Introduction: Defense from What?
have been many discussions in India, where Yogacara originated,
and elsewhere. Inevitably it was misrepresented and the
misrepresentation repeated and copied. Now, this situation is much more
complicated than would first appear. In India there was a practice
starting from the Veda of memorizing texts by repetition and
transmitting them orally. Much later in the A.D. period there was a
writing down of such scriptures, and then a copying of them. The
Buddhists transmitted their scriptures in a comparable manner, and the
written-down canon was also copied. Such scriptures--whether Hindu or
Buddhist--are regarded by the respective groups as conveying truth.
Therefore, we cannot argue against copying itself. Clearly, it is what
is copied that concerns us now, namely, whether or not there are
as concerns `misrepresentations', it could be asked: What is wrong
with them? Well, some are good-natured and some are rather venomous. In
general there is a great range of such, extending from speculations and
guesses to downright lies, that are repeated and copied. In the case of
misrepresentations of the Yogacara, I accept them as usually of the
good-natured kind, whether it be a type of refutation in a Hindu
commentary, or by a Buddhist opponent of the Yogacara. An important
illustration of the Hindu type of refutation comes from the commentaries
on the Brahmasutras, where the author, say Sankara, can write in terms
of the system itself, setting forth its tenets, or attempts to express
certain tenets, while distorting an opposing system.(1) These Hindu
systems and their rival systems were established by geniuses or
otherwise brilliant persons. It might have been thought better that
these rival systems not be presented correctly, or followers might get
confused, wondering if their own system was indeed better, or whether a
certain tenet being refuted was not really identical with a tenet of
their own system. Therefore, when the Buddhist system supposedly being
refuted in such a Hindu commentary can reasonably be identified with a
sort of Yogacara position, it seems fair to conclude that it is a
good-natured type of refutation--one necessary for a successful
the same judgment could be rendered for the quarreling between
the Hindu logic school and the Buddhist logicians, which at least in
modern times is acknowledged to have sharpened the arguments on both
sides, so that when Buddhism left India around A.D. 1200, the Hindu
logic school began to stagnate for lack of intelligent opposition. I
interpret also the Madhyamika Buddhist refutation of the Yogacara
philosophical position to be a good-natured type--probably more an
exercise in discourse, with the premise that some tenets can be
expounded in themselves and certain others can be clarified by the
refutation of an opposing position. The Madhyamika-type refutation of
the Yogacara regarded its `mind-only' (cittamatra) teaching as a denial
of the external world, and further took its `store-consciousness'
(alayavijnana) as an unwarranted addition to the traditional list of
then, that these sorts of refutations are not made out of
malice, one must face the problem of what there is to defend against on
behalf of the Yogacara. I claim that this system itself is its best
defense. After all, Asanga, the founder, wrote the huge Yogacarabhumi
and the Mahayanasamgraha, while Vasubandhu, besides his commentary on
the latter, as well as on the Dasabhumika-sutra and on the
Madhyantavibhaga, popularized the system in abbreviated treatises. There
is no lack of texts explaining this system, whether in an argumentative
or in a nonargumentative fashion. Therefore, the role of the present
writer to defend the system really amounts to exposing it, taking it as
the important thing, with ancient and modern writers less important in
comparison. Accordingly, the defense of the system against anyone is
really the system's own response; and the role of the present writer is
to find this response of the system itself.
what is meant by this kind of attitude, let us take the
case of a review article I wrote for Philosophy East and West on Ashok
Kumar Chatterjee's book expounding the Yogacara. I concluded:
If Chatterjee's "Yogacara" is indeed the Yogacara person that
Vasubandhu was, then Chatterjee's book is certainly a wonderful
exposition of the Yogacara philosophy. But, if the Yogacara
fundamentals are what I have indicated above, happening to be in
rough agreement with [P. T.] Raju and with [Surendranath] Dasgupta,
the Chatterjee book is still worth reading as a philosophical exegesis
of what was traditionally held, principally by non-Yogacarins, to be the
I wrote this, it was already known to me--though not mentioned in
the review--that Chatterjee had adopted the very interpretation that his
teacher T.R.V. Murti had espoused in chapter 13 of his well-known book
on Madhyamika.(3) As has already been acknowledged, there is nothing
wrong with this kind of copying of a teacher's position. But it also
shows that if what was copied was in fact incorrect, the copy is also
incorrect. Still, we do not know yet if Murti was incorrect in that
chapter. The present essay will deal with this matter later on.
example is the more recent work by Schmithausen on the
Yogacara's alayavisana teaching, wherein Schmithausen disagreed with a
Japanese scholar, H. Hakamaya, who insisted that Asanga was the compiler
of the entire encyclopedic Yogacarabhumi. Here Schmithausen thought to
counter Hakamaya's position by noting the use of the scripture
Samdhinirmocana in some parts of the larger work and not in other parts.
About that scripture, Schmithausen concluded: "Therefore, the
Samdhinirmocanasutra, at least the portions concerned with the new kind
of vijnana distinguished from the ordinary six, was most probably
composed before the Viniscayasamgrahani but after the Basic Section of
the Yogacarabhumi."(4) It happens that a Japanese student, H. S. Sakuma,
completed a dissertation at Hamburg, approved by Schmithausen, wherein
he presented (in German) this very position stated above--hence copied.
In my review of this published work I pointed out: "One may refer in my
Analysis of the Sravakabhumi Manuscript,(5) 110-11, to a passage on the
three doors of vipasyana. This passage is virtually the same as is found
in the Samdhinirmocanasutra, ch. 8, sect. 10. This is a proof that the
author of the Sravakabhumi-agreed to be the oldest part of what Sakuma
calls Mauli Bhumi [the `Basic Section']--had available and used the
Samdhinirmocanasutra."(6) This, then, is an example of where the copying
turned out to be incorrect, and was defended against by calling
attention to what the Yogacara itself says about the matter.
two examples should clarify what is meant by misrepresentations
(or possible ones) that are defended against--not really by the present
writer as an independent thinker, but rather by the Yogacara system
itself being brought to bear upon the alleged misrepresentation. This
essay continues with two main sections below: (1) The Position of the
Yogacarins and (2) Clarification of the Position.
The Position of the Yogacarins
I present four topics: (a) Vasubandhu and an early scripture; (b)
about cittamatra; (c) about alayavijnana; and (d) about three laksana.
and an Early Scripture. Some authors treat the Yogacara
system as though it were an invention of the founders, notably the
brothers Asanga and Vasubandhu. To suggest otherwise, may we notice that
Bhikkhu Nanananda has put out a booklet on the Kalakarama Sutta.(7) This
scripture is in the Anguttara-Nikaya, Book of Fours, where it was
translated by Woodward,(8) who noticed certain textual difficulties. The
Bhikkhu used some other editions and studied Buddhaghosa's commentary on
every word for his English rendition, wherein he obviously uses
Woodward's words whenever possible. In the following I have substituted
my rendition `gods and humans' and subdivide the translation with brief,
The Kalakarama Sutta. [The Setting.] At one time the Exalted One was
staying in Saketa in Kalaka's monastery. There the Exalted One
addressed the monks, saying: "Monks." "Revered Sir," replied those
monks in assent. [As a Vedic or Greek god would talk, declaring
omniscience,] the Exalted One said: "Monks, whatsoever in the world with
its gods, Maras and Brahmas, among the progeny consisting of recluses
and brahmins, gods and humans--whatsoever is seen, heard, sensed
(sensations arising from taste, touch and smell),
attained, sought after and pondered over by the mind--all that
I know. Monks, whatsoever in the world . . . of gods and
humans,--whatsoever is seen, . . . by the mind,--that have I fully
all that is known to the Tathagata, but the Tathagata has
taken his stand upon it. [Now he denies alternatives to be construed as
in number, because the scripture is in the Book of Fours.] If I
to say: `Monks, whatsoever in the world ... of ... gods and
humans--whatsoever is seen ... by the mind--all that I do not know'--it
be a falsehood in me. If I were to say: `I both know it and know
too would be a falsehood in me. If I were to say: `I neither
nor am ignorant of it'--it would be a fault in me. [The Buddha now
how one uses the senses while avoiding the `apprehender' and the
`apprehended'.] Thus, monks, a Tathagata does not conceive of a visible
as apart from sight; he does not conceive of an `unseen', he does
of a `thing-worth-seeing', he does not conceive about a seer.
does not conceive of an audible thing as apart from hearing; he does not
of `an unheard', he does not conceive of a
does not conceive about a hearer. He does not conceive of a thing to
sensed as apart from sensation; he does not conceive of a
`thing-worth-sensing', he does not conceive about one who senses. He
does not conceive of a cognizable thing as apart from cognition; he does
conceive of a `thing-worth-cognizing', he does not conceive about
cognizes. Thus, monks, the Tathagata, being such-like in regard to
phenomena seen, heard, sensed, and cognized, is `Such'. Moreover, than
who is `Such', there is none other greater or more excellent, I
[Now verses on how ordinary persons use their senses, sometimes
called `the fantasy of normalcy'.]
Whatever is seen, heard, sensed or clung to, is esteemed as truth by
Midst those who are entrenched in their own views, being `Such' I hold
none as true or false.
barb I beheld, well in advance [i.e., at the foot of the Bodhi
whereon humans are hooked, impaled.
"I know, I see, `tis verily so"--no such clinging for the Tathagatas.
scripture clarifies the Buddhist prescription "to see things as
they really are," since it implies that one should simply see without
adding anything. Recall the ancient Hermes epigram found on a Grecian
urn: "Who shall say more, will lie."(9) As to the `barb' of the verse,
the Pali is sallam, equivalent to the Vedic word salya, which down the
centuries means an `arrow', on which one is impaled (if such be the
case). I have cited this scripture in this essay because I do believe
that Vasubandhu's popularizing treatises have such a scriptural source
in the background.
Cittamatra. Past writings on this topic uniformly render the term
cittamatra as `mind-only', and so do I also in my own former essays.
Notice that such renditions take the topic outside India by way of Asian
translations and essays in English and European languages. Inside India,
where the term originated, the words citta and matra appeared as such
with connotations of Sanskrit words; outside, there was the connotation
of the words `mind' and `only'. I propose to consider these words with
their connotation in the Sanskrit language.
current lexical work I have learned various usages of the term
matra.(10) The lexicons of India recognize a neuter form matram and a
feminine form matra, so the adjectives can go with the neuter or the
feminine. Of the two definitions for the neuter form--`all, the
entirety' (kartsnya), and `restriction to the instance'
(avadharana)--the adjective `only' could agree with 'restriction to the
rendition `only' works for the other definition--the entirety. For
example, there is the compound sthanamatra, in the meaning `a place in
general', thus any and all places and excluding what is not a place.
When cittamatra is understood in this way, the citta is unmixed with
anything that is not citta. Accordingly, if we suppose in the compound
cittamatra that matra means `only', then is only one of the two senses
intended, or can it be both?
is well to point out that other adjectives are feasible from the
feminine matra. The definition `any measure' (mane) yields the entry in
Apte's Sanskrit-English dictionary paromatra (`vast') for the spatial
measure and, in the lexicons, tatkalamatra (`at once') for the temporal
measure--and also alpa (`a trifle') as well as aksibhaga (`a mirror').
As adjectives applied to cittamatra, we could say `amounting to
mind'--both spatially and temporally. Or we could say `just mind' and
`mirroring mind', and, for the latter, perhaps also `being mirrored by
a previously published essay, I presented Asanga's statement on
cittamatra, which I translated in part.(11) It is clear that the
Buddhist opponents did not criticize on the grounds that Asanga denied
the existence of the external world (which, of course, he did not do).
As the first attack, the opponent states: "It is not valid that there is
a mind-only in the sense of a `continuous substantiality' (dravyatas),
because it contradicts scripture." The opponent is asked: "How does it
contradict scripture?" That person responds: "He (the Buddha) said, `If
the citta consisted of lust defilement (upaklesa) and consisted of
hatred and delusion defilement, it could not become liberated.'" Asanga
replies: "But what is the objection to that?" He seems to mean that we
accept what the Buddha taught, and so if the citta does not consist of
these defilements (or contain them), it would be liberated; hence your
scriptural appeal cannot deny to mind-only a `continuous
substantiality'. The opponent does not give up, and retorts: "Mind-only
by itself is invalid, because if there is not two together, when one
does not resort to representation (vijnapti) of lust, etc., one would be
free (of those defilements) [which we know is not the case]."
opponent is obviously a follower of the Buddhist Abhidharma, which
teaches that there is no citta without a caitta (= caitasikadharma)--a
`mental'.(12) That we cannot have a citta by itself is the second
meaning I treated above, the `totality' sense, all citta, unmixed with
anything else. But Asanga appears to espouse this `totality' sense. His
response starts by saying "There is no fault in a prior representation,"
There is what was said by the Bhagavat, to wit, "concomitant (sahaja)
feeling (vedana), idea (samjna), and thinking-volition (cetana)," and
what was said (by Him), to wit, "These natures (dharma) are mingled,
not unmingled, so these natures are not objects individually separated
out; or when separated out (not objects) for reference as distinct, or
clear, or different." To demonstrate the meaning of the mingling he
used the simile of the light of a butter lamp.(13) Accordingly, if they
were not concomitant, it would also have been improper to say they
cited a scripture in Buddhist Sanskrit equivalent to the
Mahavedalla-sutta of the Pali canon Majjhima-Nikaya. Apparently for
Asanga, concomitance does not deny a pure citta any more than it denies
a pure feeling, and so forth. Ordinary thinking is not able to separate
out the individual factors, because they are mingled. Presumably, it
takes a yogin to separate the mental items. Then this yogin can arrive
at a pure citta. The Patanjali Yogasutra seems to have a similar idea
when it refers to `cessation of the modifications of the citta'
I shall deal with three subtopics: (1) the phrase cittamatram yad
uta traidhatukam, (2) the Yogacara theory of ekagracitta, and (3) the
theory that Vasubandhu denies the external world.
Phrase Cittamatram yad uta traidhatukam. There is a rather famous
passage in the Buddhist Dasabhumika-sutra, its Sixth Stage (bhumi). This
has been cited as cittamatram yad uta traidhatukam, and the scripture's
translation by Megumu Honda, as revised by Professor Johannes Rahder,
understands this to mean "This triple world is mind-only."(14) Sylvain
Levi, in a learned note at the beginning of his French translation of
Vasubandhu's Vimsatika (the Twenty Verses), presents several versions of
this formula, in each case showing the form traidhatukam, and he appears
to recognize that the Dasabhumika-sutra is the source of such
nonscriptural citations of the entire formula.(15) This passage has
seemed to support the claim that the Yogacara denies the existence of
the external world, as here, `three worlds'--of desire, form, and the
formless worlds, according to the usual Buddhist dogmatics.
one of the difficulties of the usual translation and consequent
interpretation is that this scripture, the DaKabhumika-sutra, is not
really a Yogacara scripture. Indeed it is a basic scripture of Mahayana
Buddhism and is not devoted to the particular philosophical view of the
Yogacara. The particular phrase was inserted by that scripture within a
discussion of Dependent Origination, which is important to all Buddhist
schools. Vasubandhu, in his great commentary on the Dasabhumika-sutra,
gives two explanations for this phrase, neither of which promotes any
particular theories of the Yogacara school, even though the first
explanation uses the Yogacara theory of multiple vijnanas.(16) This
first explanation goes into the Buddhist theory of waywardness (vipary
asa), and then announces that the `mind-only' passage was promulgated so
that one may be liberated from the `store consciousness' (alayavijnana)
and from other perceptions (vijnana). Here, the sense of matram seems to
be `amounting to', that is, `amounting to mind', where the `mind'
(citta) here stands for all the vijnanas, counted as seven or eight.
These are the perceptions (vijnanas) based on the five outer-directed
senses, the manovijnana based on the mind (manes) as an inner sense
organ, and the `store consciousness' (for 7), or number 7, the `defiled
mind' (klistamanas), plus the `store consciousness' for 8.
of Madhyamaka persuasion had been commenting, probably only
the first six vijnana would be mentioned. Therefore, it is not a
particular Yogacara teaching here that is meant, but simply that these
various perceptions are what lead to waywardness (viparyasa) and then to
rebirth according to the precepts of Dependent Origination. Vasubandhu
was entitled to interpret the term cittamatra by the set of vijnana
because the Abhidharma gives the terminological set citta, manes,
vijnana, which allows their mutual substitution in certain contexts;(17)
and Candrakirti's autocommentary on his Madhyamakavatara also changes
the cittamatra of the famous formula about the `three worlds' to
vijnanamatra in his section attempting to refute the `store
second explanation has to do with the formula of Dependent
Origination, and here the `thought' is that of the Buddha, who realized
the formula with just one thought (ekacitta) and then taught it in a
twelvefold way. Here, the meaning of matram is the temporal sense `at
once' or in temporal sequence when the twelve members are taught one
after another, thus taking time, eventually to have the theory that the
twelve amount to three lives.
this information from Vasubandhu shows that the rendition of
traidhatukam as `three worlds' is incorrect. The Sanskrit term is a
derivative noun from `three worlds' (tridhatu). The derivative nouns of
Sanskrit have to be interpreted by each such term. For example, Gautama
is the derivative of Gotama; this is a family-type of derivative such
that Gautama is the descendent of Gotama. The derivative paurusa from
purusa (a human person or man) applies more to the present case, because
as a masculine noun paurusa can mean `human action' (karman) and `the
weight that one man with both hands can raise upwards'
(urdhvavistrta-dohpaninrmana).(19) We can also interpret traidhatukam in
two ways. Both ways take the derivative here to be the formula of
twelvefold Dependent Origination, that is, that it is derived from, or
faithful to, the three worlds, so realized in one moment's thought by
the Buddha and then taught in twelve terms. This one-moment's thought is
a variety of cittamatra. The other interpretation of the twelvefold
formula is that it is the way of rebirth through waywardness due to the
set of vijnana, and this is a different interpretation of cittamatra.
But rebirth requires--that is, is related to--the three worlds. Neither
of Vasubandhu's explanations in this commentary justifies the
interpretation that the passage, so misrendered--to wit, "This triple
world is mind-only"--has anything to do with denying the existence of
the external world.
natural question, then, is how should that Sanskrit phrase be
rendered? I would offer: the derivative of the three worlds is only
mind. That is to say, whatever may be the `three worlds' in a minimal
sense--whatever else is attributed to them, an elaboration of them, a
product of them--has been added by the mind. In one case it was added by
the mind of the Buddha--the twelvefold formula of Dependent Origination.
In the other case it was added by the mind of other sentient beings.
Hence, such an observation agrees with the scripture translated above,
the Kalakarama Sutta, which gave the message that to see things as they
really are, one must not add anything. Apparently the six senses (when
normal) see things as they really are, and it is the perceptions (vi
jnana) based thereon that do the adding. But this is the interpretation
of cittamatra when it is tantamount to the set of vijnana. This shows
that it was proper for me to cite that scripture as a background of
the large chapter 3 of Vasubandhu's Abhidharmakosa and his own
commentary is devoted to the worlds, which, when in two sets, are the
world of sentience (sattva-loka) and the support world
(bhajanaloka).(20) He was therefore very familiar with the respective
attributions of these two kinds of worlds. He would undoubtedly know of
the scripture in the Digha-nikaya, III, called Agganna-sutta (though in
the Buddhist Sanskrit Agama version)--a scripture on the Buddhist theory
of genesis. One may read the account in the translation of the Pali
version that, after a long period, this world passes away. This is the
Indian theory of cycles, with the passing away and emergence of the
`support world'. The scripture mentions that with the passing away the
sentient beings had retreated to a `higher' realm--that of Form
(rupa-dhatu)--and while the world was plunged in watery darkness, they
were `made of mind', fed on joy, and were self-luminous. The earth
reappeared like a scum on the cooling water and became endowed with
color, odor, and taste. The sentient beings tasted this and found it
very sweet. As they ate more and more, they gradually lost their
self-luminance, while the moon and sun became manifest. The beings had
evidently fallen into the realm of desire.(21) In such an account we see
the prior disappearance and reappearance of the support world, and that
the sentient world follows suit. Accordingly, the way of translating
that phrase about cittamatra and the traidhatuka that I have had to
discard, namely that the three worlds are dependent upon citta, would
have been in direct violation of the Genesis story. I don't believe that
Vasubandhu would have taken a position in his brief popularizing works
in direct opposition to the Buddhist scriptures.
Yogacara Theory of Ekagracitta. It was pointed out above that the
Buddha was credited with realizing the entire formula of Dependent
Origination with "one-moment's thought." In that place, the term was
ekacitta. The theory of Buddhist meditation sets forth a goal called
samadhi, which is defined as ekagracitta. This term has frequently been
rendered "one-pointed thought (or mind)." Then what is meant by the
Sanskrit word agra? Since this is an important term in the Sanskrit
language, the Indian lexicographers have had to make it a defined word.
Over four of their lexicons include for it the definition alambana.(22)
Literally, this means a `support'. Many years ago I learned that this is
the basic term in Buddhism for the `reflected image' in the mind,(23) w
hich is what one should meditate upon, given that it is an appropriate
object for such meditative purposes. It follows that such a definition
permits a translation for the entire compound ekagracitta, namely "mind
on a single meditative topic." Also, see Vasubandhu's Abhidharmakosa,
chapter 1, for a similar distinction between the outer sense object
(visaya) and the mental reflection (alambana).(24)
three of these four lexicons also defined the expression
ekagra, namely as "a single continuity" (ekatana) and as "being without
discord" (anakula). So the expression ekagracitta can be further
explained for the practice of a yogin, to wit, "the mind or
consciousness as a single continuity and as without discord."(25) The
two definitions go together, because the continuity would be broken if
the mind were subjected to discord. And in such a case there would not
be a "single meditative object." This is an implication of the
definition I found, namely ekatana, because this contrasts with a
well-known Buddhist term, samtana, which is usually understood as a
"stream of consciousness," but which the foregoing information suggests
should be better rendered as a "mingled stream of consciousness." I
should call attention to the fact that the prefix sam- in Sanskrit is
cognate with an Into-European prefix that is present in English as con-
in the sense of "together," that is, that in the ordinary case of a
samtana, the continuity (tana) is composite.
when a modern Tibetan author, Lozang Tsewang, in a published
seminar paper, writes, "Lord Buddha says in the sutras: `The
consciousness of a sentient being is of a single stream'"(26) (but does
not identify the sutra so saying), the cited passage might mean what I
called attention to above, that although the mental factors are mingled,
a yogin--if successful--may separate out the citta and find it
consisting of a single stream. But it is important to notice that the
yogin does this with his own mind, and because he is engrossed in a
samadhi. He does not do this separating out in the minds of other
sentient beings, who cannot verify that the consciousness is of a single
stream, for the reasons that the mental factors are mingled and that
they have not learned how to go into the appropriate samadhi. This
Tibetan author attempted to apply that cited passage in a philosophical
treatment of the Yogacara position. But it is not valid to argue that
something is the case with all minds (and their presumed objects) when
it was only the case of a certain yogin and his personal attainment.
Theory that Vasubandhu Denies the External World. Certainly some
readers decided that Vasubandhu's twenty-verse treatise (the Vimsatika)
involves a denial of the external world. And such persons are entitled
to say: suppose we grant your previous argument about cittamatra and the
traidhatuka. Even so there are verses in that Vasubandhu treatise that
appear to deny the external world. Before going into those particular
verses, it is well to notice the situation in his Abhidharmakosa.
Abhidharma Buddhism recognizes six senses and their objects, from the
sense of eye with its object of formations (in shape or color) to the
sense of mind (manes) with its object of natures (dharma). But this does
not mean that one necessarily perceives such sense objects. So Buddhism
taught that there is a `perception' (vijnana) based on the eye, and so
with the other senses as bases (ayatana). Because the senses had the
power to apprehend those various objects, they were given the Sanskrit
name indriya, a word which means `a power'. It follows immediately that
`perception' is powerless; that is, it is unable to contact the object
directly, but must depend on whatever the sense organ comes up with.
presents the manner in which Vasubandhu treated this matter in
his Abhidharmakosa:(27) rupam pancendriyany arthah pancavijnaptir eve ca
(1, 9) ("Formation is five sense organs, five objects, and
non-representation"); caksuh pasyati rupani sabhagam na tadasritam
vijnanam drsyate rupam na kilantaritam yatah (1, 42) (Bareau's
translation: "When it is in condition to work, the eye sees formations;
the consciousness which is leaning on it does not (see formations),
because formation, then being concealed, is not seen"). If we grant that
perception fails to reach the object that was apprehended by a sense
organ, we begin to notice how perception, perhaps automatically, adds to
the sensory evidence, as was discussed above. And this may help to
explain why the same art object is viewed differently by different
persons--presumably it was about the same as a sense object, but the
subjective perceptions evaluate it quite divergently.
to the twenty-verse treatise. The translation from the Chinese by
Hamilton(28) was reprinted in A Source Book by Radhakrishnan and
Moore.(29) We learn there that these twenty verses are on
vijnaptimatrata, there rendered `Representation-Only'. It is somewhat
embarrassing to me to point out what the editors have added in their
footnote on page 328, because I have always had great admiration for
both Radhakrishnan and Moore. They say: "A better translation of
vijnaptimatrata would be `ideation-only,' since `representation'
suggests rather than denies external reality." Well, no fair
mistranslating a Sanskrit term just to make one's theory come out right!
That remark was made because of thinking that Vasubandhu's treatise
denies external reality.
Levi's edition(30) of the Sanskrit for the twenty verses numbers
them as twenty-two. His verse 1 can be taken as introductory, and his
verse 22 as concluding. The translation from the Chinese, as presented
by Hamilton, starts with Sanskrit number 2 as the first verse. This is
Levi's introductory verse with my translation:
vijnaptimatram evaitad asadarthavabhasanat/
yatha taimirikasyasatkesacandradidarsanam //
This just amounts to representation, as the sight of unreal hair, moon,
etc. of one with an eye-caul--because being the (subsequent)
manifestation of an unreal artha (external thing).
introductory statement does not deny an external object. Instead
there is a mental representation that amounts to tinsel, `fool's gold',
a false wealth. Vasubandhu appears to mean that the mind imagines an
external artha in front, but the mind has only a report or
representation of what the sense organ had sensed.
This is Levi's
verse, and what for both the Chinese and the Tibetan is
verse 1, with my rendition (Levi had to reconstruct the Sanskrit):
yadi vijnaptir anartha niyamo desakalayoh /
samtanasyaniyamas ca yukta krtyakriya na ca //
representation lacks an external object (artha), there is no
(aniyama) of space and time; there is no certainty of the composite
stream (of consciousness) and agency is not valid.
notice again that Vasubandhu does not here deny an external object,
because the sentence makes a supposition, "If .... " The verses go on to
make a distinction between the beings of different destinies, gods,
humans, hungry ghosts, and so forth, in how they view externals. Thus
the gods see the river sparkling with gems, humans see it as good to
drink, and the hungry ghosts (preta) find the river full of unclean
things rendering it unfit to drink. We notice this difference in terms
of destiny classes that it involves the use of external things. Again,
it is not a denial of external objects, but a claim that different
destinies have a different addition (philosophically false) to what was
there is what is numbered verse 16 in Hamilton's translation, where
he renders the first part as follows: "As has been said, the apparent
object is a representation. It is from that memory arises" (the Sanskrit
for this: uktam yatha tadabhasavijnaptih; smaranam tatah). Hamilton
translates the commentary on this:(31) "As we have said earlier,
although there is no external object, a sense representation, visual,
etc., appears as an outer object. From this comes the later state with
its memory associate, the discriminated mental representation, appearing
as a seeming former object. Then we speak of this as a memory of what
has been already experienced." Notice that Hamilton translated the first
sentence of this commentary as though there is denial of the copula
("although there is no external object"). But when we consult the
Sanskrit that Levi edited, we find the sentence worded differently:
"Even in the absence of an external object" (vinapy arthena). It is
necessary to translate this way to make sense of the comment that
Hamilton translates from Chinese:(32) "That is, he defends his position
by saying that there must have been this object immediately received in
the past by the five organs of sense, eye, etc. [so that] in the present
the intellective consciousness is able to hold it in memory."
have probably all had such an experience, when concentrating on some
problem or passage, if someone comes to the door and says, "Dinner is
ready" and, not immediately hearing a response of the type "O.K. I'm
coming," continues, "I said: `Dinner is ready'!"--whereupon the
concentrating person responds, "I heard you the first time"--not exactly
as this person now says it, because if we mean by hearing the actual
sounds as heard at the time the sounds are made, this is true just for
the reception by a sense organ (of hearing), and not true for auditory
perception. According to Vasubandhu, as the Chinese commentary here
understood him, the person remembered the words "Dinner is ready."
"even in the absence of an external object," that is, even though
that sound is no longer sounding, a person may hear it as a memory
image. Due to the actual experiences of yogins, this situation was taken
for granted. There is a celebrated case associated with what in Buddhist
history is called the Second Council, to determine if certain erring
monks should be ousted from the Samgha: all the senior monks were called
to assemble for the hearings, but one of these monks was in the deep
concentration called nirodha-samapatti. According to the story, upon his
emerging from the samadhi a divinity gave him the message, whereupon he
sped to the meeting.(33) It should be admitted that this yogin
remembered the message, while the instigating sound was no longer s
when we examine the text more carefully, we find that Vasubandhu
does not deny the existence of external objects in this and in the
previously cited materials, even though the translator, just by his
manner of translating, made it appear so. Besides, two authors of recent
books translating a number of the Vasubandhu treatises agree that
Vasubandhu does not deny an external object (Kochumuttom(34) and
to Murti's chapter, previously alluded to, a few words will suffice.
A reader of that chapter, supposedly on the `absolutism' of Vedanta,
Madhyamika, and Vijnanavada, will readily find out that the Yogacara
position (called here `Vijnanavada') is set forth, not from Yogacara
books, but from their rival Vedanta and Madhyamika books. Having decided
that the opponents must be right, when he then cites a Yogacara treatise
it must be made to agree with Murti's supposition. So, referring to the
Madhyantavibhaga, he says, "The constructed subject-object world is
unreal; but this does not make the abhutaparikalpa unreal; for, it is
the substratum for the unreal subject-object duality. It is, however,
non-conceptual."(36) So abhutaparikalpa, which means "the imagination of
what did not (really) happen," is `non-conceptual'! I conclude that
Murti in this chapter does not advance the understanding of Yogacara
Alayavijnana. When we turn to the Yogacara theories that devolve
about the term alayavijnana, we notice that what must have been a hotly
contested point even in the time of the Buddhist master Asanga is still
in present times disputed. I already mentioned that there is a
two-volume work on the topic by Schmithausen. Here, under a heading
"Introduction and Original Meaning of Alayavijnana," he points to a
passage in the Samahitabhumi portion of the Yogacarabhumi about a person
in the deep concentration called nirodha-samapatti, and that it is a
continuance of the alayavijnana with its seeds, which shows that even
though various other mental functions have ceased, vijnana itself has
not ceased, and these seeds will bring forth the evolving types of
vijnana when the person emerges from the samadhi.(37) Schmithausen calls
this the "Initial Passage" here and a number of times later on in his
work. He goes on to claim(38) that the passage, although not stating
this explicitly, implies that the continued presence of the alayavijnana
has kept alive that yogin who is in nirodha-samapatti. And if that is
so, then the alayavijnana must also be associated with the moment of
conception in the womb,39 and so this is the vijnana on which
Name-and-Formation (nama-rupa) arises in dependence--in the usual
sequence of the Buddhist Dependent Origination. Accordingly, he insists
that the pratisamdhi (or `linking', the `rebirth') kind of vijnana is
that initial alayavijnana that descends into the male-female element
union in the womb.(40)
the course of his investigation he was led to disagree with various
Japanese scholars who understood these matters differently from him. I
shall have to evaluate whether his conclusions are consistent with the
positions of Asanga, the founder of the Yogacara, or of Vasubandhu, the
great popularizer thereof.
happens that Asanga himself--obviously responding to a number of
challenges and condemnations of this alayavijnana position--gave his
answers in the opening section of his exegetical section called
Viniscayasamgrahani, which I employ in the Tibetan version in the
Tanjur. There we learn that the `store consciousness' is the abode of
seeds (alaya-vijnanam bijasrayah).(41) Asanga claims that this is a
secret teaching of the Bhagavat, citing a well-known verse from the
Samdhinirmocana-sutra about the adanavijnana (the consciousness that
`takes' [seeds]).(42) But the reader of this section cannot avoid the
conclusion that Asanga is convinced that of the many references to
vijnana in the old Buddhist Sanskrit canon (the four Agamas) that he
employed, they cannot all be explained as the standard six perceptions
based on the six sense organs, but that there are various contexts of
this Sanskrit expression which justify it to be understood differently.
as Asanga continues in his defense of this type of vijnana he sets
forth three reasons that surprisingly were not referred to by
Schmithausen. I shall cite the Tibetan along with my translation below
each passage, and then follow with a discussion--insofar as it is
ci'i phyir kun gzhi rnam par ses pa med na lus kyi tshor ba mi rung
/ 'di Itar tshul bzhin nas tshul bzhin ma yin pa sems par byed pa
rjes su rtog par byed pa'am / sems mnyam par bzhag pa'am / sems mnyam
ma bzhag pa gcig cig /us la tshor ba rnam pa du ma rnam pa many po sna
tshogs gang dag `byung ba'i rigs na snang ste / de'i phyir yang kun gzhi
rnam par ses pa yod do /(43)
Why in the absence of alayavijnana is the body's feeling not
is this way: when positing that the mind attends in the right manner
imagines in the wrong manner; or that the mind is equipoised, then
not equipoised, there appear principles that bring forth a multitude of
of varied kinds of feelings in a certain body. Therefore, there
ci'i phyir kun gzhi rnam parses pa med na sems med pa'i snyoms per
mi srid ce na / 'di Itar 'du ses med pa la snyoms par zhugs pa'am/
la snyoms par zhugs pa'i rnam par ses pa lus dang bral ba kho nar
ma bral bar mi 'gyur bas / de'i phyir si ba kho nar 'gyur ba zhig
Idan 'das kyi de skad du / de'i rnam par ses pa ni lus dang bral ba
yin no zhes gsungs pa'i phyir ro /(44)
2. Why in the absence of alayavijnana would there be no possibility of
without thought (acittika-samapatti)? It is this way: the
is in non-ideational equipoise (asamjnika-samapatti) or is in
(nirodha-samapatti) only occurs when it is absent from the
and would not occur [that way] when not absent from the body). For that
it was only for the case of death that the shagavat declared,
vijnana is not absent from the body."
ci'i phyir kun gzhi rnam par ses pa med na 'chi 'pho mi rung zhe na /
'pho ba'i tshe'i rnam par ses pas lus ro stod dam / ro smad du drod
byed cing spong la yid kyi rnam par ses pa ni nam yang mi 'byung ba
bas / de'i phyir lus len par byed pa'i kun gzhi rnam par ses pa kho
dang bral bas lus kyi drod yal ba dang / lus la tshor ba med par snang
bar zad kyi / yid kyi rnam par ses pa dang bral bas ni ma yin te / de'i
phyir yang mi rung ngo/(45)
3. Why in the absence of alayavijnana is there no feasibility of
transmigration? It is this way: the vijnana at the time of [dying and]
transmigrating leaves when the warmth of the upper and lower parts of
the body fades away, and the manovijnana certainly does not occur and is
[at that time]. On that account, only in the absence of the
which takes a body does feeling (vedana) get lost in the body, but this
does not happen through the absence of manovijnana. So there is no
feasibility (in the absence of alayavijnana).
of the first conclusions about these three passages is that
Schmithausen was wrong in concluding that alayavijnana was necessary so
that a meditator who is in the trance state nirodha-samapatti would not
die there. Certainly, if Asanga had thought so, this would be the place
for him to have said so. It is a case when vijnana departs from the
body, but the person does not die (on that account). But when a person
does die, vijnana does depart from the body. However, the three passages
certainly require further explanations. A certain amount of explication
should come through considering certain Schmithausen claims as were
alluded to above.
to his view that the alayevijnana is the kind of vijnana that
descends into the male-female element union in the womb, supposedly
bringing life thereto, there is Asanga's own explanation in the early
part of the Yogacarabhumi:
[tatra] sarvabijakam vipakasamgrhitam asrayopa adanad
There, the `store-consciousness' all-seeded and restrained by
maturation, after taking a body, faints (or falls unconscious, or
ca kalaladese tad vijnanam sammurcchitam so 'sya bhavati tasmin
hrdayadesah / (47)
Where that vijnana faints in a place of the kalala (initial form of the
it [that place] becomes for it [the embryo] at that time the
of the heart.
that in this account, the initial form of the embryo after
conception is already there when the `store consciousness' enters. The
term kalala is used in Indian medicine for the initial embryo. So it is
a case like the Genesis account already mentioned where the support
world precedes the subjective element. Thus while Schmithausen was right
about the role of alayavijnana to represent in some way the vijnana that
is the third member of Dependent Origination, and so to fall into the
womb (in the human case), it is clear that Asanga does not, and would
not, ascribe to vijnana the role of conferring life--as Schmithausen
claimed. These remarks help to explain somewhat the third of Asanga's
defenses of the alayavijnana.
Schmithausen claimed, and repeated his claim, that this
alayavijnana that falls into the womb is the pratisamdhi (rebirth) type
of vijnana. Apparently because some pandits were espousing such a theory
at the time of Vasubandhu, he countered it in no uncertain terms in his
Dependent Origination commentary: "It is not right that the
pratisamdhivijnana is by way of samskara [the second member of Dependent
Origination]. It is true that vijnana arises [as the third member]; the
Name-and-Formation (nama-rupa) [as the fourth member] arising on that
basis is the time of pratisamdhi--this is the faultless position."(48)
Gunamati claims that Vasubandhu's Abhidharmakosa identifies the
pratisamdhi (`linkage' = reincarnation) with the skandha kind of
vijnana.(49 ) He presumably means chapter 3, the introductory paragraph
to k. 14, where the first line in the Sanskrit text has the expression
"five upadana-skandha," and in the next line it says "The moment of
pratisamdhi is the birth in the destinies (gati)" (upapattibhavo gatisu
helps to make sense of the foregoing when, in his famous
work Visuddhimagga, he sets forth under the topic of the fourth member
of Dependent Origination that the naman (of nama-rupa), when initially
developing in the womb, consists of three aggregates. That is, he
defines it, vedanadayo tayo khandha, "the three personal aggregates,
feelings, etc.,"(51) thus leaving out vinnana (the Pali way of writing
vijnana), while the standard Abhidharma listing of the naman part
presents the four members, including vijnana. This forces the well-known
canonical passage holding that nama-rupa arises in dependence on vijnana
and that vijnana arises in dependence on nama-rupa to mean that vijnana
is added to the naman to make four.
short, when the vijnana that is number 3 in Dependent Origination
falls into the womb, it is there the `store consciousness' that is a
store of seeds. As to the word `seed' (bija), Asanga gives a set of
words that are roughly definitions, the bija-paryaya: realm (dhatu),
lineage (gotra), basic nature (prakrti), basic cause (hetu), the real
set (satkaya), elaboration (prapanca), store (alaya) the taking
(upadana), suffering (duhkha), foundation of reifying views
(satkayadrstyadhisthana), and foundation of `I am' pride
(asmimanadhisthana)--and any others belonging to the same set.(52)
Besides, Asanga gives a list of various outcomes of these seeds: family
(kula), strength (bala), bodily appearance (rupa), length of life
(ayus), enjoyments (bhoga), and so on; and of these effects, principally
good (subha) and bad (asubha) karma is the cause.(53) Later, Asanga told
how this `store consciousness' gets its seeds:
evam avyakrta dharma kusalakusalavyakrtan dharman avahanti /
tadyatha kusalakusalavyakrtabijakam alayavijnanam avahanti /(54)
Thus, the indeterminate natures (avyakrtadharma) bring the virtuous,
unvirtuous, and indeterminate natures, as follows: they bring (them) to
the `store consciousness', which is seeded with the virtuous,
unvirtuous, and indeterminate.
it was mentioned that the Yogacara frequently presents a list
of eight vijnana, of which number 7 is the `defiled manes' and number 8
is the alayavijnana. Now, in Buddhist commentarial exegesis of the
scriptures, there was a problem with the vijnana that is third in
Dependent Origination, and was said to have a `vision' of the birthplace
and so to be attracted thereto. Now, even in the Yogacara, it would
hardly be feasible to identify this third member with the alayevijnana,
since a store of seeds could hardly be called visionary, except for the
envisioning of effects, as an acorn might be said (poetically or
metaphorically) to foresee the oak tree. But however we might credit the
acorn with such an ability, we should all admit that the acorn cannot
imagine where it will grow. Vasubandhu was well aware of the difficulty,
so in his Mahayasamgraha commentary he said:
Besides, when the manovijnana that is defiled witnesses the
birthplace, the intermediate state [between death and rebirth] comes
to an end. That it "faints" means that the manovijnana comes together
with the male and female generative elements, [ana] experiences a
single [moment of] bliss, whereupon the manovijnana faints [i.e.,
becomes unconscious or inactive], and on the basis thereof, a different
sort of manovijnana enters."(55)
Asanga had already denied that the manovijnana (i.e., the one
based on the manes as the sixth sense) is operative at the time of
death, Vasubandhu must mean the seventh vijnana, what in later Yogacara
was referred to as the `defiled mind' (klista-manas). Later, Vasubandhu
comments: "Therefore, the manovijnana that faints is not [i.e., is no
longer] a manovijnana, but is a vipaka-vijnana [i.e., a resultative
kind]; and that is `all-seeded'."(56) He therefore admits that it was
the `defiled mind' that falls into the womb and, once there, is called
alayevijnana. This separate category of a `defiled mind' seems to have
been adopted for the death vision, for the visions during the
intermediate state (antarabhava) for the birth vision, and perhaps also
(just my own speculation) for hypnogogic states during a lifetime.
as this `store consciousness' is credited with having a store of
`all' dharmas, we should recall that in Buddhist Abhidharma it is the
manovijnana that has those dharmas supposedly as object, while it is the
sixth sense manes that has those dharmas directly as object. In the
theory of Dependent Origination alluded to above, for a vijnana that
arises in dependence on Name-and-Formation, Gunamati explains that the
manovijnana is the main one: "Given the set of six vijnana only
manovijnana is the fastening (fib. sbrel ba) of pratisamdhi (the
reincarnation)."(57) This means that the fourth member
Name-and-Formation is necessary for the Abhidharma manovijnana to
operate; so the remaining vijnanas based on the-five outer-directed
senses, must require the fifth member, the Six Sensory Bases
(sadayatana), to operate.
the foregoing should clarify that the chief demand to have the
concept of alayavijnana was for Asanga's third reason--the theory of
rebirth. Schmithausen's theory that the initial place is the context of
a special yoga state--that is, nirodha-samapatti--is hardly tenable. The
discussion at some length in the early part of the Yogacarabhumi shows
that it was the Buddhist arguments over whether some persons had or did
not have the potentiality of Nirvana,(58) and the attempt to justify
events of the present day as the effect of previous lives, that demanded
a carrier of a seed-nature. This must be why Asanga allows for the
operation of alayavijnana when the `evolving perceptions' are in
then there are the first and second reasons that Asanga mentioned.
These are quite difficult because yoga states are implicated. When
Asanga mentioned, as cited above, that vijnana entered the embryo at the
place where the heart would form, this makes it clear that when he said
in the second reason that in the case of those two kinds of samapatti
the vijnana is absent from the body, it means that vijnana had left its
`heart' location, and was somewhere else for the time being (--in the
head?). This suggests that the yogin is in a sort of cataleptic state.
And this seems to be why, in the first reason, Asanga had credited the
`store consciousness' with enabling the feelings of the body. Thus, in
certain trance states there is a local or more general loss of feeling,
and Asanga claims that to understand what is going on in these
specialized states, one must posit a `store consciousness'. This appears
to be enough for these three reasons, as far as this essay is concerned.
would also take too much space to go into the matter of the
transmutation of the `store consciousness' (alayavijnana-parivrtti), and
the matter of which advanced persons--a Buddha, an arhat, advanced
bodhisattvas, and so on--have transmuted this `store consciousness' and
so do not possess it.(60) But this teaching shows that the alayavijnana
theory does not require it for the yoga experience of ekagracitta as it
was discussed above, which implies a special condition of the evolving
perceptions, especially manovijnana (the perception based on the mental
sense of mind, manas).
Three Laksanas. There is considerable treatment already in Western
sources on the three laksanas of Yogacara theory that are also called
the three svabhava. Nagao has written a spirited essay on the topic,
"The Buddhist World View as Elucidated in the Three-Nature Theory and
its Similes."(61) As Nagao describes the three, they are the imagined
nature (parikalpita-svabhava) the other-dependent nature
(paratantra-svabhava) and the consummated nature
(parinispanna-svabhava). And he explains: "The `imagined' nature,
therefore, is characterized by `unreality' and `total nonexistence'."
"In contrast to this, parinispanna or `consummated' means perfect, real,
and existent and connotes `reality,' `truth,' `real existence', or `the
absolute'." "Between them is the third nature, called paratantra, the
`other dependent'. It exists, but only by depending on some other
entity."(62) Nagao cites an important observation from Vasubandhu's
Trimsika, k. 21 c-d: "When the other-dependent nature obtains a state
absolutely free of the imagined nature, it is then the consummated
is the conclusion of the present writer that this system of three
natures is very close to what is found in the writings of the earlier
and famous Nagarjuna, with the difference that the Yogacara thought it
was improving in describing what is going on. I allude to two verses in
the latter's Acintyastava, 44-45:(64)
hetupratyayasambhuta paratantra ca samvrtih / paratantra iti proktah
paramarthas tv akrtrimah // 44
svabhavah prakrtis tattvam dravyam vastu sad ity api / nasti vai
kalpito bhavo paratantras tu vidyate // 45
Convention, with dependence on other(s) (paratantra), arises from a
cause and from conditions. This dependence on other(s) has been
announced (by Thee). The Absolute is not fabricated. (44)
It (the Absolute) is termed self-existence (svabhava), primary nature
(prakrti), reality (tattva) substance (dravya), abiding essence (vastu),
the really existent (sat). An entity (bhava) when imagined does not
exist, but (exists) when its dependence on other(s) is found. (45)
the complete agreement with the Yogacara that an entity when
imagined does not exist, but does exist with its dependence on another;
and the same word for dependence on another (paratantra) is used.
Nagarjuna, like the Yogacarin espousers of the three svabhava theory,
has an absolute, defined by six terms. The only seeming difference is
when Vasubandhu holds that this absolute nature is the other-dependent
nature when the latter is free of the imagined nature. This amounts to
accepting for Dependent Origination (pratityasamutpada) that there is
both a conventional and an absolute explanation. Even if Nagarjuna
accepts this, it would probably be difficult to draw it from his works.
brief account of the three natures should suffice. Clarification of
I present three clarifications: (a) in terms of subject and object,
(b) in terms of Buddhist logic, and (c) in terms of Western discourse.
in Terms of Subject and Object. Previously I discussed the
position in Vasubandhu's treatise of twenty verses and rejected the
claim that he denied external existence. But more needs to be said about
this. Vasubandhu in this brief work of his used the term artha for the
presumed external object. Thereby he could only implicate the five
outer-directed sense bases that have these arthas as objects. The word
artha in its general Indian usage stands for property and goods, and of
course these stay behind when a person passes to the other world. This
usage of the term seems to be involved in its etymology.(65) The sixth
sense, manes, has the set of dharmas as object; and by the Indian theory
of transmigration, some of these may well be held to transmigrate.
us now consider the treatment in Matilal's book Perception. In an
"Analysis of Perceptual Illusion" he has a subsection on two Buddhist
analyses of illusion, admitting that he follows Vacaspati Misra for
these materials; and this is quite proper, because he aims just not to
misrepresent the Hindu author. Matilal summarizes the Yogacara position
in seemingly well-stated sentences, which I shall number: (1 ) "The
object-form is an integral part of the awareness itself, each awareness
being different from another by virtue of this unique object-form which
appears in it." (2) "The object-form does not come from outside." (3)
"In fact when the object-form is projected outside or externalized, we
are said to have an awareness of the external object."(66) In all three
statements, Matilal attributes to the Yogacara that there is an
`object-form' in the mind. But when we refer to what Vasubandhu said (in
his Vimsatika), we find him explaining the `representation-only' not as
an artha, the external object, but as an `unreal object form'
(asadartha)--a poor copy of the external object. Such a term affirms,
rather than denies, what is connoted by the term artha, when it is
employed by itself. Vasubandhu apparently used the term artha instead of
the standard Buddhist visaya to fend off misattributions of the position
he followed. Still, he was misunderstood, and these misunderstandings
were copied over and over.
there is the essay in Philosophy East and West by Prasad, who, like
Matilal, is a good writer on these topics.(67) He cites Vasubandhu's
treatise on the three natures, the Trisvabhavanirdesa, as follows:
is the conception of that which is nonexistent? [The answer is]
[mental projection] / For by it, the nonexistents are imagined; and
inasmuch as the mind imagines objects, they do not exist at all.
What is that which is presented in cognition? The nonexistent which is
[or imagined]. How is that presented in cognition? In the form
a twofold appearance [of the apprehender and the apprehended] / What is
in cognition that does not exist? That by which the twofold
the Sanskrit for the verses as is found in Kochumuttom,(68) I
accept Prasad's translation as on the whole correct. In the case of the
second of his cited verses, his rendition may possibly mislead the
reader, as suggesting that a nonexistent external has appeared in the
mind. I believe that here we should adhere more literally to the
Sanskrit: tatra kim khyati asatkalpah katham khyati dvayatmana ("What
appears there? The imagination of an unreal. How does it appear? As the
subject-object duality"). This subject-object duality is found discussed
in Buddhist literature generally by the terms grahya (the apprehender)
and grahya (the apprehended). The avoidance of the two is the topic of
the scripture presented above, the Kalakarama Sutta. Thus, when Vasuband
hu writes such verses, he evidently believes that they are consistent
with the Buddha's teachings.
in Terms of Buddhist Logic. Stcherbatsky and others
claimed that there is a pronounced influence of Yogacara philosophy on
Buddhist logic of the Dignaga-Dharmakirti lineage.(69) If one is to
credit the Yogacara with such influence, one should state the influence
rather specifically. I have previously published three essays on this
matter, "Yogacara and the Buddhist Logicians,"(70) "A Reconsideration of
Dharmakirti's `Deviation' from Dignaga on Pratyaksabhasa,"(71) and
"Dharmakirti and the Yogacara Theory of Bija."(72) In the first of these
essays I translated Dignaga's brief treatise, the Alambanapariksa, which
deals with the theory of `atoms', which Vasubandhu also dealt with in
his twenty-verse treatise. I also cited the commentator Dharmottara's
passage explaining the so-called atoms as constituting color (varna),
while the shape (samsthana) was added, presumably in
Representation-Only. The article "A Reconsideration ... " defended
Dharmakirti's position that there are four kinds of falsification of
perception (pratyaksabhasa) and that this was also Dignaga's position.
The one on Dharmakirti and bija showed that Dharmakirti was amenable to
this `seed' way of talking.
the many authors who claim this Yogacara influence is C. L.
Tripathi, who wrote a book titled The Problems of Know/edge in Yogacara
Buddhism that included a treatment of Buddhist logic. He is among the
many who think that Yogacara Buddhism denies external existence, saying:
"Vasubandhu ... categorically denies the existence of the external
world."(73) Yet this same author has a chapter "Object of Perception"
presenting the position of Buddhist logic that the object called
svalaksana is the only real.(74) He calls it a `particular', using the
Western terminology of `universal' and `particular'. This author seems
not to realize that if the Yogacara does indeed deny external objects,
it opposes Buddhist logic--and so why treat Buddhist logic in a book
with such a title?
I shall continue the discussion with the `falsification of
perception', because this apparently agrees with the Yogacara
`representation-only'. Vasubandhu's commentary on the Mahayasamgraha,
its chapter 2, speaks of `representations' (vijnapti) belonging to the
body, the body-possessor, and the eater. Here `body' means the five
realms (dhatu) of (sense organs), eye, and so forth. `Body-possessor' is
the `defiled mind'. The `eater' (or enjoyer) is the realm of mind
(manodhatu), that is, mind (manes), the sixth sense.(75) Another
relevant passage is in Vasubandhu's Karmasiddhiprakarana (I translate
from the Tibetan): "There are two kinds of citta: (1) what collects its
seeds (= alayevijnana); (2-a) what has it (i.e., the alayavijnana) as a
mental support (alambana), namely the `defiled mind' (klistamanas);
(2-b) what has images (akara) of it (i.e., the alayavijnana), namely the
manovijnana; [and] (2-c) what have differing distinctions, namely the
five outer-directed perceptions (vijnana)."(76) So the `defiled mind'
knows the (subconscious) alayavijnana seeds, while the manovijnana
distinguishes the (conscious) images that the seeds have sprouted into.
Then we notice that the traditional set of six vijnana amounts to
`representations' because this system uses the convertible terminology
cittamatra, vijnanamatra, and vijnaptimatra. Five of the six are
representations of the five sense organs (called the `body'), while the
sixth one is a representation of the sixth sense, manes.
we can compare with the four kinds of error of Buddhist logic.
Dharmottara's commentary on Dharmakirti's Nyayabindu, following
Dharmakirti's Pramana-viniscaya, listed four causes of error. They are:
(1) cause of error found in the object, for example the whirling
firebrand taken as a wheel; (2) cause of error found in a place, for
example embarking in a boat, where the trees on the shore are moving;
(3) cause of error found within, for example being troubled by hatred;
and (4) cause of error found in a sense organ, such as the `caul',
causing the conch shell to appear yellow.(77) It is reasonable to
compare with these four causes of error--even though two are placed
externally--since we have shown above that Vasubandhu does not deny
externals as being existent, provided they arise in dependence on
1, cause of error found in the object, amounts to representations
based on the sixth sense organ, manes. Number 2, cause of error found in
a place, also amounts to representations based on the sixth sense organ,
manes. Thus, for numbers 1 and 2, the representations are called
manovijnana. Skipping to number 4, cause of error found in a sense
organ, the representations are the five outer-directed perceptions.
Notice that this illustration of a caul on the eyes is precisely the
example used by Vasubandhu in the introductory verse to his treatise
with twenty verses (above, subsection "The Theory that Vasubandhu Denies
the External World").
going to number 3, the cause of error found within, with the
example of hatred, this clearly goes with representations of the
`defiled mind'. About the `defiled mind', Asanga's Paramartha-gatha,
39-41, contains these points: "The defiled mind (klistam manes) always
arises and ceases together with defilements (klesa)"; "On another
occasion it is born pure"; and "That which was defiled, here in the end
is purified, with its intrinsic light (prakrtibhasvara)."(78) This shows
that the `defiled mind' requires defilements, which are dharmas.
the four causes of error found in those texts of Buddhist logic
are reasonably based on the Yogacara theory of `representation-only',
and this is certainly a better solution than my old attempt in the
Bhandarkar journal to associate these four causes of error with the four
pratyaksa of the Buddhist logic system.(79) Accordingly, these causes of
error are not a theory that the world is an illusion, since the causes
of error can be `seen through', appreciated for what they are, in
in Terms of Western Discourse. Modern Western philosophers
have concerned themselves with getting proper descriptive statements
regarding perception and its associate functions. Since such topics can
generate much writing, I shall restrict my comparisons to issues raised
in an article by Bijoy H. Boruah, "Seeing in the Mind's Eye."(80) He
concerned himself with theories of the author Gilbert Ryle and others.
He was clearly impressed with the phraseology by Elizabeth Anscombe,
`intentional seeing' and `material seeing'. The example was: suppose we
look at a painting of Gandhi (the `material seeing'), and then go away
and ask ourselves, "What did I see?" Then these authors are forced to
use the same expression, namely the translation into English of
vijnaptimatra as `representation-only', but say `representational
seeing', which is `intentional seeing', while Wittgenstein used words
like `seeing as'. But then Boruah has to decide that `seeing in the
mind's eye' is not representational seeing, and this conclusion forces
him to reject various theories by Ryle and others. Thus, to see in the
mind's eye is to have an image that involves both "doing and achieving"
and so is distinguished from "abstract or purely conceptual
contemplation." But Boruah recognized (speaking `metaphorically') that
this image "emerges only when the finger of thought touches the right
cord of sentience." He thus admits that a sentience precedes this image.
it might have been the opponent--was cited above for a
consistent remark, which I would now render: "Besides, there is no
particular fault in understanding that there is a prior
representation."(81) This apparently means, "At least there is one thing
we can agree on--there was some kind of prior representation."
Therefore, for the Yogacara theory of the `store consciousness', the
seeds stored here cannot sprout unless there is a prior
`representation-only' triggered by sensory input or by the `defiled
mind'. And the sprouting seeds would provide the images alluded to in
illustrate the independence of the image from sentience as well as
from abstract thought--which Buddhism calls `discursive thought'
(vikalpa, etc.), one could take the example of the composer Beethoven.
He first took classes in music, heard Bach, and so forth, and this was
represented in his mind, `heard as'. Later, he composed great works of
music, but this did not involve hearing with his ear--because he was
becoming stone deaf. Therefore, his auditory imagery of music became
independent of sentience, that is, in terms of hearing external sounds,
and was also independent of discursive thought. He did have discursive
thought to change his score--"I should change it thus"--or to fume
against other persons, and so forth, which is outside his composing
music. Of course, even in his deafness he still had visual input from
of Beethoven also illustrates the description of the
image--here auditory imagery (heard by the sixth sense, manas)--as a
sort of doing and achieving. The Yogacara issuance from the `store
consciousness' that is expressed in the language of `seeds' (bija)
amounts to a compatible way of talking because the seed can do
something, achieve a result, say, a shoot. Since this issuance from the
`store consciousness' can be of the indeterminate as well as of the
virtuous and the unvirtuous, there are a vast number of images possible.
Since these images can also be called representations, this shows the
active nature of these representations.
is hoped that this essay, including the preceding three
clarifications, may help to elucidate Yogacara Buddhism.
that the Yogacara system has been somewhat clarified on behalf of
anyone who wants it clarified. The misrepresentations of this system
that have appeared in older as well as in recently published works are
not more correct simply by being copied over and over. Of course, the
Yogacara put its trust in the subjective search for truth by way of a
samadhi. This rendered the external world not less real, but less
valuable as the way of finding truth.
tide of misinformation on this, or on any other topic of Indian lore
comes about because authors frequently read just a few verses or
paragraphs of a text, then go to secondary sources, or to treatises by
rivals, and presume to speak authoritatively. Only after doing genuine
research on such a topic can one begin to answer the question: why were
those texts and why do the moderns write the way they do?
ago I knew that the Yogacara position was misrepresented in many
works, ancient and modern. Only recently was the means to defend
Yogacara Buddhism put in my hands, as has been detailed above. I do not
care whether or not others are convinced by the arguments presented
Cf. Gregory J. Darling, An Evaluation of the Vedantic Critique of
Buddhism (Delhi: Motilal Banarsidass, 1987), p. 371, taking notice of an
incorrect portrayal of Buddhism in Vedantic criticism.
A. Wayman, "The Yogacara Idealism," Philosophy East and West 15 (1)
T.R.V. Murti, The Central Philosophy of Buddhism (London: George
Allen and Unwin Ltd., 1955), chap. 13, "The Madhyamika, Vijnanavada and
Vedanta Absolutism," pp. 311-328.
Lambert Schmithausen, Alayevijnana: On the Origin and the Early
Development of a Central Concept of Yogacara Philosophy, pt. 1, Text;
pt. 2, Notes, Bibliography and Indices (Tokyo: The International
Institute for Buddhist Studies, 1987), here, pt.1, pp. 13-14.
Alex Wayman, Analysis of the Sravakabhumi Manuscript, University of
California Publications in Classical Philology, vol. 17 (Berkeley,
Alex Wayman, review of Hidenori S. Sakuma, Die
Asrayaparivrttitheorie in der Yogacarabhumi ..., 2 vols. (Stuttgart:
Franz Steiner Verlag, 1990), in Journal of the American Oriental Society
113 (1) (1993): 144.
Bhikkhu Nanananda, The Magic of the Mind: An Exposition of the
Kalakarama Sutta (Kandy, Sri Lanka: Buddhist Publication Society, 1974),
F. L. Woodward, The Book of the Gradual Sayings, vol. 2 (London:
Pali Text Society, 1952), pp. 26-28.
William G. Doty, "Hermes' Heteronymous Appellations," in James
Hillman, ed., Facing the Gods (Dallas, Texas: Spring Publications,
1984), p. 131.
See Lozang Jamspal, ed., Abhidhanavisvalocanam of Sridharasena
(Narita: Naritasan Shinshoji, 1992); and see Alex Wayman, trans.,
Abhidhanavisvalocanam of Sridharasena (Narita: Naritasan Shinshoji,
Alex Wayman, "Doctrinal Affiliation of the Buddhist Master Asanga,"
in N. H. Samtani, ed., Amala Prajna: Aspects of Buddhist Alex Wayman
Studies (Delhi: Sri Satguru Publications, 1989), p. 214 and n. 74,
furnishing the Tibetan in transcription for the passage.
See, e.g., Th. Stcherbatsky, The Central Conception of Buddhism and
the Meaning of the Word "Dharma" (Calcutta: Susil Gupta, 1961), p. 7:
"Consciousness, it is stated, never arises alone, since it is pure
sensation, without any content."
For this simile, see I. B. Homer, trans., The Collection of the
Middle Length Sayings (London: Luzac, 1967), vol. 1, Mahavedalla-sutta,
p. 355, where, in the case of a burning oil lamp, "the light is seen
because of the flame and the flame is seen because of the light."
This translation is in the Sata-Pitaka series, Indo-Asian
Literatures, vol. 74, published by the International Academy of Indian
Culture, New Delhi, 1968; and the passage is at p.189, where the
translator also offers: "What belongs to the triple world, that is (of)
Sylvain Levi, Materiaux pour l'Etude du systeme Vijnaptimatra
(Paris: Librairie Ancienne Honore Champion, 1932), p. 13.
When the Japanese scholar, Professor Ryushin Ohminami, was at
Columbia University some years ago studying this scripture with the
Vasubandhu commentary for a work in Japanese, he kindly gave me both the
Peking Tibetan Tanjur edition (PTT) and the Derge Tibetan edition for
Vasubandhu's commentary. I have read the beginning of the Peking
edition, but for the present topic have used just the Derge edition
(published by Delhi Karmapae Chodhey and printed at Mujeeb Press, 1976),
the part I discuss at pp. 200 and 202.
As for the terminological set of three, as in Vasubandhu's
Abhidharmakosa, chap. 2, k. 34a-b, cf. Louis de La Vallee Poussin
(Paris, 1923), p.177, where the autocommentary explains, "Citta is so
named because it accumulates (cinoti); named manes because it knows
(manute); named vijnana because it distinguishes its object (alambanam
Cf. the Peking Tibetan canon (Japanese photo edition, PTT), vol.
98, p.127.5-1, citing, "khams gsum po 'di ni rnam par ses pa tsam
mo"--where rnam par ses pa tsam mo = vijnanamatra, and is asserted to be
these three worlds.
(19) Wayman, Abhidhanavisvalocanam of Sridharasena, p. 192.
Cf. Louis de La Vallee Poussin, L'Abhidharmakosa de Vasubandhu:
Troisieme Chapitre (Paris: Paul Geuthner, 1926).
Cf. T. W. and C.A.F. Rhys Davids, Dialogues of the Buddha, pt. 3
(London: Luzac, 1957), pp. 77-94.
For three lexicons which include the definition alambana, Anundoram
Borooah, Nanarthasamgraha (Gauhati: Publication Board, Assam, 1969),
text, p. 3, under "Agra," cites the Medinikosa, the Visvaprakasa, and
the one by Hemacandra. For the fourth one, see Jamspal,
Abhidhanavisvalocana, p. 278, no.1625A. Besides, alambana is a
definition for agra in these two lexicons of the Deccan College, Poona,
series: Nanarthamanjari, by Raghava, ed. K.R.V. Sharma (1954), and
Dharanikosa by Dharanidasa, ed. E. D. Kulkarni (1968). The translation
by A. Wayman, Ethics of Tibet: Bodhisattva section of Tsong-kha-pa's Lam
rim chen mo (Albany: State University of New York Press, 1991), adopted
the rendering `area' for the term agra, as though an area in the mind
That is, while preparing my Analysis of the Sravakabhumi Manuscript
(cited note 5 above).
(24) de La Vallee Poussin, trans. of the Abhidharmakosa, chap.1, p. 52.
See Borooah, Nanarthasamgraha; definitions from Medinikosa and from
Visvaprakasa. The same in Jamspal, Abhidhanavisvalocana, p. 297, no.
Tsewang, "The Mentalism of Dignaga and Dharmakirti," in Doboom
Tulku, ed., Mind Only School and Buddhist Logic: A Collection of Seminar
Papers (New Delhi: Tibet House and Aditya Prakashan, 1990), p. 15.
See Andre Bareau, "Abhidharmakosakarika of Vasubandhu: Index,"
reprinted from VAK (Poona: Deccan College), no. 3 :45-83.
Clarence H. Hamilton, trans., Wei Shih Er Shih Lun, or The Treatise
in Twenty Stanzas on Representation-Only (New Haven: American Oriental
Sarvepalli Radhakrishnan and Charles A. Moore, A Source Book in
Indian Philosophy (Princeton: Princeton University Press, 1957), pp.
Sylvain Levi, Vijnaptimatratasiddhi (Paris, 1925; reprint Shanghai,
(31) Hamilton, Wei Shih Er Shih Lun, p. 61.
(32) Ibid., n.114.
The story is cited in F. D. Lessing and A. Wayman, trans.,
Introduction to the Buddhist Tantric Systems (Delhi: Motilal
Banarsidass, 1978), p. 65.
Thomas A. Kochumuttom, A Buddhist Doctrine of Experience (Delhi:
Motilal Banarsidass, 1982), p.1, contends that the Yogacara is really a
"realistic pluralism" rather than, as it is usually described, an
"absolute idealism," and so on.
Anacker, Seven Works of Vasubandhu (Delhi: Motilal
Banarsidass, 1984), p.159, rejects the theory--claimed by some persons
to be the purport of Vasubandhu's Twenty Verses--that consciousness
unilaterally creates all forms in the universe.
(36) T.R.V. Murti, The Central Philosophy of Buddhism, p. 319.
(37) Schmithausen, Alayavijnana, pt.1, pp. 18 ff.
(38) Ibid., p. 31.
(39) Ibid., pp. 36-39.
(40) Ibid., pp. 5 ff.
(41) Asanga, in his Viniscaya-samgrahani, PTT, vol.110, p.235.1.
(42) Ibid., p. 235.2.
(43) Ibid., p. 235.4-6.
(44) Ibid., p. 235.4-8.
(45) Ibid., p. 235.5-2.
The Yogacarabhumi of Acarya Asanga, ed. Vidhushekhara Bhattacharya
(Calcutta: University of Calcutta, 1957), p. 24.4-5.
(47) Ibid., p. 24.18-19.
Vasubandhu, Pratityasamutpadadi-vibhanganirdesa, PTT, vol. 104, p.
287.3-3, 4: / de teas na 'du byed kyi rkyen gyis zhing mtshams sbyor
ba'i rnam par ses pa yin par rigs pa ma yin gyi / ... rnam par ses pa ni
'du byed kyi rkyen gyis yin no / de'i rkyen gyis nying mtshams sbyor
ba'i tshe ming dang gzugs yin no zhes bye ba de Ita bu'i lugs 'di ni
skyon med pa yin no /
Gunamati, Pratityasamutpadadi-vibhanganirdesa-tika, PTT, vol. 104,
p. 335-4-2: / chos mngon mdzod las / mtshams sbyor phung po'i rnam par
ses zhes smras pa.
(50) P. Pradhan,
ed., Abhidharmakosabhasyam of Vasubandhu (Patna: K. P.
Jayaswal Research Institute, 1975), p. 124.
Visuddhimagga of Buddhaghosacariya, ed. Henry Clarke Warren,
revised by Dharmananda Kosambi (Cambridge: Harvard University Press,
1950), p. 477, par. 187.
(52) Bhattacharya, The Yogacarabhumi of Acarya Asanga, p. 26.18-19.
(53) Ibid., p. 25.13-14.
(54) Ibid., p. 109.13-15.
Vasubandhu, in Derge edition of Tibetan Tanjur, Sems tsam, vol. Ri,
f. 135a-5, 6: / de yang nyon mongs pa can gyi yid kyi rnam par ses pa
skye ba'i srid pa la dmigs nas bar ma do'i srid pa 'gag par 'gyur ro /
brgyal ba de la zhes bye ba ni yid kyi rnam par ses pa khu ba clang
khrag dang Ihan cig grub pa dang / bde ba gcig par 'gyur ba ste / yid
kyi rnam par ses brgyal par gyur pa de la brten nas yid kyi rnam par ses
pa gzhan nyid 'jug par 'gyur ro /
Vasubandhu, Derge, Sems tsam, Ri, f. 69a-5,6: / de'i phyir rnam par
ses brgyal ba gang yin pa de ni yid kyi rnam par ses pa ma yin gyi / de
ni rnam par smin pa'i rnam par ses pa ste de sa bon shams cad pa'o /
Gunamati, Pratityasamutpadadi-vibhanganirdesa-tika, PTT, vol. 104,
p. 337.3-3: / rnam par ses pa'i tshogs drug go zhes gsungs kyang yid kyi
rnam par ses pa kho nas nying mtshams sbrel ba yin pa.
Asanga alludes to this in the Yogacarabhumi early section, in
Bhattacharya, The Yogacarabhumi of Acarya Asanga, p. 25.1-2.
Buddhist Insight: Essays by Alex Wayman, ed. George R. Elder
(Delhi: Motilal Banarsidass, 1984), p. 330.
(60) Ibid., p. 330.
Gadjin M. Nagao, Madhyamika and Yogacara, trans. Leslie S. Kawamura
(Albany: State University of New York Press, 1991), chap. 6, pp. 61-74.
(62) Ibid., p. 62.
(63) Ibid., p. 70.
Among the editions, there is one in Chr. Lindtner, Nagarjuniana
(Delhi: Motilal Banarsidass, 1987), and another in Fernanda Tola and
Carmen Dragonetti, "Nagarjuna's Catustava," Journal of Indian Philosophy
See the essay by M. A. Mehendale, "Etymology of the Word Artha-,"
in his Nirukta Notes series 1 (Poona: Deccan College, 1965), pp. 42-46.
Bimal Krishna Matilal, Perception (Oxford: Clarendon Press, 1986),
Chakravarthi Ram Prasad, "Dreams and Reality: The Sankarite
Critique of Vijnana-vada," Philosophy East and West 43 (3) (July 1993):
(68) Kochumuttom, A Buddhist Doctrine of Experience, p. 93.
Cf. F. Th. Stcherbatsky, Buddhist Logic (reprint, New York: Dover
(70) Alex Wayman,
in Journal of the International Association of
Buddhist Studies 2 (1) (1979): 65-78.
Wayman, in Annals, Bhandarkar Oriental Research Institute
(Diamond Jubilee Volume), 1977-1978, pp. 387-396.
Alex Wayman, in Ernst Steinkellner, ed., Studies in the Buddhist
Epistemological Tradition: Proceedings of the Second International
Dharmakirti Conference, Vienna, June 1 1-16, 1989 (Wien: Verlag der
Osterreichischen Akademie der Wissenschaften, 1991), pp. 419-430.
Tripathi, The Problem of Knowledge.... (Varanasi: Bharat-Bharati,
1972), p. 333.
(74) Ibid., pp. 150-151.
Vasubandhu, Derge, Sems tsam, vol. Ri, f. 143b-4: / lus dang / lus
can dang / za ba po'i rnam par rig pa zhes bye ba de la / lus ni mig la
soys pa'i khams Inga'o / lus can ni nyon mong pa can gyi yid do / za ba
po ni yid kyi khams so /
vol. Si, f. 141 b-3: / sems ni rnam pa gnyis te / de la gcig
ni de'i sa bon rnams bsag pa yin no / gnyis pa ni de'i dmigs pa dang /
rnampa dang / bye brag the dad pa dag gis sna tshogs pa yin no /
Wayman, in Annals, Bhandarkar Oriental Research Institute, pp.
Alex Wayman, Analysis of the Sravakabhumi Manuscript, p. 173;
reprinted in Elder, Buddhist Insight, p. 340.
(79) Wayman, Annals, Bhandarkar Oriental Research Institute.
Boruah's essay appeared in Journal of Indian Council of
Philosophical Research 6 (3) (May-August 1989): 119-130.
Since this remark is important, I should cite the Tibetan (alluded
to in note 11 above): / rnam par rig pa sngon 'gro ba nyid du rtogs
pa'ang de Ita na nyes pa khyad par med pa kho nar 'gyur ro /
Copyright 1996 University of Hawaii Press
Copyright 1997 Information Access Company