b2; Giithiirthasa7J7grahasiistra 262b3.
78. Vyiikhyayukti143a6.
79. Vyiikhyiiyukti 146a1; Bu ston (OBERMILLER) I 80.
80. Vyiikhyiiyukti 146a3; Bu ston (OBERMILLER) I 80.
81. All are cited at Bu ston (OBERMILLER) I 82-83.
82. Vyiikhyiiyukti 143b2 foll.; Bu ston (OBERMILLER) 181.
JIABS 23.2 326
"because it brings about comprehension" (go bar mdzad pa niei) by
"with a single utterance [the Buddha] brings understanding to world-
systems without end" (dbyans gcig gis 'jig rten gyi khams mtha' yas par
go bar mdzad pa). Under the tenth, "because it comprises [all]
qualities", he cites the Tathagataguhya on the sixty qualities of the
Buddha's voice.
At folio 143a4, Vasubandhu gives a definition of the term "treatise"
(Sastra), and cites, without attribution, a verse:
non mons dgra roams ma Ius 'chos pa daTi,
nan 'gro srid las skyob pa gan yin te,
'chos skyob yon tan phyir na bstan bcos fe,
gnis po 'di dag gian gyi lugs la med.
The same verse is cited by Candrakirti in his Prasannapada: 83
yac chiisti klesaripun Sal{lfriiyate durgatito bhaviic ca,
tac chiisaniit triil)agul)iic ca siistraTf1, etad dvayal{l niisti.
It is also cited by GUlfaprabha in his Pancaskandhavivarm:za, which is a
commentary on Vasubandhu's Pancaskandhaprakarw:za.
84
GUlfaprabha
ascribes it to Acarya A.ryadeva (slob dpon 'Phags pa lha).
Vasubandhu cites the Arthavistara, an important sutra of the (Mula)-
Sarvastivadins, three times (140a8; 134h6; 134a3).85 At folio 144b5, he
cites and explains a verse on how the Dharma is no longer properly
explained, and refers to two losses: the loss of the letter and spirit of the
Agama, and the loss of understanding. Similar sentiments are expressed
in the final verses of the Abhidharmakosa (8:39, 41).
83. Louis de LA V ALLEE POUSSIN: Madhyamakavrttii: Mulamadhyamakakiirikiis
(Miidhyamikasutras) de Niigiirjuna, repr. Osnabrlick 1970, p. 3,3-4. According
to LA VALLEE POUSSIN, n. 3, the verse is missing in the Tibetan.
84; P5568, Vol. 114, sems tsam, hi, 68a8. The verse is also cited at Bu ston
(OBERMILLER) I 42.
85. J.-U. HARTMANN has identified Sanskrit fragments of this text from Central
Asia, and demonstrated that it belonged to the $a!sutrakanipiita: see his
"Fragmente aus dem Dlrghagama der Sarvastivadins" in ENOMOTO,
HARTMANN, and MATSUMURA: Sanskrit-Texte aus dem buddhistischen Kanan:
Neuentdeckungen und Neueditionen, G6ttingen 1989, pp. 40-46.
ill. Sources of the Vyakhyayukti
Citations from the canon of the Sravakas
SKILLING 327
All of the "Hundred Extracts" are culled from the scriptures of the
Sravakayana. Inaddition to the "Hundred Extracts", in the Vyakhyayukti
Vasubandhu cites briefly or refers to numerous other sutras. Those
referred to in Chapters 1 and 2 belong exclusively to the canon of the
Sravakas; those cited or referred to in Chapters 3, 4, and 5 belong to
both the canon of the Sravakas and to the Mahayana. Chapter 5 has
relatively few citations. As a whole, most of the citations and references
are from the scriptures of the Sravakayana. Most citations are in prose.
As in his Pratftyasamutpadadivibhmiganirdea,
and Gatharthasa1[lgrahasastra, Vasubandhu cites verses under the rubric
gtithti.
The specific source of the "Hundred Extracts" and the other citations,
with a very few exceptions, is the canon of the (MUla)Sarvastivadins.
This can be established by their phraseology. For example, extract 2,
on the Buddha's declaration of his enlightenment, corresponds to a
passage in the MUlasarvastivadin Sanghabhedavastu, as does the citation
from the *CetalJparyayabhijfianirdesa (see Appendix 5.1) as amplified
by GUl;tamati. 86 Furthermore, Vasubandhu cites by title a number of
sutras which, so far as I know, are unique to the (MUla)Sarvastivadin
tradition, such as the Arthavistara, the * Ja{ila, the Paramarthasiinyata,
the Mayajala, and the Hastatarjopama Siitras.
87
The Paramartha-
siinyata, the Hastatarjopama, and the * Ja{ila (under the title Bimbisara)
Siitras are also cited by Vasubandhu in his as
are many of the sutras cited without title. The (MUla)Sarvastivadin
affiliation also holds for the citations in GUI.lamati's Fka, since
GUI.lamati simply amplifies the references given by Vasubandhu. It also
holds for the and the other works of Vasu-
bandhu to be mentioned below.
86. I suspect that this is a reference to a section of a larger work (whether siitra or
vinaya), rather than the title of an independent work.
87. For the Ja!ilasiitra (known also as the Bimbisiira- or Bimbisiirapratyudgamana-
siitra), see Peter SKILLING, Mahiisiitras: Great Discourses of the Buddha, Vol.
II, Parts I & II, Oxford 1997, pp. 267-333 (for the title Ja!ilasatra, see p. 278).
For the Miiyiijiilasiitra see ibid., pp. 227-65.
88. Cf. Bhikkhu PAsADIKA, "Uber die SchulzugehOdgkeit der Kanon-Zitate im
in H. BECHERT (ed.): Zur SchulzugehOrigkeit von
Werken der Hfnayiina-Literatur, Part 1, G6ttingen 1985, pp. 180-90.
nABS 23.2 328
The "few exceptions" are citations from the recensions of other
schools. As in his and Karmasiddhiprakaral}a,
in such cases Vasubandhu usually, but not always, names the school in
question (see Appendix 5.2). An interesting example is at folio 33a1,
where, after analysing the phrase svarthaf!! suvyafijanam, Vasubandhu
notes that "some recite sa-arthaf!! sa-vyafijanam" (kha cig ni don dan
ldan pa dan, tshig 'bru dan ldan pa ies 'don te). Here neither Vasu-
bandhu nor his commentator GUI).amati say who these "some" are, but
the reading agrees with that of the Theravadin tradition, satthaf!!
sabyafijanaf!! (Majjhimanikaya 1179,7, etc.).
Citations from the scriptures of the Mahayana
A list of Mahayana scriptures cited by title is given in Appendix 6. In
addition, Vasubandhu cites several without title. The Saf!!dhinirmocana
is also cited by Vasubandhu in his Kannasiddhiprakaral}a and referred
to in his Pratftyasamutpadadivibhanganirdea.
Sources ofVasubandhu's methodology
It is probable that Vasubandhu derived most of his terminology and
methodology from earlier works, and that his originality lay in his
arrangement and use of traditional material. He makes frequent
references to agama (tun: see Appendix 5.1), sometimes in verse; in
most cases the style and phrasing of the citations shows that he does not
mean the canonical Agamas, but is referring to an unidentified exegeti-
cal tradition. As is common in Buddhist sastra literature, Vasubandhu
occasionally refers to the opinions of "some" (kha cig = kecit: 67a6;
95b8) or "others" (gian dag = anya: 80b8, in verse; 112b4). As noted
above, the Vyakhyayukti contains many other verses in a non-canonical
style; whether any of these were composed by Vasubandhu remains to be
determined.
As a manual of exegesis. the Vyakhyayukti may be compared with the
Pali Nettipakaral}a of the pre-Buddhaghosa Theravadin tradition, so far
as I know the only other major treatise devoted to this subject that has
come down to us. Apart from the occasional use of similar terms, such
as saf!!kldabhagfya, and a similar interpretation of the mahapadda,
which might derive from a common source, I have not noticed any
striking correspondences between the two texts (but a thorough study
remains to be made). Exegesis is also dealt with by Asanga in his
SKILLING 329
Yogacarabhilmi
89
and in the Sal'J'lkathyaviniscaya, the last chapter of his
Abhidharmasamuccaya.
90
A passage in the latter employs categories
similar to those of the Vyakhyayukti (36b5-37al).91 Only a thorough
comparison of these texts will reveal their relationship.
IV. Translators and Translation
Since more or less the same team of translators rendered all three works
into Tibetan, we are presented with a unified body of work. The works
are listed in the lDan (or Lhan) dkar Catalogue of the early 9th
century.92 Visuddhasirpha and were also responsible for
the Tibetan version of Vasubandhu's KarmasiddhiprakaralJ-a (P5563;
KSP pp. 174-75) and the accompanying Fka by SumatisTIa (P5572;
KSP p. 175), as revised by Srikiita (dPal brtsegs).
I have not been able to find any biographical information for any of
the translators, who also worked on the following texts:
93
VisuddhasiIpha et al.: P166 (dhiirmyJ); 351 (id.); 831; 855; 856; 874; 973
Visuddhasiqlha, Sarvajiiadeva, Prajiiavarman, Ye ses sde, dPal brtsegs: P767
Visuddhasiqlha, dPal brtsegs: P5593 (Yasomitra's Abhidharmakosavyakhya)
SakyasiIpha, P940
Visuddhasiqlha, Devendraralqita: P935
P994
Jinarnitra, et al.: P827
Vidyakaraprabha:
94
P141 (dhiiral}l), 508 (id.)
MaiijusrIvarman et al.: P367 (dharal}l)
Sarvajiiadeva, dPal brstegs: P891; 896; 923; 999; 1013
Sarvajiiadeva et al.: PI030 (Vinayavastu); 1034
89. Cf. WAYMAN: "Doctrinal Affiliation," p. 203.
90. Ed. Pralhad PRADHAN, Santiniketan 1950, pp. 102 foll.; French translation by
Walpola RAHULA: Le Compendium de la Super-doctrine (Philosophie) (Abhi-
dharma-samuccaya) d'Asmiga, Paris 1971, pp.l77 foIl.
9!. PRADHAN, p.103,9-11; RAHULA, pp.178-79. Cf. also Abhidharmasamuccaya-
142,5-9.
92. That is, the Vyakhyayukti and GUJ.].amati's !fka: Marcelle LALOU, "Les textes
bouddhiques au temps du roi Khri-sroiJ.-lde-bcan", Journal Asiatique 1953,
649, 650 (in section XXIV, rnam paries pa 'i bstan bcos): [649] rnam par Mad
pa'i rigs pa, 1800 slokas, 10 bampos; [650] rnam par Mad pa'i 'grel pa, yon
tan blo gros kyis mdzad pa, 4500 slokas, 15 bampos. I suspect the Sutrakhaw!a-
sata is included with the Vyakhyayukti.
93. For this list I have used the catalogue volumes of the Otani reprint of the Peking
Kanjur and Tanjur.
94. Vidyakaraprabha worked with dPal brstegs: P402, 432, 433.
nABs 23.2 330
The corpus of tl).eir work shows that the translators collaborated with the
well-known Jinamitra, Ptajiiavarman, dPal brtsegs, and Ye ses sde on
the translation of Dhara" and Sutra, the latter both of the M$ayana and
Sravakayana (and, in the case of Sarvajfiadeva, on the Vinaya). Their
activity belongs the period of intensive translation during the "First
Propagation" (sna dar) of Buddirism in Tibet, in about the beginning of
the 9th century. Their familiarity with the thought of Vasubandhu shows
itself in a lucid rendering, equal to the best standards of Tibetan
translation.
v. The Influence of the Vyakhyayukti
Influence in India (1): The Tarkajvala of Bhavya
The Tarkajvala is an encyclopredic work composed by Bhavya (circa
500-70?) as a commentary on his Madhyamakahrdayakarikas.95 It
contains an important defense of the Mahayana, which in some cases
seems to rely on the Vyakhyayukti.
96
At one point Bhavya cites a verse
list of sutra titles that appears to be from the Vyakhyayukti:
97
(VyY 114b3) (Tarkajviilii 180a5)
bla ma kun dga' 'od sel dan,
sdug bsnal phun ston sa ston dan,
'char ka ston fiid chu las skyes,
gan po sa mtsho chu sin dan,
my a nan 'das dan yul 'khor srun,
'phro ba'i mdo dan de biin gian,
yan dag bsdus pa'i gii fiams phyir,
mtsha' dag min par rtogs pa yin,
bla ma kun dga'pa ta l ~
sdug bsnal phun po sa ba dan,
'char byed skyod pa 'chan ba dan,
gan po spans par ma pa dan,
mya nan' das Ian yul 'khor skyon,
ies bya gian yan 'gro ba'i mdo,
rtsa ba 'i sdud pa ma tshan bar,
rgyas pa min par rtogs par bya.
Influence'in India (2): The Arthaviniscayasutranibandhana
The Arthaviniscayasutranibandhana (Arthav-n) is a commentary on the
ArthaviniScayasutra, a long compendium of many of the basic terms and
categories of Buddhism. The Nibandhana, composed by VrryaSrIdatta at
95. For references to Bhavya and his date and work, see Peter SKILLING, "A Citation
from the *BuddhavaT[lsa of the Abhayagiri School," Journal of the Pali Text
Society XVIII (1993), pp. 168-69.
96. It is possible that some of Bhavya's material is taken from a common source,
rather"than directly from Vasubandhu. The extent of his debt to Vasubandhu can
only be judged when the two works have been properly edited.
97. P5256, Vol. 96, dbu ma, dza.
SKILLING 331
Nalanda during the reign of King Dharmapala (latter half of the 8th
century),98 cites or paraphrases the Vyiikhyiiyukti a number of times.
Vlryasrldatta cites three verses from the Vyiikhyiiyukti, without naming
his source:
(1) (Arthav-n '72,4) tatM ea iiha:
srutvii satrasya miihiitmya'!l sratur iidarakiiritii,
srava1}adgraha1}a'!l syiid ity iidau viieii'!l prayajana'!l.
(VyY 34a1) 'dir smras pa: (VyY133a7) snar:
mda dan ehe ba nid thas na, mda sde'i ehe ba nid thas na,
nan pa dan ni 'dzin pa la, mnan pa dan ni bzun ba la,
nan pa pa ni gus byed pas, nan pa pa ni gus byed 'gyur,
thag mar dgas pa brjad par bya. de phyir thag mar dgas (corr. from dgans)
pa brjad. ees bstan pa yin.
This verse is cited by Bu stan (OBERMILLER) I, p. 17.
(2) (Arthav-n 77,2) yathaktam:
aisvaryasya samagrasya rupasya yasasaJ:! sriyaJ:!,
jniinasyiitha prayatnasya bhaga iti srutif:t.
(VyY 46b7) 'dir Mad pa:
dban phyug dan ni gzugs bzans dan,
brtsan 'grus phun sum ldan pa yi,
grags dan dpal dan ye ses dan,
drug pa dag la beam ies Mad.
This verse is also cited by Haribhadra in his AbhisamayiilaT[lkiiriiloka.
99
(3) (Arthav-n 247,5 - in verse?) padiinii'!l punaJ:! pi1}iirthaJ:!:
bhanktvii vibandha'!l siistrtvasampada'!l ya'!l hi yena saJ:!,
yatMgataJ:! siistrkarma yae ea yais ea parigrahaJ:!. ity etat paridfpitam iti.
(VyY 45b7 - prose) tshig rnams kyi bsdus pa'i dan de ni bgegs beam nas stan
pa phun sum tshags pa gan du gan gis, ji ltar gsegs pa dan, stan pa 'i phrin las
gan yin pa dan, gan dag gis yans su gzun ba ies bya ba 'di yans su bstan ta.
Several prose passages, such as parts of the sections dealing with the
four brahmavihiira
100
and with the four srota-iipattyariga
101
are based on
the Vyiikhyiiyukti. Otherwise, VIryasrIdatta relies heavily on the Abhi-
98. See Arthav, introduction, pp. 133-36; Arthav-n 312,3-6. I have not been able to
trace any references to VlryasrIdatta in Tliranatha.
99. WOGlliARA: AbhisamayiilaJ'!lkiiriilakii, p. 7,27, introduced by atha vii.
100. Cpo Arthav-n 193,3 -195,1 with VyY 65b8-66a8.
101. Arthav-n 240-59. Since the VyY does not deal with this subject, the parallels are
scattered here and there and do not follow the same order. Cp., for example,
Arthav-n 248, 5-249,5 (on the Dharma) with VyY folios 49a, 49b, and 50a;
Arthav-n 250, 5-252,2 (on the Sarp.gha) with VyY folio 51a; Arthav-n 256,
6-258,6 (on sfla) with VyY folio 82b.
JIABS 23.2 332
and Yasomitra's Vyiikhyii (as shown by Samtani in
his notes).102 For his interpretation of pratftyasamutpiida, he uses Vasu-
bandhu's Pratftyasamutpiidiidivibhailganirdesa.
103
Other passages may
be traced to the Giithiirthasa""grahasiistra, and to the commentaries on
Vasubandhu's PancaskandhaprakaralJa by GUl).amati and PphivIbandhu.
Thus Vlryasddatta's work may be described as a compendium of !be
thought of Vasubandhu as applied to the interpretation of the Artha-
viniscayasiitra. Its intrinsic value is enhanced by the fact that it supplies
us with examples of the Sanskrit terminology and phraseology of the
Vyiikhyiiyukti and the PratftyasamutpiidiidivibhaiiganirdeSa. Conversely,
since there is no Tibetan or Chinese translation of the Nibandh(ma,
doubtful readings or difficult passages in that text can be clarified by
consulting VIryasddatta's sources in Tibetan translation.
104
(A second and quite different commentary on the Arthaviniscayasiitra,
the anonymous ArthaviniScayatfkii [P5852], which is preserved only in
Tibetan translation, takes a quite different approach. A casual study
shows a preference for the interpretations of AsaJiga. The only parallel
to the Vyiikhyiiyukti that I have traced is the mention of the three types
of pot, discussed above.)
Influence in India (3): The Abhisamayalarpkliraloka
The Abhisamayiila""kiiriiloka was composed by Haribhadra, who resided
at the TrikU.!aka Vihlira during the reign of King Dharmapala, and hence
was roughly contemporary to VIryasddatta. Haribhadra gives five
citations of verse from the Vyiikhyiiyukti. The first, which is also cited in
the Arthaviniscayasiitranibandhana (see above), is introduced without
title by the phrase atha vii. The second, which names the source as the
VYiikhytzyukti, lists the five methods by which the siitras are to be
explained.1
05
Haribhadra also cites kiirikiis from the Kosa.10
6
In his
102. See also SAMTANI's introduction, pp.134-35.
103. A reference to the Sanskrit fragments edited by TUCCI has been noted by
SAMTANI. In addition there are numerous parallels to the complete Tibetan
version.
104. For example, the variant sunirjita for Arthav-n 194, 1, is
confirmed by VyY 66a4, mi mthun pa'i phyogs sin tu thul ba'i phyir.
105. See above, n. 54. For the other three citations see PRAPOD & SKILLING,
"Vasubandhu on Travel and Seclusion" (above, n. 56).
106. WOGlliARA 220,23; 222,2-26; 224,2; 225,7.
SKILLING 333
Sphu!iirtnii Abhisamayiilaftkiiravrtti, Haribhadra cites the Kosa and the
Vi'!!satikii.
107
These references show that the Vyiikhyiiyukti was studied (along with
the other works of Vasubandhu) in the great monasteries of Northern
India during the reign of Dharmapala. Tibetan tradition records that Sa
skya PaI).9ita studied the Vyiikhyiiyukti under Indian pa1}tj.itas,108 which
shows that the work retained its importance in India up to the 12th
century. Since, however, the work has not been edited, translated, or
analyzed to date, the full extent of its influence upon Indian siistra
literature has yet to be determined.
Influence in Tibet (I): Sa skya Pa1}tj.ita and the Vyakhyayuktil09
The Vyiikhyiiyukti was one of the sources used by Sa skya Pal).<;iita Kun
dga' rgyal mtshan (1182-1251/2) for his mKhas pa 'jug pa'i sgo, which
"described for the fIrst time to Tibetans the literary values and scholastic
methods of Indian Buddhists", and set forth "the three main activities of
a traditional scholar: composition, teaching, and debate". The Vyiikhyii-
yukti, cited by Sa pal)., lends its structure to the second part on
exposition.
110
Influence in Tibet (2): Bu ston and the Vyakhyayukti
Bu ston, in his renowned History of Buddhism (Chos 'byuft) classes the
Vyiikhyiiyukti among the "Eight Treatises" of Vasubandhu (OBER-
MILLER, I 56-57), and describes it as "teaching the proper [method] of
exposition and hearing [of the the Dharma] in the light of the doctrine
107. B h i k ~ u SAMDHONG Rimpoche (ed.): AbhisamayiJlmikiJravrttj Sphu!iJrthiJ,
Sarna!h 1977, Tibetan text, p.1l8,1, 118,12-13, respectively.
108. See following section for reference.
109. The information and quotations in !his section are from David JACKSON's
foreword to Mkhas pa la 'jug pa'j sgo, An introduction to the principles and
concepts of Indo-Tibetan scholasticism, Sakya Centre, Dehra Dun 1983. See
also JACKSON's The Entrance Gatefor the Wise (Section III): Sa-skya Pm:ujita
on Indian and Tibetan Traditions of PramiJlJa and Philosophical Debate, 2
Vols., Vienna 1987.
110. JACKSON (1987), Part I, pp.195-96: bsad pa la 'jug pa'i le'u, p. 28, rnam Mad
rigs pa las. See also Part II, Appendix I, for Glo bo mkhan chen's outline of !he
section.
JIABS 23.2 334
of mind only" .111 His citations begin at the very beginning of his work
(OBERMILLER, I 9), with a verse which he attributes to the Bodhisattva-
pi!aka, also found in the Udanavarga and cited by Vasubandbu in both
the Vyakhyayukti and the Gathasa1'[tgraha.
112
He then cites at length the
interpretation of the verse from the Vyakhyayukti (a similar explanation
is given in the Gatharthasa1'[tgrahasastra), followed by several further
citations. In his History he cites the Vyakhyayukti by name 22 times, and
cites GUI).amati's Tfka once)!3 There are also several un attributed
citations (138 = 97b5; I 81 = 140b7-141a1).1l4
The Vyakhyayukti and modern scholarship
The Vyakhyayukti has been neglected by modem scholarship, perhaps in
part because it is not available in Sanskrit or in Chinese translation.
Although much has been written about Vasubandhu, the Vyakhyayukti is
scarcely mentioned. The only early scholar who seems to have consulted
it extensively was OBERMILLER, who, in his translation of Bu ston's
History (see above), meticulously traced the Vyakhyayukti references.
LAMOTTE, in the introduction to his Le Traite de l'Acte (p. 176)
describes the Vyakhyayukti as "un ouvrage lexicologique", while
Ill. de Itar yin yan Mad nan 'thad par ston pa rnam Mad rigs pa: Bu ston rin chen
grub, Bu ston chos 'bywi, Garis ljoris ses rig gi fiiri bcud, Kruri go bod kyi ses
rig dpe skrun khan 1988, p. 31,20.
112. Bodhisattvapi!aka, Chap. XI (Ulrich PAGEL: The Bodhisattvapi!aka: Its
Doctinres, Practices and their Position in Mahayiina Literature, Tring (UK)
1995 (Buddhica Britannica Series Continua V), pp. 341, 344; Udiinavarga 22:6;
Vyiikhyiiyukti 136b3; Giithiisa7!lgraha 240b4; GiithiirthasaJ!lgrahasiistra
257a6; also cited at Samathadeva, Abhidhannakosa-upiiyikii!fkii, P5595, Vol.
118, tu, 51b2.
113. See OBERMILLER's Index, Part II, p. 230.
114. A study of the influence and application of the Vyiikhyiiyukti and its principles in
the Land of Snows is a chapter of Tibetan intellectual history that remains to be
written, lying beyond the range of this already bloated monograph and the
competence of its author. Suffice to add here that it was regularly cited by the
great 19th century scholar Jamgon KONGTRUL (Kori sprul Blo gros mtha' yas,
1813-99) in his Ses bya mtha' yas pa'i rgya mtsho: see for example Jamgan
KONGTRUL Lodra Taye, Myriad Worlds: Buddhist Cosmology in Abhidhanna,
Kiilacakra and Dzog-chen, translated by the International Translation Committee
of Kunkhyab Chaling founded by the V.V. Kalu Rinpoche, Snow Lion
Publications, Ithaca (New York) 1995, pp. 88-89; Buddhist Ethics, translated by
the International Translation Committee founded by the V.V. Kalu Rinpoche,
Snow Lion Publications, Ithaca (New York) 1998, pp. 68, 76, etc.
SKILLING 335
ANACKER calls it a "history of the Buddhist Canon".1
15
Both descrip-
tions are inaccurate. A better one is offered by Hajime NAKAMURA:
the Vyiikhyiiyukti "teaches how to interpret and explain the content of a
sutra. In the fourth chapter of this work the assertion that Mahayana can-
not be considered as Buddhism is set forth and Vasubandhu refuted
it".116
Exceptions to the neglect include an article by Jose Ignacio CABEZON,
"Vasubandhu's Vyiikhyiiyukti on the Authenticity of the Mahayana
Sutras"ll7 and the work of Kazunobu MATSUDA and other Japanese
scholars. liS
VI. Conclusions: The Importance of the Vyakhyayukti
As a source for Agama studies and sutra citations
In the "Hundred Extracts", we have 109 citations from the (MUla)-
Sarvastivadin scriptures, selected for their significance by Vasubandhu
himself. Most of them are repeated in full in GU.Q.amati's Fkii. In the
Vyiikhyiiyukti, there are many other citations; though generally these are
very brief, they are frequently amplified by GU.Q.amati. In addition, the
Vyiikhyiiyukti contains numerous references to sutra titles or to
personages; in some cases where Vasubandhu gives only the title,
GU.Q.amati supplies a citation which helps us to identify the sutra. A
number of the citations are not found in the Abhidharmakosa or other
works of either Vasubandhu or the V a i b h a ~ i k a tradition. In addition,
there are occasional references to the scriptures or interpretations of
other schools, and references to and citations of Mahayana scriptures.
As a source for Mahiiyiina studies
The study of the origins of the Mahayana and its relationship to the
Sravakayana is one of the most important and challenging fields of
115. ANACKER,op. cit., p.1; p.155, n. 62.
116. NAKAMURA, Indian Buddhism, p. 27l.
117. In Jeffrey R. TIMM (ed.): Texts in Context: Traditional Hermeneutics in South
Asia, Albany 1992, pp. 221-43.
118. See (all in Japanese) the two articles by Susumu YAMAGUCHI mentioned by
NAKAMURA, and Kazunobu MATSUDA, "Buddha's Teaching and its Meaning
based on Passages in the Vyakhyayukti by Vasubandhu," Journal of Buddhist
Studies and Humanities, Otani University, 63-1 (1983), pp. 79-80; "On the Two
Truths in the Vyakhyayukti by Vasubandhu, Notes on Vasubandhu - Part
Two," Journal of Indian and Buddhist Studies 33-2 (1985), pp. 750-56.
JIABS 23.2 336
Buddhist studies. The Vyiikhyiiyukti is one of the few Indian works
extant that treats at length with the validity of the Mahayana and its
relationship to the Sravakayana scriptures, in the form of a highly
literate debate.
ll9
It is thus a primary source for studies on this subject.
As a source for "Vasubandhu studies"
On the evidence of his Abhidharmakosa alone, Vasubandhu has been
rated one of the most learned and interesting Buddhist thinkers of all
time. Since the Vyiikhyiiyukti is a mature, confident, and comprehensive
work, it cannot be ignored in future studies on Vasubandhu. As noted
above, controversy remains about who he was, when he lived, his
relationship to Asanga and the Mahayana, and to the "Elder Vasu-
bandhu". The present study shows that the works of Vasubandhu' slater
period are all interrelated, and that they are related to his earlier
In order to gain a deeper insight into the development of
Vasubandhu's thought, we need proper critical editions of these works,
and - since they cannot be fully appreciated in isolation - a thorough
concordance. This would put "Vasubandhu studies" on a firm footing.
120
As a manual of exegesis
The three points listed above should be sufficient to establish the
importance of the Vyiikhyiiyukti. But they must be subordinated to the
purpose (prayojana) of the text itself.121 The Vyiikhyiiyukti sets out the
principles of exegesis or hermeneutics. Needless to say, the principles of
exegesis described by Vasubandhu may be applied not only to the siitras
that he cites, but to Buddhist literature in general.
On the basis of these points, I venture to state that the Vyiikhyiiyukti
can no longer be safely ignored, and that it is one of the most important
Buddhist works remaining to be studied.
119. Another is Bhavya's Tarkajvtilii, which depends for some of its ideas on the
Vyiikhyiiyukti.
120. I am preparing a critical edition of the three works of the "Vyiikhyiiyukti
literature" .
121. We might say that while the emphasis of the Kosa is the clarification of
categories and classifications (by nature of its being an Abhidharma text), the
emphasis of the Vyiikhyiiyukti is meaning and interpretation (dgons pa,
abhipriiya).
Appendices
Appendix 1: Concordance of the first ten extracts
Chilpter 1
[1] VyY-sii 19<i3-6; VyY 32a3 foIl.; VyY-! 2b1
SKILLING 337
On hearing and realizing the Dharma; parallel to Anguttaranikiiya V 6, 1-6, on
sutadhana.
Chapter 2
Four extracts "in praise of the Buddha" (VyY 45b5, salis rgyas kyi bsliags pa
las brtsams nas ni ... mdo sde'i dum bu bii dag ste):
[2] VyY-sii 19a6-7; VyY 45b7; VyY-! 20a1
The "ten epithets of the Buddha" = Arthaviniscayasiitra 4S,9-11
[3] VyY-sii 19a7-19b1; VyY 47al-48a3; VyY-t 22b5
On the Buddha's proclamation of his enlightenment = Sanghabhedavastu I
136,9-13
[4] VyY-sii 19b1-3; VyY 48a3-bS; VyY-p3b7
On the Buddha as "tamer of the untamed"; parallel spoken by King Pasenadi in
the Angu!imiilasutta, Majjhimanikiiya 86, II 102,19.
[S] VyY-sii 19b3; VyY 48bS-49aS; VyY-t 26a3
"The Tathagata is a knower of the path ... " = Vinayavastu!fkii, PS61S, Vol. 122,
'dul ba'i 'grel pa, dzu, 347a4. Cf. Majjhimanikiiya 107, III 6,8.
Three extracts "in praise of the Dharma" (VyY 49a5, chos kyi bsnags pa las
brtsams na[s] ni ... mdo sde'i dum bu gsum ste):
[6] VyY-sii 19b3-4; VyY 49a7-S0a3; VyY-! 27aS
Stock description of the Dharma = Arthavinikayasiitra 4S, 12-46, 1
[7] VyY-sii 19b5-6; VyY SOa3-S0b2; VyY-! 28a2
[8] VyY-sii 19b6-7; VyY SOb2-8; VyY-! 30a2
One extract "in praise of the SaIp.gha" (VyY SOb8, dge 'dun gyi bsliags pa las
brtsams nas ni ... mdo sde'i dum bu ste);
[9] VyY-sii 19b7-8; VyY SOb8-51b2; VyY-t 30b3
Stock description of the Saq1gha = Arthaviniscayasiitra 46,3-4
One extract on "the side of defilement" (VyY 51b2, kun nas
non mons pa'iphyogs las brtsams nas ni ... mdo sde'i dum bu ste):
[10] Vy-sii 19b8-20aS; VyY 51b2-S2a2; VyY-! 31a4
Appendix 2: Verses in the Vyakhyayukti
Appendix 2.1. SaIpgrahasloka
Chapter!: 44b6; 4Sa5
Chapter 2: 50b6; 80b3; 80b7
Chapter 3: 100b2; l09bS
Chapter 4: 123'a8
Chapter 5: 134a2; 146b6; 148b6; lS2a5; lS2b5.
Appendix 2.2. Introductory phrases to verses
'di Itar smras pa: 34a1 (= Arthavinikayasiitranibandhana: see above)
'dir smras pa: 33b5 (= Abhisamayiila'!lkiirii[oka: see above); 34a5; 59bS; 60b 1
JIABS 23.2 338
'dir: 45a7
'dir ymi: 55a7
'dir ymi smras pa: 61a2, 8; 65b5; 78b2;. 87b8
'dir yan bsad pa: 1l0b6; 127a7; 141a1
'dir Mad pa: 46b7 (= Arthaviniscayasiitranibandhana: see above); 49b6
de la: 69b1
mam grans gian yan: 50a7; 80b4
gian yan: 48a8
gian yan go rims biin: 48a7
gian yan smras pa: 69b3
yan 'dir biad pa: 112b5
yan gian dag na re: 80b4
yid ches pa 'i lun las: 87b1
lun las ni: 60b2; 69b6
Appendix 2.3. Verses without introduction:
45b2; 104a5; 106a5; 142a7; 144a3; 144a5; 144bl, 2,3; 144b5; 145a7; 147a3,5;
147b1,4; 148a1; 148b7; 149a5, etc.
Appendix 3: Thirteen terms with diverse meanings
[1] VyY38b4: gzugs = riipa
[2] VyY38b6: mtha' =anta
[3] VyY 39a2: mchog = agra
[4] VyY 39a8: 'jig rten =loka
[5] VyY39b4: zan zin = a m i ~ a
[6] VyY39b7: 'byun ba = bhiita
[7] VyY 40a4: pa da =pada
[8] VyY 40b2: chos = dharma (cited at Bu ston I, p. 18)
[9] VyY 41a2: spans pa = prahiil}a
[10] VyY 41a6: tshul = nyaya
[11] VyY 41b4: las kyi mtha' = karmanta
[12] VyY 42a2: phunpo =skandha
[13] VyY 42a5: bsdupa = saT[lgraha
Appendix 4: Concordance of Pali and Sanskrit Titles
Titles are given only when the correspondence between versions is certain, in
traditional order of Pali Nikayas, by Sutta number for Dighanikaya (DN) and
Majjhimanikaya (MN) , by Pali Text Society volume and page numbers for
SaT[lyuttanikaya (SN) and Anguttaranikaya (AN). When possible a restored Sanskrit
title is given (otherwise unattested or tentative titles being preceded by an asterisk).
Further details can be found under the appropriate titles in Appendix 5.
Dfghanikaya
DN 16
DN 17
DN34
Mahiiparinibbiina-su
M ahasudassana-su
Dasuttara-su
= Mahaparinirval}a-sii
= Mahasudarsana-sii
= Daiottara-sii
Majjhimanikiiya
MN 13
MN 14
MN28
MN61
MN78
MN82
MN 107
MN 108
MN 115
MN 121
MN 139
Sal'ftyuttanikiiya
SN III 65-66
SN III 86
SN III 105-6
SN V 154-58
SNV 320-22
Ariguttaranikiiya
AN 1229
ANN 402-5
ANN 162-66
AN N 100-6
SKILLING 339
Mahiidukkhakkhandha-su = *(Prathama-)DuJ:.!khaskandhika-sii
CiiJadukkhakkhandha-su = *DuJ:!khaskandha-nirdea (?)
Mahiihatthipadopama-su = Hastipadopama-su
.Ambalatrhikii-riihuloviida-su = Riihula-sii
SamalJamalJtjikii-su
Ra!!hapiila-su
GalJakamo ggalliina-su
Gopakamoggalliina-su
Bahudhiituka-su
CiiJasufifiatii-su
AralJavibhariga-su
Sammiisambuddha-su
Khajjanfya-su
Ananda-su
Bhikkhunfviisaka-su
Vesillf-su (also Yin III 68)
Gadrabha-su 122
Siliiyiipa-su
Uttara-su
Suriya-su
= *Paficiirigasthapati-su
= Riiwapiila-sii
= *Gmj.aka-sii
= *Gopaka-sii
= Bahudhiituka -sii
= Siinyatii-sii
= AralJa-sii
= Arhat-sii
= bZa' ba lta bu'i mdo
= PiirlJa-sii
=
= Ri dags zlog gi mdo
= *Gardabha-sii
= *Acandrikilputra-sii (?)
= Uttara-su
= Saptasiiryodaya-sii
Appendix 5. Sravakayana sources and references
Conventions: In Appendices 5 and 6 references are given in the order VyY-sii, VyY,
VyY-t. References without any preceding siglum are to the primary text, VyY.
References are given first by known or restored (indicated by preceding asterisk)
Sanskrit title, followed by Tibetan title for those for which the Sanskrit is unknown.
When the source of a citation is known but not given by title in the text, the title is
enclosed within brackets. Secondary elements of titles (Prathama-, Arya-, etc.) are
placed in parentheses. Further: -[sii] indicates that -sutra is not included in title as cited;
(vs) = verse; (?) indicates a doubtful restoration or possible parallel, etc.; ariga indicates
one of the twelve arigas of the Buddha's teaching; = indicates parallels rather than exact
equivalents. When minor corrections have been made to the Tibetan of the Peking
edition, the original reading follows the corrected form in curling brackets: { ... }.
Appendix 5.1. Literary references
*Acandrikaputra-sii (?): Ma zla ba'i bu mo'j mdo (VyY-t Ma zla ba'j bu mo'i bu)
= AN N 402-5, Silayiipa-su, in which a Candikaputta plays a role
37a3; VyY-! 10b6, 7
Adbhutadharma, aiJ.ga: rMad du byun ba'i chos kyi sde
VyY-sii 31b6; VyY 97b1; VyY-t 122b5; 124b1 (def.)
122. Title from Syiimaratrhassa Tepi!akal'ft, Vol. 20, p. 308 penult.
nABS 23.2 340
(Pratharna- rMad du byun ba dan po' i mdo
127b6 (= DN 11109; AN N 307); VyY-t 159a4
*Anagiimi-sii: Phyir mi 'on ba'i mdo
VyY-t 153a2
Abhidharma: Chos rnrion pa
VyY-sii 25b8; VyY 77a4; 97b3; 124b1; VyY-t 86a1; 124b5, 7; 162a7
Abhidharmakosa: Chos mnon pa'i mdzod
VyY-t 181a8
Abhivinaya: 'Dul ba rnrion pa
VyY-sii 25b8; VyY 77a4; VyY-t 86a1; 162a7
AraI,la-sii: Non mons pa med pa'i mdo = MN 139, AraI,lavibhanga-su
74a8; VyY-t 79a5 = MN ill 230,15-17
Arthavistara: Don rgyas pa = P984, Vol. 39
134b6; [43a8 (= P984, Vol. 39, mdo, su, 198a2); 74a2 (= VyY-sii no. 67);
134a3; 140a8]; VyY-t 166a7
Arhat-sii: dGra bcom pa'i mdo = SN ill 65, Sammasarnbuddha-su
127b1; VyY-t 157a8; 158a5
Avadana, anga: rTogs pa bIjod pa'i sde
VyY-sii 31b5; VyY 97a6; 141a1 (see Nanda, Gopilla); VyY-t 122b5; 123b8
(def.). See also Drrghala-.
Agarna: Lun
33b2;34b1;36a6;36b5;44a1;48b2;49b8;51a8;54a5;55b4;57a3;58 a1;60b2
(vs); 62b4 (vs); 65b6; 69b6 (vs); 74b6; 76a1; 76b5; 77b7; 79a4; 81b4; 87b1
(vs); 103a8; 104a4; 112b7; 116b2, 3; 125bl; 145a3, 7; VyY-t 6a3; 58a6; 92b8;
93a1;94a6; 133b4; 134a1,3,6; 147b8; 148a1,2
Ananda-sii: Kun dga' bo'i mdo
114b3; 114b7
Ayul}.paryanta-[sii]: Tshe'i mtha' = P973, Vol. 39
116b4
Aryasatya-sii: 'Phags pa'i bden pa'i mdo sde
124b8; VyY-t 153b4 (= SN V 425,14-17, etc.)
ltiv[ttaka, anga: De Ita bu byun ba'i sde
VyY-sii)lb6; VyY 97a7; VyY-t 122b5; 124a5 (def.); 124b5
Uttara-sii: bLit ma'i mdo = AN N, Uttara-su (cf. uddana p. 172,15, Uttaro)
114b3; 114b4 (= AN N 166,7-10); 115a3
Udana, anga: Ched du bIjod pa'i sde
VyY-sii 31b5; VyY 97a4; VyY-t 122b5; 123b6 (def.)
[Udanavarga] 123
36a8 (= Uv 27:25ab, 26ab); 40a5 (= Uv 26:29b); 40a7 (= Uv 4:1ab); 103b7
(=Uv 1:3); 40bl (= Uv 29:25ef, 27ef, 2gef, 31ef); 53a8 (= Uv 28:21); 104a1 (=
Uv 26:9); 113b7 (= Uv 23:17); 114al (= Uv 29:24); 118a2 (= Uv 29:24a);
118a2 (= Uv 29:23a); 125a4 (= Uv 29:24a); 136b3 (= Uv 22:6). See also Gatha.
Udayi-sii: 'Char ka'i mdo
114b3; 115b1 (cp. SN N 224,29-30?); VyY-t 162a6; 163a4
123. References are to Franz BERNHARD (ed.): Udiinavarga, G6ttingen, Vol. I, 1965,
Vol. II, 1968 (Sanskrittexte aus den Turfanfunden X) = Uv.
Uddiina: sDom
VyY-sii 25a4
-gatha: - gyi tshigs su beadpa41a7; 61b5; VyY-t51a6
Upadesa, anga: bsTan la dbab pa bstan pa'i sde
VyY-sii 31b6;.VyY 97b2; VyY-t 122bS; 124b7 (def.)
Upali-sii: Ne bar 'khor [ba'i] mdo sde
VyY-t 147b7
Ekottara: gCig las 'phros pa
116b5
*KadalI-sii: Chu siil gi mdo
114b3; 115b8
SKILLING 341
*Kapinjala-jataka: Goil rna sreg gi skyes pa'i rabs (ef. Mahavyutpatti [Sakaki] 4892)
VyY-t 124a6
rTsis mkhan [gyi] mdo = MN 107,
VyY-t 171a8
*Gati-sii: 'Gro ba'i mdo
114b3; 116a5; VyY-t 153a2, 3
*Gardabha-sii: Luil boil gi mdo = AN I 229, Gadrabha-su
51b5; VyY-! 3la7 (= AN I 229,11-25)
Garbhavakriinti [-sii]: milal {mal} du 'jug pa
116b3; VyY-t 153a2
Gatha: Tshigs su bead pa
33a6 (see Marokta-); 51b6; 68b7 (= Uv 26:lOd); 73al (= Uv 30:33); 73a5; 84a4;
114a7; 118a2 (= Uv 29:24a); 118a2 (= Uv 29:23a); 123b3; 125a4 (= Uv
29:24a); 128a7 (dGe 'dun phal chen sde pa: Mahasiirp.ghika); 144b6; 150b1
VyY-t 48b8; 75a8 (yathoktaJ!l sthaviriiniruddhena); 100b1, 3; 130a7; 148b3
-ailga: - pa'i sde: VyY-sii 31bS; VyY 97a3; VyY-t 122b4; 123b5 (def.)
Geya, anga: dByans kyis bsnad pa'i sde
VyY-sii 31b5; VyY 40b7; 97al, 2; 140a4; VyY-t 122b4; 122b7 folio (def.)
*Gopaka-sii: Sa mtsho'i mdo (cf. Mahavyutpatti 1071, GopI [pal = Sa 'tsho rna)
= MN 108, Gopakamoggalliina-su
114b3; 115b6 (= MN ill 13,26-32); ef. Samathadeva, milon pa'i bstan beos, thu,
111b4
*Ceta!:tparyayabhijna-nirdesa: Sems kyi mam grans milon par ses pa bstan pa
37a8 (cf. Sanghabhedavastu [Gnoli] II 248,21); VyY -! lOa7
Jati!a-sii: Ral pa can gyi mdo sde = Bimbisarapratyudgamana-sii, P955, Vol. 38
[VyY-sii 20bS]; VyY 55b5
Jataka, anga: sKyes pa'i rabs kyi sde
VyY-sii 31b6; VyY 97a8; VyY-t 122bS; 124a6 (def.). See also Kapifijala-,
Sunetra-.
Tripitaka: sDe snod gsum
VyY-! 124b5, 7
DaSottara[ -sii]: bCur bskyed pa = DN 34, Dasuttara-su
44a4
DIrghala-avadiina: Riil po len gyi rtogs pa brjod pa
97a7; VyY-t 124a2
jIABS 23.2 342
(?): sDug bsilal gyiphuil po ston pa [bstan pa?]
= MN 14,
115a6 (= MN I 92.26)
* (prathama- sDug bsilal gyi phuil po can gyi mdo dati po
= MN 13, Mahadukkhakkhandha-su
88b3; VyY-! l08aS (= MN I 85,22-29); 108b1 (= MN 188,6-13)
*Deva(ta)-sil: Lha'i mdo
73a8; VyY-! 75b3
Dharmamudra-sii: Chos kyi phyag rgya'i mdo
VyY-! 153a3
*Nandaka[-sii]: dGa' byed
116b3
Nidana, ailga: G1eil gii'i sde
VyY-sii 31b5; VyY 97a5; VyY-! 122b5; 124b5, 6
Nipata: Le'u
116b3; VyY-! 1481, 2
*Nyagrodha-sii: Nya gro dha'i mdo
48bl; 137a7; VyY-p4b8; 168a2; 171a7
*Pancailgasthapati-sil: Yan lag lila pa'i phya mkhan gyi mdo MN 78,
S amllI!amllI!9.ika-su
37a6 (= MN II 27,5-10); VyY-! 11b2 (= MN II 26,16-22)
Paramarthasiinyata[ -sii]: Don dam pa stoil pa fiid
124a7; 127b7; VyY-t 159b6
Piin;ta-sii: Gail po'i mdo = SN III 105-6, Ananda-su
114b3; 115b5 (= SN III 105,10-11)
PratItyasamutpada-sil: rTen em 'brei bar 'byuil ba'i mdo
98b2 [98b4, de nid las; 98b7]
Pravacana, dvadasanga: gSuil rab, yan lag bcu grus po
VyY-t 124b5
Bahudhatuka[-sii]: Khams mail po pa = P963, Vol. 38 = MN 115, Bahudhatuka-su
127b1; VyY-! 157a8
Bimbisarapratyudgamana-sii: gZugs can sfiiil po bsu ba['i] mdo = P955, Vol. 38
VyY-! 152b8
dGe sloil ma'i mdo sde = SN V 154-58, BhikkhunIvasaka-su
43a3 (spoken to Ananda); VyY-! 68b6
Yoils su my a ilan las ' das chen po' i mdo
= DN 16, Mahaparinibbana-su
114b3; 116a1 (=MPS 42:17, p. 396); 116a8; VyY-! 147b6
Mahasudarsana-sii: Legs mthoil chen po['i] mdo sde = DN 17, Mahasudassana-su
VyY-! 147b7
Matrka: Ma mo
97b3; VyY-t 124b3
Mi'i mdo
VyY-! 123b5; 153a1
Mayajala[-sil]: sGyu ma'i dra ba
126b1; VyY-! 155b7
SKILLING 343
*Marokta-gatha: bDud kyis smras pa'i tshigs su bead pa
33a6(= SN I, Marasarp.yutta, p. lOS); VyY-t 4a5
Yul 'khor skyon [gil rndo) = MN 82, RanJlapala-su
114b3; 116a3 (cf. MN II 6S,15 foll.)
RiihuIa-sii: sGra gqm zin gyi mdo sde = MN 61, AmbalanJlika-riihulovada-su
71aS; VyY-t 71a4 (= MN I 415,30-416,6); 71a7 (= MN I 416,35-417,1);
71aS (= MN I 417,6-7)
V dEyar tshul [gyi) rndo sde
VyY-t 147b7
Vinaya: 'Dul ba
97a5; 124bl foll.; 129b7; VyY -t 124a5; 124b6; 153a6 foll.
Bye brag tu Mad pa
lOSb5; VyY-t 129b4
Veda: Rig byed
105b2
VaituIya, ailga: rnTshuns pa rned pa'i sde
VyY-t 125a7
Vaipulya, ailga: Sin tu rgyas pa'i sde
VyY-sii 31b6; VyY 97aS; 97b5; 9Sa3; 103b2; l04a4; 113bl; 133a3
VyY-t 122b5; 124a6 (def.); 124bS
VyakaraIJ.a, ailga: Lun du bstan pa'i sde
VyY-sii 31b5; VyY 97a3; VyY-t 122b4; 123a6 foll. (def.)
VyakaraIJ.a: Bya ka ra (non-Buddhist grammatical work)
105aS; 105bl, 2, 6,7; VyY-t 135b7; 136a5
Sastra: bsTan bcos
143a4 (def.; see Bu ston I 42)
Siinyata-sii: sTon pa fiid kyi rndo = P956, Vol. 3S = MN 121, Ciilasuiiiiata-su
114b3; 115b2 (= MN ill 104,7-11); VyY-t 152bS
Gro hiin skyes kyi rndo
94b6
Sarp.yuktagama: Yail dag par Idan pa'i Iun
36b5; VyY-t 9aS (cp. SN ill 51)
Saptabhava-nama-sii: Srid pa bdun zes bya ba'i rndo
VyY-t 153a1
Saptasiiryodaya-sil: Ni rna bdun 'char ba'i rndo sde = AN
VyY-t 126bl, 2
Sujata-sil: Legs skyes kyi {kyis} rndo
VyY-t 122bS
Sunetra-jataka: sPyan legs kyi skyes pa'i rabs
99aS; VyY-t 126bl, 3. See also Sunetra.
Siitra, ailga: rndo'i sde
VyY-sil31b5; VyY 40b7; 97al, 2; 140a4; VyY-t 122b4, 6 (def.)
[Pitaka) VyY-t 124b6, 7
Hastata<;topama[-sil): Lag pa rdob pa Ita bu
102b6; l11a2; VyY-t 132b2; 143bS
Hastipadopama[-sii): gLail po che'i stabs Ita bu = MN 2S, Mahahatthipadopama-su
109b3 (= MN I 191,1S-19)
nABS 23.2 344
Katya'i bu mo'i mdo
VyY-! 78b2
sKye mched so so ba'i mdo (sde)
*****
102b5; 109a2; 109a5; 109b6; 111a1;VyY-!132a4,5; 140b5; 143b4
rGyan can gyi bu'j mdo sde
81a7
Cail mi smra ba'i mdo
116a6
Ci iig gsun ba'j mdo
131b5; VyY-! 162a6, 7
Chu las skyes pa'j mdo (see Mahavyutpatti 6142, Jalaja?)
114b3; 115b4
Nan thos la Mad pa'i mdo sde gills
41a6; VyY-! 14b8
rTa cail ses bion bzails kyi chos kyi mam grails
115a6
gNas pa'i mdo
116a7
Mig chun gi mdo sde
VyY-! 147b7
dMag ldan gyi mdo
VyY-!69a7
bZa' ba Ita bu'i mdo = SN III 86, Khajjanrya-su
98b6 (= SN III 86); VyY -! 126a1
, Od sel gyi mdo
114b3; 115a5
Ri dags zlog gi mdo
132b7 (= Vinaya III 68, SN V 320). Ri dags (= rnrga) zlog is a proper name
equivalent to the Pali MigalaI).<;lika samaI).akuttaka at Vinaya III 68,22 foll. It also
occurs at Vinayavibhanga (P1032, Vol. 42, 119a7 foll. as ri dags zlog dge sbyon
sbed.
Sara'i mdo
49b1
Appendix 5.2. Proper names
AiJ.gulimala: Sor phren
* Acandrikaputra (?): Ma zla ba'i bu mo'i bu
Ajitakesakambala: Mi pham skra'i la ba can
Anathapil).q.ada: mGon med zas sbyin
46b4; 113b8 (= MN II 99)
VyY-! 10b7
VyY-!69a5
(grhapati: khyim bdag = Sailghabhedavastu [GNOLI] I 167,1);
116a5; 131a7
VyY-! 147b2
Aniruddha: Ma ' gags pa
tshe dailldan pa 116a2
(from Mahaparinirv[il).asiitra) arya: 'phags pa 129b7
sthavira: gnas brtan VyY-! 75a7
Asvajit, rTa thul, tshe dailldan pa 84a4
Acarya: Slob dpon VyY-! 131a4, 7; 135b5, 7; 136b2, 3,8; 138a2; 146a8
SKILLING 345
*Acarya-iula: sLob dpon gyi rigs 124a2; 148a1; 153a4
Ananda: Kun dga' bo 35b8 (=MN III 110); 43a3; 114a3 [D] 881);
115b3; VyY-Pb2foll.; 17a6 foll.; 68b7; 147a6; 153a6; 159a4
tshe danldan pa Vy Y -t 192a3
arya-: 'phags pa 114b7, 8; 115a3; 115b2; VyY-t 147a8
sthavira-: gnas brtan - 115b5; 116a1
[Aryadeva, acarya: 'Phags pa lha, slob dpon] 143a
Uttara, bhadanta: Bla rna, btsun pa 114b4, 5 (see Uttara-su)
Uttara, rn1iJ;tava: Bla rna, bram ze'i khye'u (previous birth of Slikyamuni)
Uttarakaurava: Byan gi sgra rni snan pa (mams)
Udayin, sthavira: 'Char ka, gnas brtan
131a6, 8; 131b6; VyY-t 160b7, 8
VyY-t 54a7
VyY-t 78b1
VyY-! 143a6
VyY-t 69a4
VyY-t 143a6
VyY-t 192a2, 6
VyY-! 39b8; 40a2
(= AN I 66, Samathadeva tu 25b3; SutrakhaJ;1q.a no. 22)
VyY-su 20b2, 3; VyY 54a7
Uluka: 'Ug pa
Kakudha Katyayana: Katya'i bu nog can
Kapila:Serskya
KapotamalinI: Thi ba'i phren ldan rna
Katyayana: Katya'i bu
Kalama: sGyu rtsal ses (dag)
KasI: Yul ka {ko} si
Kasyapa: 'Od smns (previous Buddha)
Kasyaplya-nikaya: 'Od smns pa'i sde pa
Kosala: Yul ko sa la
*Gahanavana (?): Thibs po'i tshal
GrdhrakUta-parvata: Ri brgod bya
VyY-! 124a2
130b6; 131a6; VyY-t 160b8
32a7; 54b6 (btsun pa
'od srmis pa'i giun las); VyY-t 153a2
VyY-t 124a2
100b4
Gautama: Gau ta rna (Buddha, addressed by "outsider")
115a7,8
VyY-su30b4; VyY U5b7;
VyY-! 69b1 foll.; 117a8
(- dag) (name of a clan, addressed by Buddha) VyY-su20a4; VyY 52a2
Chanda, 'Dun pa, tshe dan ldan pa (cf. Mvy 9474; BHSD 235a) 64b8
Jambudvlpa: 'Dzam bu'i gliil VyY-su24b7; 114b4 (from Uttara-su)
Trrthika: Mu stegs can 51a8; 100a2; 115b1; 130b7; 136b5; 140a6;
dGa' ldan
DIrghala: Rin po len
See also DIrghala-avadana
Devadatta: Lhas byin
DroI].a, br1ihrnaJ;1a: Bre bo, bram ze
Dharrnaguptaka-nikaya: Chos smn gi sde
Dharrnananda: Chos dga'
Nagaraka-nama Slikyanam nigama
VyY-t 68b8; 123a8; 143a6; 147a5; 161a2
130b6; VyY-t 160b8
VyY -! 124a2 (past King of Kosala)
VyY-t lOb7
40a8 (= AN Il37,26-27)
VyY-t 153a2
143b8 (birth as a chu srin)
Sa kya mams kyi gron rdal gron khyer ies bya ba
NadI-kasyapa: Chu klun 'od smns
115b2
46b4; VyY-! 22a7
nABS 23.2 346
Nanda: dGa' bo
kulaputra: rigs kyi bu
46b4
VyY-su 27b6; 28a3, 6, 8; VyY 85a3;
86a5;VyY-t 101b6 foIl.; 103b7 foil.
gopaIaka: ba lail. rdzi ' 140b8
(avadana in which Nanda crushes a frog with a stick = Gilgit
Manuscripts [DUTI] ill-I 51,1-6)
Nandaka, dGa' byed, tshe dail.ldan pa VyY-t 50b2
Nikaya: sDe pa 116b2, 3; 124b7; VyY-t 147b8; 148a1; 153a2
sarva-: - thams cad 103b3, 5; 124b2; 129b5
- bco brgyad 124a7; 124bl
Nirgrantha: gCer bu pa VyY 1I5a7 (- lag pa bsgren ba dag);
VyY-t 147a4 (mams; addressed); 162a8; 162b2
Nirgrantha Jiiatiputra: gCer bu pa giien gyi bu VyY-t 69a5'
Nyagrodha: Nya gro dha VyY-t 168a2.
See also Nyagrodha-su.
Parivrajaka: Kun tu rgyu (dag)
PuraJ?akasyapa: 'Od sruns rdzogs byed
PUfJ?a-maitraYar:t1putra: Byams ma'i bu gail. po
See also PUfJ?a-su.
39b3
VyY-t 68b8, 69a4
115b5 = SN ill 105,10.
Biihyaka: Phyi rol pa
Brahma: Tshail.s pa
32b5,6; 37a6; 46b2; 47b1, 4, 6; 48a8;
49a7;49b3;50a3,4;50b7;67a2;68a3
VyY-su 19a7; VyY 47a4; 47b1
-loka: 'jig rten
Brahmadatta: Tshail.s pas byin (past King of KaSl)
Bhadravarglya, bZail. po'i sde pa drug cu dag
Mathura: Yul bcom brlag
Maskan GosaITputra: Kun tu rgyu gnag lhas kyi bu
Mahiikasyapa: 'Od sruns chen po
VyY-sil22b4
VyY-t 124a2
44b1
106b7; VyY-t 137b3, 4
VyY-t 69a4
116b2; VyY-t 147b8
gSus po che chen po, tshe dail. ldan pa
43b6; 100b5 (= MN I 292)
Mahanama, Siikya: Min chen {can}, Siikya- 115a6
(= MN I 92,26; cf. *Dul;1khaskandha-nirdesa)
Maharaja: rGyal po chen po 36a7; 115b4
Mahasarp.ghika-nikaya: dGe 'dun phal chen sde pa 128a6 (verses of)
Mahasudarsana, Raja: Legs mthon chen po, rgyal po VyY-sil 28b2; VyY 86b5
Mahlsasaka-nikaya: Sa ston pa'i sde 124a8; VyY-t 153a2
Mahlsasaka-siltra (?): Sa ston gyi mdo 14b3; 115a8
Mahendrasena, raja: dEail. chen sde, rgyal po VyY-t 192a5
Mara: bDud VyY-sil19a7; 20a5, 6; VyY 33a6; 45b8; 47a1, 4; 52b3, 4,5
Muni: Thub (vs) 32al
Maudgalyayana: Maud gal gyi bu 116a7; VyY-t 147b4; 162b4
Rajagrha: rGyal po'i khab 115a7, 8; VyY -t 192a2
Riihula: sGragcan zin VyY-t 74a4 foIl. See also Riihula-sil.
V dEyar tshul 115b6. See also V
VaraJ?asi: Bara J?a si VyY-t 161a3
Vinaya-dhara: 'Dul ba 'dzin paVyY-t 124a5
'Dul {gdul} ba'i bye brag tu smra ba (mams) VyY-t 133b2
Vinaya-saI!lgIti: 'Dul ba yan dag par bsdus pa
Vipu1aparsva-parvata: Ri nos yans
Bye brag tu srnra ba (dag)
SKILLING 347
124b6
Sakra, Devanam-indra: brGya byin, lha mams kyi dban po
115a7
VyY-! 131a8; 146al
114b4; 115a3
(from
Sakya: Sakya 115b2
Sakyamuni: Sa kya thub pa 129b6; 130a3; 132b8; VyY-t 163a8
Sanputra: Sari'i bu VyY-sii 29b8; VyY 84b4; VyY-! 112b3; 113a1; 162b4
tshe dan 1dan pa - 43b6; 100b4; VyY-t lla2
arya: 'phags pa - 84a3; 134a3 (from Arthavistara-sii);
dge sImi.
134b6 (id.); 140a8 (id.); 143b3; VyY-! 100a8; 166a7
VyY-t 162a7
Saradvatlputra: Sa ra dva ti' i bu
Sravakayana: Nan thos kyi theg pa
VyY-t 100a8
114b2; 116bl; 120b6,8;
125b4; 127bl; 129b5,6; 130a7
43b4; VyY-t 123a5
VyY-sii 21b3, 4; 29bl, 2;
VyY 59a6; VyY-t 45b2, 3; 109a6,7
89a7
VyY-sii 30b4; VyY-! 117bl
SravastI: mNan yod
Srol).a: Gro Min skyes
arya- : 'phags pa -
brahmana: bram ze
see also'Srona-siitra
Svetaketu, devaputra: Tog dkar po, lha'i bu (Sakyamuni as bodhisattva in
SaI!lgIti: Yan dag par bsdus pa
-kara: - sdud par byed pa
VyY-! 161al
116b3,4; 124b7
76a8; 115a4; 124b5; VyY-! 84b8
116b2; 124b6; VyY-t 147b8 -pa'i gii po
see Vinaya-
SaI!ljayIVairaqIputra: sMra 'dod kyi bu rno'i bu yan dag rgyal ba can
(Arya-)Samrnatlya-nikaya: 'Phags pa kun gyis bkur ba'i sde pa
Sarviirthasiddha, bodhisattva
VyY-t69a4
124a7
Don thams cad grub pa, byan chub serns dpa'
Sunetra, Sast:r: sPyan legs, ston pa po
VyY-! 161al
99a4; VyY-! 126bl, 3
See also Sunetra-jataka.
Sthapati: Phya mkhan VyY-t llb3
*****
Ka tya'i nu rno srnyug rna can 74a3 (Cf. Mvy 3798 srnyug rna mkhan = vel).ukara)
sGra sgrogs 115b7
sTen {rten} rna (dge sIon ... mams) 115b8; VyY-t 147bl, 2 (= Paiicabhadravarglya)
dMag ldan VyY-sii 23b6; VyY 36bl; 70b3; VyY-t 69a5, 8; 69bl; 85b4
sMra mkhas, khyim bdag VyY -! 123bl, 4
'Dzambu'i grib rna can
= Pall Jambukhadaka
'Od se1: name of monk
01 rna se can (?)
Zas gtsan rna'i sras
Sa ga, tshe dan 1dan pa
VyY-sii 29b8; VyY-t 112b3; 113a2, 6 (= SN N 252-53)
115a5; see - gyi rndo
115b7
129b6; VyY-! 17b4 (Sakya-); 161al (cf. Mvy 24)
VyY-sii 25b5; VyY 76b8; VyY-t 15b8 foIl. (= MN I 301)
nABS 23.2 348
Appendix 6. Mahayana sources and references
Appendix 6.1. Literary references
Prajiiaparamita: Ses rab kyi pha rol tu phyin pa brgyad ston pa '
122a8 (= Vaidya, p. 165,14-20)
UpayakausaIya-sii: Thabs la mkhas pa'i mdo sde
131b6, 7; 132a5; VyY-t 162b5
Gatha: Tshigs su bcad pa
128b5 (from BhavasaI11kranti, P 892, Vol. 35, rndo, tshu 186b2
Tathagataguhyaka: De Min gsegs pa'i gsan ba
138al (= cit. in Mahayanasiltralrup.kara, ed. S. Bagchi, Darbhanga, 1970, pp.
78,5 foil.)
Tathagatotpatti-nirdesa-sii: De biin gsegs pa skye ba bstan pa'i rndo
VyY-t 163b2; 173b6
*Mahayanasiltranta-srup.graha: Theg pa chen po'i rndo sde bsdus pa
134a6
Prajiiaparamita: Ses rab kyi pha rol tu phyin pa
116b7; (de fiid las) 116b8 (= Dutt 21.11-19); 117a4 (= ibid. 23.3-5); 117a5 (= ib.
25.4-9); 117a7 (= ib. 29.14-18); 117b1 (= ib. 30.3-5); 117b2 (= ib. 32.4-15)
VyY-t 133b6
[BhavasaI11kranti-sii: see Gatha]
Mahaprasadaprabhavana-rnahayana-sii:
Dad pa chen po skye ba, theg pa chen po'i rndo (PSI2)
119a7
Lokottaraparivarta: 'Jig rten las ;das pa'i le'u
131b7; VyY-t 162b5
(Arya-)VimaIakIrtinirdesa: (,Phags pa) Dri rna rned par grags pas bstan pa
119a2 (LAMOrrE, p. 106, vs 4)
Satasiihasrika [Prajiiaparamita]: sTon phrag brgya pa
97b6
Srup.dhinirmocana-sii: dGons pa nes par ' grel ba'i rndo
123b2 (LAMOrrE p. 67); 123b4 ( = LAMOrrE p. 80); 129a7 ( = LAMOrrE p. 35)
*Samantaloka-sii: Kun tu snan ba'i rndo sde
119a4 '
Sarvapul).yasamuccaya[-sil]: bSod nams thams cad bsags pa'i rndo
123a1
SiitraIrupkara: mDo sde'i rgyan
155b3
Appendix 6.2. Proper names
Brahma: Tshans pa
118b2,5
Mahayana: Theg pa chen po
97aS (= Vaipulya); 97b5 (id.); 97b6 (def.); 98a3; 103b3; 104a3, 4; 113a8; 113bl,
2; 114aS; 116b5; 118b6, S; 119aS;120a7; 122a6, 7; 122b5; 123a6, 7;123b1;
124a5,6; 125a2,3; 126b3; 128a8; 128b5(vs); 129b4; 130a8; 132b7,8; 133a3;
VyY-t 152b8; 153b8; 156a4; 163b7
SKILLING 349
Mara: bDud
122a8; 122b4,5, 6,7
Ratniikara, bodhisattva: dKon mchog 'bYUIi gnas, byail chub sems dpa'
119a3 (from VimalakIrtinirdesa)
Santamati: Zi ba'iblo gros:
138al (from Tathiigataguhyaka)
Subhiiti: Rab 'byor
103b5; 122a8 (from VyY-t 133b7; 134a5
Appendix 7. Sources of the verses of the GiithiisaIpgraha
124
1) Avadilnasataka (EST 19) 251; 4:112 (Pradhan 267). Cf. Dasabala-
srImitra (Saf!1skrtasaf!1skrtaviniscaya, P5865, Vol. 146) , TIO mtshar bstan bcos,
flo, 37b8; Bu ston (OBERMILLER) I 103.
2) Udilnavarga 15:9-11; Siikarikilvadilna P1014, Vol. 40, mdo u, 302a8-b2.
3) Udiinavarga 28:40.
4) Saf!1yuttanikilya I 133, Therfgilthil200, 201; Mahilvastu I (EST 14) 23-24.
5) Divyilvadilna (EST 20) 103; 7:34 (PRADHAN 416) 1099; Dasabala-
srlmitra, no mtshar bstan bcos fio, 268a4; Samathadeva, Abhidharmakosa-
upayikil!fkil, P5595, Vol. 118, milon pa'i bstan bcos, thu, 103a3.
6) Cf. Saf!1yuttanikilya I 5, Arafifiasutta.
7) Udilnavarga 32:8.
8) Udilnavarga 10:9.
9) Udilnavarga 5:24.
10) Udiinavarga 22:6; Samathadeva, mnon pa'i bstan bcos tu, 51b2; Vyilkhyilyukti,
sems tsam si, 36b3; Bu ston (OBERMILLER) I 9; Bodhisattvapi!aka, Chap. XI
(Ulrich PAGEL: The Bodhisattvapitaka: Its Doctinres, Practices and their
Position in Mahilyilna Literature, Tring (UK) 1995 (Euddhica Britannica Series
Continua V), pp. 341, 344.
11) Udilnavarga 28:1; Samathadeva, milonpa'i bstan bcos, tu, 242a6; Ta chih tu [un
= Etienne LAMOTTE: Le traite de la grande vertu de sagesse de Nagilrjuna
(Mahilprajfiilpilramitilsilstra) II, [Louvain 1949] 1967, p. 1075.
12) Divyilvadiina (EST 20) 53. Cf. Udiinavarga 30:5.
13) Udilnavarga 20:16; Madhyilntavibhilga!fka (ed. Susumu YAMAGUCID, Nagoya,
1934, p.150.13; ed. R.c. PANDEYA, Delhi 1971, 114; c, din Pratftya-
samutpildildivibhmiganirdeia (p5496, Vol. 104), mdo tshogs 'grel pa, chi, 30b8
(giithii).
14) Udilnavarga 30: 11; a in Pratltyasamutpiidiidivibhanganirdesa, mdo tshogs ' grel
pa, chi, 26b2 (giithii).
15) Udilnavarga 28:2 (var. in d).
16) Udilnavarga 33:2.
17) Udilnavarga 10:1.
18) Udilnavarga 1:5; Divyilvadilna (EST 20) 561.8-9.
124. The enumeration follows the commentary, the Gilthilrthasaf!1graha-silstra,
which treats some as groups. It is therefore different from the numbering in
ROCKIDLL & SCHIEFNER. For further occurrences of the verses consult the
appropriate verse in BERNHARD's Udiinavarga.
nABs 23.2 350
19) Udiinavarga 1:3.
20) Udiinavarga 1:22.
21) Not traced.
Appendix 8: SigJa and references for the Vyakhyayukti texts
(1) Vyiikhyiiyukti = rNam par Mad pa'i rigs pa, by Vasubandhu, translated from
Sanskrit into Tibetan by Visuddhasirpha, Sarvajiadeva, and
P: Peking Tanjur 5562, sems tsam si, 31b8-156a5 (Repr. Vol. 113, 244.5.8-
294.4.5)
D: Derge Tanjur 4061, sems tsam si 29a2-134b2
G: Golden (Ganden) Tanjur, sems tsam si (Repr. Vol. 66, A22-99)
(2) Vyiikhyiiyuktisutrakha/yJasata = rNam par Mad pa 'i rigs pa 'i mdo sde'i dum bu
brgya, by Vasubandhu (dByig giien), translated from Sanskrit into Tibetan by
Visuddhasirpha, Sarvajiadeva, and Maiijusrlvarman:
P: Peking Tanjur 5561, sems tsam si, 19a2-31b7 (Repr. Vol. 113, 239.5.2-
244.5.7)
D: Derge Tanjur 4060, sems tsam si, 17bl-29a2
G: Golden (Ganden) Tanjur, sems tsam si (Repr. Vol. 66, A12-21)
N: Narthang Tanjur, mdo si, 18al-29a1
(3) Vyiikhyiiyukti!fkii = rNam par Mad pa'i rigs pa'i Mad pa, by GUI;Lamati (Yon
tan blo gros), translated from Sanskrit into Tibetan by Visuddhasirp.ha,
Sakyasirpha, and
P: Peking Tanjur 5570, sems tsam i, 1a1-194a6 (Repr. Vol. 114,95.1.1-173.5.6)
D: Derge Tanjur 4069, sems tsam si, 139bl-301a7
G: Golden (Ganden) Tanjur, sems tsam i (Repr. Vol. 66, Cl-120)
Colophon
Although the product of over twenty years of devotion to the Vyakhyayukti literature,
the present article is a compilation of information culled at various times for various
reasons, rather than a thorough and comprehensive study. Thus I feel more like one
who has periodically plundered an ancient, overgrown stupa of a part of its treasures
than a careful drchceologist who has systematically laid bare it structure in full detail. I
aver without reluctance that the Vyiikhyayukti is Vasubandhu's most brilliant work,
far surpassing the already brilliant Abhidharmakosa, and that for historical,
philological, and philosophical purposes it is the most important single and
compact
125
Indian sastra in Tibetan translation awaiting study and translation. It is
with this somewhat missionary zeal that I put forward this paper, incomplete and
imperfect as it is, in the hope that it will generate interest in the important but hitherto
ignored work, and serve as a guide to at least some of its treasures.
125. I have used these two adjectives to exclude that other mine of treasures, the vast
Yogiiciirabhami literature associated with AsaIiga, which equally deserves a full
study and translation.
The XIIIth Conference of the International Association
of Buddhist Studies
. .
8-13 December 2002
Bangkok, Thailand
First Circular: October 2000
The 13th conference of the International Association of Buddhist Studies
will be held in Bangkok, Thailand, from Sunday the 8th to Friday the
13th December, 2002. Chulalongkorn University will host the
conference.
Organizing Committee
Honorary President: HRH Princess Maha Chakri Sirindhorn
Chairperson: Dean, Faculty of Arts
Deputy Chairs:
Director, Buddhist Studies Center
Associate Dean for Research, Faculty of Arts
Secretary General: Assistant Professor Prapod Assavavirulhakarn
Assistant Secretary: Peter Skilling
Board Members:
Chair of Department of Eastern Languages
Chair of Department of Philosophy
Prof. Wit Wisadevet
Prof. Preecha Changkhwanyeun
Assoc. Prof. Suwanna Satha-anand
Martin Perenchio
Academic Programme
The programme will consist of panels and sections. Each panel will be
organized and supervised by its convener. The sections will be arranged
by the Organizing Committee of the conference.
Journal of the International Association of Buddhist Studies
Volume 23. Number 2.2000
nABS 23.2 352
1. Panels
The following panels are being tentatively offered:
Louis Gabaude [gabaudel@loxinfo.co.th]:
Preaching and teaching the Buddha's message within modernity:
Evolution and diversity offorms.
Pierre Pichard [pichboud@sac.or.th]:
The Buddhist monastery: Sharing roles among Sangha, King and
Laity.
Cristina Scherrer-Schaub [Cristina. Scherrer -Schaub@orient.unil.ch]:
Power of words and words of power: Satra and Dhtlral}f and their role
in the transmission of Buddhism outside India.
Prapod Assavavirulhakarn [prapod.A@Chula.ac.th]:
A Trans-textual Tradition?
Pat Chirapravati [pchirapravati@asianart.org]:
Expressions of Buddhism in pre-modem South-East Asia.
Shoryu Katsura & Tom Tillemans: [katsura@hiroshima-u.ac.jp]:
Buddhist Logic.
Rob Gimello [gimello@fas.harvard.edu]:
The Practice of Scripture Commentary in Mediaeval Chinese
Buddhism.
Haiyan-Hu von Hinliber:
Buddhist Studies in China.
Richard Salomon [rsalomon@u.washington.edu]:
Further Discoveries and Studies of Gandharan Buddhist Manuscripts.
Jens Braarvig and J.-U. Hartmann:
Buddhist Manuscripts: Recentfinds.
Charles Muller [acmuller@human.toyogakuen-u.ac.jp]:
The Changing of the Medium: Trends and Models in Digital-Based
Buddhist Research.
Florin Deleanu [fwjb8913@mb.infoweb.ne.jp]
New Discoveries of Buddhist Manuscripts in Japan.
2. Sections
The sections will run parallel to the panels. Papers will be 20 minutes
long, followed by 10 minutes for questions.
If you wish to present a paper in the sections you are requested to
prepare an abstract of about 600 words and send it to the organizers of
the conference. Abstracts may be sent bye-mail or in printed form.
lABS 2002 353
The abstract must state clearly the primary sources used, the problem
to be investigated, and the approach to be followed. Papers must be up
to international academic standards and make a valuable contribution to
Buddhist studies.
The deadline for submission of abstracts is 30 September 2001. The
Organizing Committee reserves the right to accept or reject a paper on
the basis of the abstract and to assign it to a particular section.
General
Chulalongkorn University
Chulalongkorn University was founded in 1915 by King Rama VI in
honour of his father, King Chulalongkorn (Rama V). The conference
will be held on the university's quiet green campus in the heart of
Bangkok. The month of December is the beginning of 'winter' or the
'cool season', when temperatures plummet to 35 degrees.
Conference fees
Pre-registration (before 30 September 2002): SUS 180 .
Registration (from 1 October 2002): SUS 200.
This includes conference fee, lunches and tea-breaks during the
conference, reception, and an excursion to the ancient capital of
Ayutthaya with banquet. (It does not include hotels). The method of
payment will be announced shortly.
Accommodation
Accommodation within walking distance will be available from about
US$20 to $300. A list of hotels giving special conference rates will be
provided on the website and in the second circular. Participants must
contact the hotel of their choice and make their own arrangements.
Food
Vegetarian and non-vegetarian lunch will be provided during the
conference (from Monday 9 December to Friday 13 December). The
lunches and the banquet are included in the registration fee.
nABS 23.2 354
Visa
Visa requirements vary according to the nationality of the visitor. Please
check with your travel agent or with a Thai consulate or embass-y to be
sure you get the right type of visa, if required.
Second Circular
A second circular will be sent to all those who return the enrolment
form enclosed with this first circular, or to those who request it by
e-mail. The second circular will be sent in January 2002.
Correspondence
All correspondence relating to the conference should be addressed to:
IABS 2002
Faculty of Arts
Chulalongkorn University
Bangkok 10330
Thailand
Timetable at a glance
Submission of preliminary form: 31 January 2001
Submission of abstracts: 30 September 2001
Second Circular: January 2002
Pre-payment: before 30 September 2002
Conference: 8-13 December 2002
The Organizing Committee reserves the right to m.ake changes to the
programme as it sees fit.
JIABS 23.2 355
-new in paperback-
The American
Encounter
with Buddhism,
1844-
1
9
12
Victorian Culture and
the Limits of Dissent
THOMAS A. TWEED
With a New Preface by the Authar
"Indispensable reading for anyone interested in the history
of Buddhism in America." -Jan Nattier, Indiana University
"An important contribution to the sometime rancorous
debate about what American Buddhism ought to look
like and how it can best be adapted to American culture
in ways that will insure both its success in the future
and its integrity." - Richard Hughes Seager, author of
Buddhism in America
"One of the finest books I have read in a long time .... I
find the account particularly worthwhile for its sensitivity
to how the American Buddhists and their Asian allies worked
to shape their account of Buddhism to address the core
value concerns of Victorian America .... I highly recommend
it." -William H. Swatos, Jr., Sociology of Religion
280 pp. $16.95 paper
jJublisbiny excellence since 1922
THE UNIVERSITY OF NORTH CAROLINA PRESS
AT BOOKSTORES OR [800]8486224 I www.uncpress.unc.edu
JIABS 23.2 357
Guidelines for Contributors to JIABS
Two paper copies of articles for publication should be submitted to the
editors. Material in English, French and German will be considered for
publication, with the proviso that a summary in English should accom-
pany manuscripts written in French or German.
For Sanskrit and Pali, use the standard system of transcription as given
in, for example, A.L. Basham, The Wonder that was India, Appendix X.
Transcription of Chinese should preferably be according to pinyin, and
Tibetan should be in the system of the American Library Association -
Verein Deutscher Bibliothekare, using n, ii, z, S, rather than the ng, ny,
zh and sh of the Wylie system. Hyphenation may be used for Tibetan
proper names, if wished, but should not be used elsewhere. Japanese
should be transcribed according the system to be found in Kenkyiisha's
New Japanese-English Dictionary.
In addition to the mandatory two paper copies, the JIABS requests
contributors also to send their articles on computer disks. Please clearly
label the disk with format, name(s) of relevant files, and the word-
processing program used to create the files. In addition, please translate
the article to ASCII code on the same disk if possible. For more infor-
mation on computer matters, contact the editors.
The International Assocation of Buddhist Studies
Colette Caillat
President
Gregory Schopen
Vice-president
Oskar v. Hiniiber
General Secretary
Cristina Scherrer-Schaub
Treasurer
Regional Secretaries:
Janet Gyatso (Amherst, MA, USA)
S. Katsura (Hiroshima)
Peter Skilling (Bangkok)
D. Seyfort Ruegg (London)
Members of the Board:
R. Buswell (Los Angeles), H. Durt (Kyoto), R. Gupta (Shantiniketan),
K. Kimura (Saitama), D. S. Lopez, Jr., (Ann Arbor), A. Macdonald
(Paris), D. Seyfort Ruegg (London), E. Steinkellner (Vienna),
T. Tillemans (Lausanne), A. Yuyama (Tokyo)
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. Subscriptions to the nABS for libraries and other institutions are $70. Dues
and sUbscriptions may be paid in the following ways:
1. by direct bank to bank transfers in US dollars or in the equivalent amount in
Swiss francs. The transfers should be to the IABS accounts at the Banque Cantonale
Vaudoise, Pl. St-Franc;:ois, CH-1001 Lausanne, Switzerland (SWIFT CODE:
BCVLCH2L, clearing no. 7677), account numbers, 983.51.04 for US dollars and
983.51.02 for Swiss francs.
2. by Visa or Mastercard.
3. by cheque payable to the "Association internationale d'etudes bouddhiques
(IABS)." There will be no supplementary charges for cheques drawn in Swiss
currency on Swiss banks. Otherwise, please add 10 dollars or 15 Swiss francs to cover
our processing charges.
4. by Eurocheque in Swiss francs.
Prospective members from developing countries may contact the Treasurer concerning
subsidized membership rates, Dues and subscriptions are payable per calendar year by
December 31 of the previous year. Payments other than direct bank transfers should be
sent to Professor Cristina Scherrer-Schaub, Section de langues et civilisations
orientales, Universite de Lausanne, BFSH 2, CH-1015 Lausanne, Switzerland.
Abstracts of the articles published in this issue available at: http://www-orient.unil.ch