On Vinaptimatra Passage In Samadhinirmocanasutra VIII

Author(s): L Schmithausen
Publisher: L Schmithausen
Catalog link: https://jainqq.org/explore/269662/1

Page #1

Sardinirmocanasira VIN 7
ON THE VINAPTIMATRA PASSAGE IN SAMDHINIRMOCANASUTRA VIIL 7*
By Lambert Schmithausen, Hamburg
1. In an earlier paper" I tried to show that Samdh VIII. 7-8 can be regarded as the starting point of the doctrine of 'cognition (or peception) only ---
. This paper is due to the most stimulating effect of two terms of fruitful collaboration with my dear friend and colleague Noritoshi Aramaki. Abbreviations: Bh, etc. Mahayanasamgrahabhisya (see $13) Both
Bodhisattvabhumi (d. U. Wogihara). Iwata 9 .
X ( RAU M BONNU Tokyo 1981 Jinag JA nagarble, Aryamaitreya kevala parivartalagyam, ed.. Now (see: None
M
Katano
HR (MMFERRAT O R ), Kyoto 1975, MAV(Ah) MadhyAntavibhaga(blsyn) (ed. Nagao). MSE Mahayanasangraha (ed. Lamotte).
o ll of the Tibetan text of MSg I and II with reconstruction into Skt. in Nigao
1982 (q. v.). Nagao 1982 W FOR E, Tokyo 1982. Notawa RARE. ARR OW MIERCURORID, Kynin 1957. S 1969 L Schmithausen, Der Nirvana-Abschnitt in der Viniscayaarahant der Yogach
26 rabbimih. Wien 1969. S 1973 id, Spirituelle Praxis und philosophische Theorie im Buddhismus, in: Zeitsclur. .
Missions wiss. u. Religionswiss. 57/1973, 161 ff. 5 1976 id., On the Problein of the Relation of Spiritual Practice and Philosophical Theory
in Buddhism, in: German Scholars on India, vol. II, 1976, 235 f (Pnglish version
of a part of S 1973). S 1982 id., Versenkungspraxis und erlosende Erfahrung in der Sravakubhomi, in: Epl
phanie des Heils, hrsg. G. Oberlimmer, Wien 1982, 59 s. Serdh Sandhinirmocanasitra (ed. Lamotte). Sampdhvy Samdhinirmocanastravyakhyann of Byan chub rdzu "phrul (see JManag. 1): Tj
No. 5845 # * . BRAU E R . Tokyo 1931.
Taisho ed of the Tripitaka in Chinese (TI) etc. see 84 and 5
Peking-Tanjur, ed. D. T. Suzuki, Tokyo/Kyoto 1955-61. U. U, see 14 Y
YogAcarabhumi (ed. V. Bhattacharya).
Tibetan translation of the Yogichrabhomi: Tj Nos. 5536-5543. 1) S 1973, 167 f.; S 1976, 240 f.
Saraki

Page #2

Samulinermocanastra VII 7
435
434
Lambert Schmithausen (vijfaptimatra) proper, documenting at the same time that this doctrine was developed in the context of a specific yogic practice of deliberately visualizing and contemplating, in meditative concentration, certain kinds of images and afterwards still in meditative concentration effacing them. The aim was either as in the Sravakabhomi—to achieve an ever clearer reconstruction of these images, or in certain (mostly Mahayana) contexts of the Viniscaya. samgrahap -to become free from all phenomenal experience in order realize transphenomenal True Reality (tathata). To efface images at will is possible because they are creations of one's own meditative concentration-nothing but seeing or knowing or recollecting [them)(darfana matramvaj nanamatram w pratism (i) matram ), as a Satra quoted in the Sravakabhomi puts it's In the Samdhinirmocanastra, however, what alternates with visualization cum contemplation of images is not so much their effacement as acts of meditative contemplation contemplating these visualizations and contemplations themselves. The idea of such "reflexive" acts of meditative contemplation can again be traced to the Sråvakabhim. But in the Sravakabha mi this "reflexive" contemplation does not seem to have any close connection with image visualiration, and its aim is rather to realize impermanence, unsatisfactoriness, etc.. of even the meditating mind itself, in order to fully comprehend the four Noble Truths. In contrast with this, in the Samdhinirmocanastra "reflexive" contem plation envisages the awareness of the ideality of the images perceived in meditative concentration the fact that they are not different from the contemplating mind in order to effect the realization of universal True Nature (tathata), understood, by Samdh VIII, as the ideality of all phenomena". It is in the context of the Samdhinirmocanastra's setting forth its view of the ideality of images vizualized and contemplated in meditative concentration (VIII.
7) and of the subsequent generalization of this view (VIII. 8) that the concept 'vijnaptimatra- obviously alluding to the "nominalist" doctrine of 'denomi. nation only' (prajnapti-matra) of the Bodhisattvabhominis introduced.
2. The crucial passage in which the term first appears is in the initial part of Sandh VIII. 7 where the question is put whether the images perceived in meditative concentration (samadhi gocarapratibimba) are something different from mind (citta) or not. The answer is that they are not, and the reason for their non-difference is stated as follows (Lamotte's translation and text):
"Parce que ces images ne sont rien qu'idée. J'ai dit que l'objet de la connaissance se définit (Idée-sans-plus)" (grugs brilan de ruam par rig pa sam du tad pa'i phyir tel .....nam par ses pa'i dmigs pa nam
par rig pa sam gyis rab tu phye ba yin no les das bsad do ID. 3. To be sure, Lamotte's translation of the second sentence fits both Hsuantsang's Chinese version (see $ 12. 1) and the Tibetan text as Lamotte gives it, but the latter is not quite correct. As is clear from a look into the block prints of Peking") and Derget, the transmitted text has not ram par ses pa'i but rnam par les pa ni, a reading confirmed by Yoshimura's edition of Samdh VIII", by its quotation in the Yogackrabhami, by the commentaries of Jha nagarbha") and Bynh chub rdzu 'phrul!), and by the quotation of Samdh VIIL 7 in the Mahayanasamgraha (II. 7, where Lamotte again reads pa'i but indi cates in the critical apparatus that the xylograph as well as the Bhasyal have pa ni). In view of this almost if not altogether unanimous testimony of the sources the reading pa'i has, in spite of the arguments adduced by Nozawa, to be abandoned in favour of pa ni:
10) S 1973, 166 and 171 ; S 1976, 243 1. 11) Peking Kanjur (ed. Suzuki) vol. Au 2961. 12) Derge Kanjur (Nyingma Ed, Dharma Publ. 1981, vol. 18), mDesde an 27
(
leaf 1232)
2) S 1982, 62 f, esp. 67 f. 3) S 1973, 169 I.; S 1976, 242. 1) Sravakabhomi (ed. Shukla), 199, 171.; S 1973, 167; S 1976, 239 f. 5) Samdh VIII. 9. 6) S 1982, 79. 7) As this ideality of the image is prefiguration or partial manifestation of tathata (Samdh
VUIL 9), the famathalipayand practice of the Samdh would seem to be also related to the Samathawi Malyand practice of t h 109 f. which seems to propound an alternation of contemplation of the features (or images?) (nimitta) of dharmas and of concentration on their ?)
sible Reality-as-such ( stomatra, =tathatamate acc. to BoBh 41, 18; p. 293, 27 f.). 8) Samdh Vill. 9 (end). 9) Samdh VIII. 20. 2. 3.
13) Sh. Yoshimurn, Comparative Study in Chinese and Tibetan Texts of the Yoga-vibhang
parivarta in the Samdhinir mocanastra, Kyoto 1959, p. 7. 11) Y, 174 b 8. 15) Janeg. 22, 31. 16) Sampdhvy co 193 b 2 17) Cp. also MS n. 62+n. 3; Sasaki, p. 48 (Yamaguchi's ed.). 189 Cp. also Iwata, 78+n. 3. 19) Nozawa, 206 n. 2. 20) This is also the opinion of Nagasawa, Katano and Suguro! Me Iwata, 78 . 3). Co also ns.
13 and 17 of this paper.

Page #3

Lambert Schmithausen
[T(a)]
rnam par ses pa ni dmigs pa rnam par rig pa tsam gyis rab tu phye ba yin no (zes has bśad do) II.
4. This text admits of two translations:
[T1] "Mind (vijnana) is constituted by (prabhavita", is characterized by, or consists in) mere cognition (vijnaptimatra) of an object (alambana)."
436
[T2] "Mind is an object that is constituted by mere cognition." The Tibetan version of our sentence in MSg II. 7, viz.
[T(b)] rnam par les pa ni dmigs pa rnam par rig pa tsam gyis rab tu phye ba can yin no (zes nas blad do) II,
involves a third translation:
[T3] "Mind has an object that is constituted by mere cognition",
5. As the Tibetan translation is not unambiguous, the question arises what the Sanskrit original unfortunately lost-may have looked like. Nozawa, in view of Hsuan-tsang's rendering (see § 12: [H]), proposed
'vijnanalambanam vijñaptimatraprabhavitam (...), but this reconstruction is irreconcilable with what has to be accepted as the correct text of the Tibetan translation. It is also in conflict with Buddhaśanta's rendering of the quotation of the sentence in MSg II. 7:
[Bu] (?) 唯所明謙説。
The value of Buddhasanta's version lies, among other things, in the fact that it is frequently (though not always) literal to the extent of preserving the Sonder wordyof the original Sanskrit without adapting it to the requirements of Chinese syntax. Thus, if there is no cogent reason against it, reconstruction should follow the word order of [Bu]. This means that the sentence we are concerned with started with alambana, wheres vijnana came in the end, immediately 21) For this term see S 1969, 109 ff.; D. Seyfort Ruegg, la théorie du tathagatagarbha et du gotra, Paris 1969, 347 ff.
22) Thus Nozawa, 192 and 206 n. 2; Suguro in: Osaki gakuho 129/1976, 43; Similarly Katano,
80.
23) Nozawa, 206 n. 2.
24) Cp. also Odani in: IBK 57/1980, 419 n. 10; Takasaki in:
(8: M),
Tokyo 1982, 37 n. 24: Vijnanalambanam vijnaptimatrena prabhavitam iti maya delitam; MSgs 63 n. 1: Vijnanalambanam hi vijnaptimatraprabhavitam, which (in contrast with the proposal in the text which is Aramaki's) seems to be Nagao's own view (cp. Nagno 1982, 290 n. 1+m).
25) T vol. 31, 101 a 22 f.
26) Cp. Aramaki in: Miscellanea Indologica Kiotiensia 4-5/1963, 36.
J
Samdhinirmocanas@tra VIII. 7
before what would correspond to tes as bad do. This is confirmed not only by the Chinese version of Dharmagupta (see $12: [Dh]), but also by the Tibetan version of both the Mahayanasamgrahabhasya and the Mahayanasam graha-upanibandhana ad MSg II. 7 which contain pratikas where the word rnam par ses pa (vijnana) immediately precedes the concluding tes nas biad do (see § 13: Bh, (3a), and § 14: U, (le) [for dmigs pa see § 14. 2]). 6. Accordingly, as on a previous occasion, I suggest the following recon
437
struction:
[S] 'alambanavi jnaptimatraprabhavitam vijanam.......
Now contrasting with this, Aramaki in his reconstruction of MSg II. 7 proposes:
(tad) alambanam hi vijñaptimatraprabhavitam vijnanam ity aham vadami.
Disregarding, for the time being, the problem of the equivalent of tes nas blad do (for which see §9), I should like to cut down, for the purposes of the present investigation, the alternative suggested by Aramaki to [A] *alambanam vijñaptimatraprabhavitam vijnanam.......
6.1 I have ignored hi though it seems to be supported by the final particle # in Bodhiruci's and Hstian-tsang's Chinese versions (see § 12: [Bo] and [H]). But this may have been induced by the question "why?" preceding (in Bodhiruci: immediately preceding) our sentence, or by logical considerations. Besides, there is no trace of hi in the commentaries. And even if hi had actually been there, the source material does not offer any clue as to its posi tion. In view of the result of the following investigation (88 13ff.)showing that there is textual support for both [S] and[A], it would seem that, if there was any hi, it can have followed neither alambana (impossible in [S]: compound!) nor -prabhavita (impossible in [A]: hi should be the second word). .6.2 As for tad(-?), the fact that it is not confirmed by the Tibetan version (including the pratikas in the commentaries) which is usually quite meticulous in rendering pronouns is a strong argument against it. There is, on the other hand, evidence for tad in Paramartha (t&, s. §12: [Pa]). But his version is the least literal of all and has most probably taken over tad from the Bhasya (see § 13. 1). The same is true of Dharmagupta whose gloss (s. 8
27) S 1969, 110; cp. S 1973, 168; S 1976, 240 f.
28) MS 63.
29) This is also true of the vocative indicated, for our sentence, by the Tibetan translation (Maitreya) and by Hsuan-tsang (kulaputra).

Page #4

Sardhinirmocanestra VIL 7
Lambert Schmithausen 12: (Dh]) too may be regarded as an expansion of tad(.)). Finally, tin [Bu] (595) to be sure corresponds to tad, but I wonder if it could not represent the fad of the preceding tat kasya hetoh.
The difference between [S] and [A] is thus essentially reduced to the presence or absence of the anusvåra at the end of the word alambana
7. It may seem disproportionate to write a paper on such a trifle as one single anusvara, and paramarthatah I should, of course, agree. But the question whether anusvara or not is decisive for interpretation of the sentence, which on its part is crucial in the context of what is, in all probability, the oldest extant passage enouncing, by the very term, the doctrine of pijnaprimatra. i. e. the central doctrine of Yogacara-Vijñānavāda. Considering the internal consistency of the passage and historical plausibility, I should not even hesitate to assume that this passage is actually the first literary expression, in Yogacara proper, of the generalization of the ideality of images visualized in meditative concen tration into an ideality of all phenomena (see $ 1). Yet, the sentence under discussion the Buddha himself seems to corroborate the fact that the images perceived in meditative concentration are nothing but cognition (vijnaprimatra), i. e, do not exist apart from the mental act of cognizing or perceiving them, by the remark that he has already taught or explained (des has bfad do) something of the kind. As this sentence contains the term vijnaprimaira, it creates the impression that the Buddha refers to an earlier enunciation of the idealist doctrine of cognition-only, and this would seem to imply that Samdh VIIL 7-9 is, against what I suggested above, not the first passage in which this doctrine was set forth. But determining the precise content of such an earlier teaching or explanation, alleged or real, depends on how one has to und stand the sentence under discussion, and this again depends on whether the anusvira was present or not.
8. Now, if the sentence had the form of [A], it would be unambiguous. From a merely formal point of view it could, to be sure, be interpreted in the sense of [T 2) (see $4). But such an interpretation does not seem to make much sense in the present context; for the thesis is not that vijnana is an object but that the object is not different from) vijnana. Thus, alambana should be regarded as the grammatical subject, and the sentence taken to mean:
[A1] "The object is constituted by nothing but cognition,
[therefore it] is mind (vijnana)." No doubt, such a statement is unequivocally idealist. And if the perfect blad has to be taken seriously, (A) would seem to imply that the sentence under discussion refers to an earlier authoritative enunciation of the doctrine of vijnaprimatra, and that Samdh VIII. 7-9 is not the literary expression of the first discovery of this doctrine.
9. The only way out of the difficulty--if (A) is to be upheld would be Aramaki's suggestion to understand ias bśad do as corresponding not to a Skt. preterite but to present. This suggestion can be based on Kasyapa.parivarta $$ 64, 65 and 127 where ses nas biad do actually represents Skt. aham....... vadami. Yet, blad is, at least from a morphological point of view, a perfect (or
- not applying to the present case- a future), and usually represents preterite forms. Moreover, in a passage corresponding to a pratika of the last part of our sentence in MSgBh, (see $ 13: Bh, (3a)), Hsuan-tsang's translation ( 13: Bh. (3a)) has the passive expression of Therefore, though I do not exclude the possibility of a present, I should consider preterite, e. & 'iti maya delitam or the like, at least equally possible. This means that with [A] the undesirable consequence indicated in $7, viz. that there must be some earlier enunciation of the doctrine of vijAaptimatra, would at least remain possibility,
10. On the other hand, this consequence does not at all apply in the case of (S); for as I have already suggested previously", [S] would allow to under stand the sentence under discussion—no matter whether it contained a preterite or a presenta a skilful utilization of a widespread Abhidharmic definition mind (vijnana). This definition, which is, in principle, etymological, defines vijnana as "the act of cognizing (lit.: making known [sc. to the cognizing 'person'] its (respective) object" (alambana or visaya (prativi japti), i.e.
er
of
30) I do not dare decide whether a finality chay vadami would, from the syntactical/stylistical
point of view, be unimpeachable (ns tam aham......iti sadami in Kasapaperivarta 64 etc. obviously in) if, as in the sentence under discussion, no stress on "
T intended. --That there actually was, in the original, some form of the pronoun of the 1st the Samdhinirmocana sentence all the versions have "T" (has, .) whereas in the Upanibanduna both versions have only the verb without personal pronoun (see $ 14: U. (3)......tes had do, U. (3) R7 ......) fnct that seems to be explicable only by an opposition maya delitam (Samh)
detitam (U) or aham adami (Sad) + vadami (U). Y son t rendered almost 31) S 1973, 168; S 1976, 241.
cerlain by the fact that in
290) Actually, the pratika of this sentence in the Bhagya (nee
13: Mhes (1)) bus #instead

Page #5

Lambert Schmithausen
Sardhinirmanastra VIII 7
visible things in the case of visual perception, etc.). If this definition is used with a view to distinguish the function of mind (vijnana, citta) from that of mental factors associated with it (caitta, caitasika-dharma), it acquires the nuance that mind cognizes its object as a whole or as such, without grasping or singling out or stressing its peculiar features, and this nuance could be expressed, as is documented also in non-Yogacara sources, by adding the word -matra to the word for "object" (alambana, etc.)" or even to the word for "cognition" (vijnapti, etc.)". Understood in this way, the sentence under discussion would not substantially depart from traditional Abhidharma, for it would mean nothing but: "I (have) declare(d) that
1
32) Cp, g. Prakarana (T vol. 26) 6935 CHRIRAM. M T 98) =Y 4, 5 (calpur
viflanan katamat? ...... papratiwijiapis): Abhidharmasamuccaya (ed. Pradhan) 12,7.: cakurvijana halamat? ....
F lamband Aratinjani: Y, 1895 (wi Manaviy prativi Aapi); Abhidharmakosa(bhasya) (ed. Pradhan, Patna 1967) 11, 6., vijay Prati wij afir-visayagi visayaw pratiwijaptirpalabdhir...... Pacarkandhaka(Tj sems-tsom si) 16 b 8: mam par les m al tema d mig! Mirman par rie Maloll ( vijnanay katamat alambawavijnaprih!). The Yogicrabbamivyakhya (Tj nemstram yi 86 b 5.) refers this de finition to the Zas Ita bu ten byn bli mdo (T vol. 2, 11 e 96 cp. Samyuttanikaya III, 87: Khajanlynutta): ram par les byed il ram par les par dyed My ma del Myir ram par des pall Mhu o to by stel tir ran par les te ma
am dat seams...... (wjaali wanati, famad vijanaskandha ity wegate mea wijani ? romani, etc.). 33) Y 69, 16: tatra kalan rawlakanam wijnen Wijn Mayall Wijna tati is abviously
the verbal equivalent of japi, and thus to be taken a "makes known (to himself", "cog
nizes": cp pretipadyate in 1. 18 ft. of the same passage. 34) Cp, ex. Y 59, 16 ff.; MAV(Bh) L8; L de la Vallée Poussin, Abhidharmakon I. 30 n.
3; id, Vijaptim.Atratasiddhi, 296. 35) MAVBh 20, 19: arthamdireditir mjanom: Madhy Antavibhigafika (ed. Yamaguchi) 31,
10: mareamiro falabdhib: Prasannapada (ed. de In Vallée Poussin) 65,2: arthamatra dariansy dittasys syarah: Sanghabhadra, Nyayanuskra (T vol. 29) 342 & 14:10MT81*
n : Yafomitra, Abhidharmakokavyakhya (ed. Wagihara) 38, 24: (wijam) spalahdhir taalumatragraharam, Skandhin, Abhidharmintra (ed. v. Velthem, Louvain 1977) 106, 11. krugla og myul gyi dhe po tm 10 or ram Mr PR M irnam Aer les pse (rendi starapraturtir WiMam). 36) Abhidharmadipa vrti) 78, 11: Palabdhimdir Nielttan. I admit that earlier evi
dence would be welcome; but in any case the fact that the source is quoted is a Vaibharika text (as are also some of those quoted in m. 32 and 35) shows that the wording of the Samdhinir mecanestra sentence under discussion could have easily been formed as well as un
derstood on the basis of the Abhidharmic definition of wij Ana 37) Of course Sotra can have recourse to an authoritative Abhidharmic formula only by put
ting it into the mouth of the Buddha. There would be no need to postulate a corresponding Sotra passage as the source of the Sandh. But seen 32 (end)!
[S1) mind (vijñāna) is characterized as Cor: consists in) merely cognizing (lit:
making known) [its) object (without stressing any of its peculiarities)." But of course the Sotra does not mean that. It has deliberately chosen the expression 'alambanavijnaptimaa (instead of, e. 8., alambana maravij Aapti.) in order to evoke, in the context of the preceding sentence that had enounced the ideality of the images perceived in meditation by qualifying them as ve japtimatra, an idealist understanding of the word vijfaptimatra. Therefore, in the context of the Satra the sentence has of course to be understood as: [S2] "(.....) Mind (vijnana) is characterized by Cor: consists of) mere
cognition of Cits) object (without there being any real object)" - an interpretation which would correspond to [T1] (see $ 1). Or, if one prefers to supply a virtual abstract suffix after vijnaptimatra.: [S3] "(.....) Mind is characterized by the fact that its] object is nothing
but cognition." Finally, [T(b)]/[T3J (see $ 4) seems to understand the compound 'alamba navijnaptimatraprabhavita as a bahuvrihi with an irregular sequence of components, translating as if there were 'vijnaprimairaprabhavitalambana, i.e.
[S4] "C......) Mind has an object which is constituted by mere cognition."
11. Yet, the preceding interpretation of the sentence under discussion is entirely dependent on the correctness of the reconstruction (S). IF(A) is to be preferred, it would be utterly baseless. It is therefore necessary to check all the pertinent sources at our disposal in order to find out which of the two reconstructions they support.
As for the Tibetan translations ( 1), it has been pointed out($ 10 that [T (b)] supports [S]. The same is true of [T(a)] if taken in the sense of [T1]. But starting from [T 2] it can also be understood as an awkward (see $ 8) rendering of (AJ. The Chinese translation of Buddhasanta (see & 5: (Bu]) is altogether ambiguous. Thus, additional evidence has to be looked for by scrutini zing the other Chinese versions ($ 12) as well as the Indian commentaries both on Mahayanasamgraha II. 7vir. the Bhasya of Vasubandhu (13) and the Upanibandhana of Asvabhava (14), and on Samdhinirmocanastra VIII. 7, viz. the commentaries of Jhanagarbha ($15) and Byan chub rdzu 'phrul (16)". 38) Perhaps somehow on the analogy of the first set of compounds treated in Wackernagel, Alt
indische Grammatik, I, 1, $ 116. 39) The commentary scribed to Asanga (Tj No 5181) does not yield any pertinent info ma
tion

Page #6

Samdhinirmocanestra VIII. 7
443
rather seems to be a syntactically simplified ad sensum translation of (S) under stood in the sense of CS 3] or, more likely, (S 4]: To say that vijnana has an object that is manifested by cognition only is, of course, equivalent to saying that the object of vijana is manifested by cognition only. An addi tional stimulus for this sytactical transformation will become manifest in $13.
412
Lambert Schmithausen After summing up the result of this investigation (17), I shall, moreover, try to evaluate the readings in terms of intrinsic probability (59 1888.).
12. The Chinese translations of Bodhiruci (Bo), Paramartha (Pa), Dharmagupta [Dh) and Hsüantsang (H) (for Buddhasanta see $5) run as follows":
[Bo] (EN) ARUS [Pa] R ite [Dh]
SOPRA
TERMO 121 I confess that I am unable to draw any useful information from [Bo] which remains obscure to me.
12. 2 [Pa): "I (have) declare(d) that there is only mind/cognition - jhana)", [that] this object which appears as some thing) visible is mani fested by mind/cognition (vijfapti?)." This translation is not quite literal, but the fact that the sentence is split up into two statements points to [A]. For some reason, Paramartha would seem to have reversed the order, i.e. placed vijnanam first and combined it with matra) while at the same time
-matra in the second part of his translation which would have to represent the subject (alambana) and the first predicate (vijnapti matraprabha vita) of (A)
12.3 [Dh): "The object (alambana) of concentrated mind is manifested by cognition only (vijnaptimatraprabhavita); I (have) declare(d) that it is mind (vijnana)."
This version unambiguously supports [A], understanding it precisely in the same way as suggested in $8.
12 4 [H): "I (have) declare(d) that the object (alambana) of mind (vijnana) is manifested by cognition only (vijnaptimatraprabhavita)."
This rendering coincides neither with [A] nor with (S), but as it takes the sentence as one predication it is hardly explicable on the basis of (A) but
18. Mahāyānasamgrahabhāsya ad MSg II. 702 Bh,
Bh, (1) dgors pa ies par 'grel Bh.
Вhph
Bhd pa'i mdo las kyanit/ (1) MORE! (1)-- Ornam par ses pa ni
P . dmigs pa ruam par rig
# R pa sam gyis rab tu phye ba car no tes das bsad
RE) do) tes gsuris pa des na (2n) dmigs pa rnam par
182c 16f.: rig pa sam gyis rab (2a)
(2a) ** (2a) tu phye ba can de ni
W 19 . Wiko (2b) rnam par rig pa
(2) ma. (26) isam nid de
AMI ||(20) (2) don gyis stor pa
(26) C UF (20) tes bye ba'i tha (2c) . (2) Rio tshig go!
L(2d) # 1 (3a) rnam par ses pe (3a)
(3a) tes nas blad do ses
(3b)
1820 14f.: bya bali (3bwa*.
(3a/b) 180 J(3b) rnam par ses pa (30) L.(30)
A R smos pa desni (3a) PUT (3) tint ie dzin gyi (3)
(30)WH O spyod yul gyi mnam R . W ALA
par ses pa bstan toll ( 51 ) 13.1, In (2a) all the Chinese versions (Bb.) seem to support (A). Dharma
4
10) (Bo): T vol. 16, 674 c 24; (PA): T vol. 31, 118 b 27 f.; (Dh]: ib. 285 b 221.[H): ib.
138 b 8: vol. 16, 698 b 2; vol. 30, 724 6. Cp. Sasaki, 31. 41) It should be noted that this part of Paramartha's rendering of our sentence coincides with
his rendering of cittametram idam in the Dalabhomikastra quotation in the beginning of
MS II. 7. Ain) Op . 43a. 12) Cp the Tibetan rendering of Hsian-tsang's version in the translation of Ydents'e' (Ven.
tshig's) commentary on the Samdh (Tj thi 117 6 61.): ......es de par ses pidigs i rom far rig pa sam las smak bar blad poli Myir roll.
11
3;
3) Bh: Tj sms-tsam li 171 b1; Bh: T vol. 31, 338 € 22-25; Bhs: ib. 285 b 29
Bhreib, 182 c 14 Cp. Iwata, 78.; Norawa, 2017

Page #7

Samdhinirmocanastra VIII 7
415
.
in
Lambert Schmithausen gupta (Bhow) is quite unambiguous: "This object is manifested by cognition only", i. e. "lad alambana vijAlaptimatraprabhavitam. Similarly Paramartha (Bhr.), whose rendering, adding "Cobject) consisting in (something) visible" and omitting matras, is identical with his translation of the respective part of the basic text (see $ 12. 2). This last is also true of Hsüan tsang (Bhu) who, here too, has "the object of mind (vijnana)" instead of "this object", but this can, in the present case, easily be explained as an attempt to concretise tad under stood as the prior member of a tatpurusa compound tadalambana. It even appears quite probable that it was precisely this interpretation of (2a) in the Bhasya that stimulated the syntactical transformation in his rendering of the basic text (cp. $ 12. 4).
On the other hand, the Tibetan version (Bh.) of (2) cannot be interpreted in the sense of (A) because as in (T(b)) (see $ 4) the particle can compels us to take dmijas pa rnam par rig pa sam gyis rab tu phye ba can as a unit, i. e. presupposes 'alambanavijflaptimatraprabhavita taken as a bahuvrihi com. pound in the sense of (S4). Thefore, in Bh, alambana cannot be the sulject of (2a), and the demonstrative pronoun de cannot be its attribute but only refer to another subject which can hardly be anything but "mind" (vijnana).
13.2 In (3) too Bho (Bhdeviates to the extent of being useless for our purpose) unambiguously corroborates (A) when it states that on account of its being included in mind the object (perceived in) meditative concentration (samadhi gocara) is mind (vijnana), i. e. takes vijnana as another predicate the grammatical subject of which is, as in (2a), the objective support i. e. alambana, though in this place represented by samadhi gocara. Bhy ("Moreo ver: By using the word) 'vijnana' (the Buddha) indicates that the object of mind in meditative concentration he has spoken of his manifested by cogni. tion only......"), inspite of its different and certainly correct) understanding of (3b) and though substituting, in the predicate, 'mind' (vijnana) by WI (+ vijnaptimatraprabhavita), still agrees with Bhon in regarding the object of
meditative concentration as the grammatical subject of the sentence.
Bh, here too, does not support the Chinese versions. It rather states that by using the word 'vijnana (vijnanagrahanena ?) the text refers to the vijana of the objects (perceived in) meditative concentration (samadhi gocara).
13.3 It is rather difficult to decide which of the two interpretations of the Bhasya is the correct one as long as the Sanskrit original is not available. Till then, we are forced to conjectures.
13.3.1 As for (2), a comparison of the various versions suggests the following original :
(20)*tad (Bhy: tad-Salambanan (Bh: na vijnapti matraprabhavitam (2b) vijñaplimatram eva (2) arthasanyam (2d) ity arthah/ This sentence is understood by Bhe as follows: "This (mind) which has an objective support (alambana) - that is constituted by mere cognition is nothing butmere cognition, i. e devoid of an (external) object Cartha)." Or, less awkwardly: "'This (mind) has an objective support Calambana) -
that is constituted by mere cognition' (means that it is nothing but mere cognition, i. e devoid of an external) object (artha)." This rendering, needless to say, would support (S). However, matters are not quite so simple as this. For by not repeating the subject in its proper place the commentator shows that it was the predicate that he wished to comment on; but, then, why did he at all refer to the subject, by means of a pronoun at that which was liable to be misunderstood as the first member of a tatusa compound tad-alambanao! Thus, the reading coinciding with (S) is not fully
the il
11
11
Bh
431) One may even get the impression that Paramartha's rendering of wijnanam in the Satra is
influenced by Bhasya (2b) (for which see $13. 3. I) and that what, in Bhat first sight
seems to correspond to (3) could just as well be a rendering of (2b)-(d). 41-44) Or: "that what has been declared by him is that the object of mind in meditative con
centration......"p. the Tibetan translation of Yuan-ta'e's commentary on the Sardh (Ti thi 118 a 3): ......rnam par fer pa smos pa mi das ruam par les pa'i pyod yul mi ties par ruam par rig pa sam la(s) shall gi......tes blad par rab ta bstan palo Il.
45) des na (at the end of (1)) which in Bh, looks as if introductory to (2) but is missing in
Rh, dnes mot admit of antisfactory interpretation if taken with (2). I should prefer to regard it as representing an anene that is to be construed with the preceding senteixe which I should retranslate into Skt. as follows:
(1)*Sandhinirmocanastre 'py wktam Balambana.......) ity anena i. , : "[cognition-only) has also been enunciated in the Samdhinirmocanastra, (vit. ) by the [sen
tence] (.....)." Bh, would have ignored the danda and wrongly have understood (1) as a reason for (2)
whereas in reality (2) is an explanation of the sentence quoted in (1).

Page #8

446
Lambert Schmithausen
satisfactory here.
On the other hand, the reading of Bhoh and Bhr. (*tad alambanam vijñaptimatraprabhavitam) does not seem to be free of problems either. For, since in the case of these versions we have to start from (A), (2a)-leaving aside tad would be nothing but a repetition of the wording of the first part of the Satra sentence, i. e. a quasi-pratika to be glossed in (2b)-(2d). Now, tad could have been either a part of the wording of the Satra, or it could have been added, by the commentator, in an anaphorical function. The first possibility is rendered altogether improbable by the absence of such a pronoun in the Tibetan text (see § 6. 2). As for the second possibility, I for one cannot detect any motive for the use of an anaphoric tad (as an attribute of alambana) in the present case; for it would be altogether superfluous to refer, by tad, to the word alambana in the Sutra sentence of which (2a) is, in this interpretation, nothing but a quasi-pratika; and there is, in the Bhasya, no occurrence of alambana or the like in the preceding context. Besides, I wonder if "arthasanya in (2c) can really be predicated of the objective support (alambana), as it cer tainly would have to be in this version.
This leaves us with the reading presupposed by Bha viz. tad-alambanam vijЯlaptimatraprabhavitam, tad- representing vijana-. This reading is, to my mind, perfectly unobjectionable. But it cannot be a mere repetition of the wording of the Satra sentence supplemented by tad- because in that sentence the noun which tad- must represent, viz. vijñana, fills a different syntactical position. The Bhasya sentence (2a) can therefore only be an explanation of the Sutra sentence, an explanation which in its turn is further explained by (2b)+(2d). Nor would, in the case of this interpretation, the fact that 'arthasanya would seem to be preferably taken as a predicate of vijñana | vijñapti" and not of alambana raise any difficulty; for as vijnana is referred to by tadand as the sentence is an explanation of the whole Sutra sentence the subject of which is vijñana, the gloss in (2b)-(2d)-which I should prefer to take, with Bhoh, as a syntactical unit governed, as a whole, by ity arthaḥ-could no doubt be understood to qualify vijnana, not alambana:
"What the Sutra says is that] its objective
support (i. e. the objective support of mind, vijñana)
is constituted by cognition only; the meaning is: (mind
46) Cp. MSg II. 6; II. 14.
lad
that
Samdhinirmocanastra VIII. 7
is) nothing but mere cognition, devoid of an (external) object (artha)." This interpretation of (2)-which is, by the way, confirmed by the fact that an investigation into a similar sentence in the Upanibandhana leads to a similar that conclusion (see § 14. 3) would imply that Bhasya (2), in spite of the fact) the reading alambanam deserves to be preferred, yet cannot be utilized as a support for (A) but is rather an explanation of (S).
13.3.2 The original wording of (3) is difficult reconstruct in a reliable way, but as far as essentials are concerned Bh, seems to make fairly good sense: "I have declared mind......: What is indicated
(dyotita" or the like) by this employment of (the word)
'mind' (vijñanagrahanena?) is the mind (that cognizes) the objects (perceived in) -
meditative concentration (samadhigocara)."
I. e. According to Bh, the Bhasya wants to make it clear that vijnana in the present passage does not mean vijñana in general but refers to the special case of the cognition or perception of mental images in meditative concentration a remark which seems to be quite to the point because it is only this specific form of mind that is treated in Samdh VIII. 7, ordinary mind being discussed only later (viz. in VIII. 8). This interpretation of (3) fully accords with the fact that (2) has most probably to be understood as a complete explanation of the whole Sutra sentence (see § 13. 3. 1). (3), on the other hand, would not be concerned with the explanation of the meaning of the sentence but would be an additional remark for the sake of reminding the reader of the specific context of the sentence.
447
As against this, the Chinese versions, esp. Bhok, seem to be based on a diffe rent syntactical interpretation, or on a different reading, of the words samadhigocara and vijñana (e. g. they might have read "samadhigocaro vijnanam dyotitah instead of "samadhigocaram vijnanam dyotitam, but there are other possibilities). Such an interpretation or reading would however seem to be inseparably linked up with the assumption that the function of (3) is to repeat, and comment upon, the second part of the Sutra sentence (by interpreting vijnana as a second predicate). It would therefore presuppose that (2) repeats, and comments upon, the first part of the Sutra sentence only. Thus, it would not agree with the result of the investigation of § 13. 3. 1 according to which 47) Cp. A. Hirakawa, Index to the Abhidharmakośabhäşys, pt. 2 (1977), 119f. (s. v. ); pt. 3 (1978), 97 (s. v. ston pa).

Page #9

Sardhinirmocannatra VIII. 7
449
448
Lambert Schmithausen (2) is originally an explanation of the whole Satra sentence. Accordingly, in (3) too the interpretation or reading supporting (A) can hardly have been the original one. It is interesting that Hsuan-tsang's version of (3), though obviously agreeing with Bhoh as regards the syntactical relation of samadhi gocara and vijnana (see 13. 2), is yet de facto nothing but a specification, with reference to the objects of meditative concentration, of the general formulation of his rendering of the Satra sentence (which coincides with his rendering of Bh (2n). Thus, from the point of view of purport, Bhe does not seem to contradict the interpretation of Bh (3) derived from Bh.
To sum up, a closer investigation into the Bhasya on the quotation of our Samdhinirmocanestra sentence in MSR 117 shows that what is likely to have been its original wording and meaning does not support (A) but rather (S).
14. Mahayanasamgrahopanibandhana on MSg II. 70
((la) dmigs pa nam par rig pa sam gyis (le) RRRR
rab in dbye (D: phye) ha can ten bya ba ni (1a) WW W . (1){(16) phyi rol gyi dmigs pa med pa'oll
(Ic) rnam par ses pa dmigs pa yin par nas
bśad doll tes bya ba ni
(2a) 'di ltar dmigs pa de rnam par rig pa (2a) (RIRE (2) Isam gyis rab tu phye ba ni
MR. (2b) de'i Mo bo nid ces bya ba'i tha tshig go! (2b) WU R AA (3) rnam par les pa ni dmigs par snart ba (3) tsam gyis rab tu phye ba yin par bfad do ERR Les bya ba'i tha Ishig go!
(E84 ). 14.1 It is advisable to start with (3) because it is quite unambiguous in U.. The sentence is intended to give the purport of the Sotra sentence as a whole:
"The meaning is: Mind has been taught to be characterized -
by merely appearing as the object." If we choose pratibhasa to render snart ba (U. #R?) and keep to the word order of the Satra sentence which is paraphrased, the crucial part of U. (3) would correspond to Sanskrit
(3) 'alambanapratibhasamatraprabhavita vijnanam. This is an unambiguous support for (S), and there is, from the point of view of U, (dmigs par), no room left for reading alambanan. U. (3) does not conflict with this since Hsuan-tsang, in rendering this paraphrase, follows the same pattern as in his rendering of the Satra sentence itself (see $ 12.1).
14.2 As for (1), it is, according to U., a pratika of the Satra sentence inter spersed with glosses. In U, it is a pure pratika, (a part of the glosses seemingly having been incoporated into the following explanation". Thus, Ue is of little if any help here. As for U.. (la) dmigs pa ram par rig pa sam gyis dbye? Phye ba can would fit only (S) (see $ 11 and 13. 1), but as the wording may have been taken over from the translation of the basie text this is not a strong argument. On the other hand, the splitting up of the pratika into the two sentences (1a)+(1b) and (Ic) and the repetition of dmigs pain (le) would seem to support (A), provided that (1c) is understood as
"I have taught that mind is the object", or, assuming a confusion of subject and predicate on the part of the translators:
"I have taught that the object is mind". But this would contradict the unambiguous testimony of U. (3). Moreover, I wonder if phyi rol gyi dmigs pa med pa (something like 'bahyalambana. virahita) - like 'arthafanya in Bh (2) (s. $13.3 1) —can really be used as an attribute or predicate of alambana, as it would have to if the text were based on (A). Therefore, I should prefer to attribute the splitting of the pratika into two separate sentences in U, to the translators, and to regard dmigs pa As a gloss of vijana, rendering not alambana but upalabdhi which is often used as a quasi-synonym of vijnana or vijflaptis. The Skt. of (1) may then have run like this:
(la) *alambanavij Aaptiprabhavitam (lb) bahyalambanavirahitarr
(le) vijanam spalabdhir iti maya defitam (or: ahar vadami) iti. This text would excellently fit (S). 14.3 (2) looks like another confirmation of (A), for it says, according to U.:
"For (that) this object (tad alambanan) is constituted by cognition only
o
18) U,: Tim tram li 271 b 1-3; U,: T vol. 31, 400 b 25-28. CA Nonaws. 200 f.; Katam,
82; Lamotte, MSc. trad. 95.
19) * in the beginning of Uç (2) seems to be a fragment of (lb). 50) See ns. 35 and 36; de la Vallée Poussin, Vijapimtitratasiddhi, 290.

Page #10

450
Lambert Schmithausen
Samdhinir macanastra VIII. 7
451
Ww
+
means that it has that for its nature (tatsvabhava)." This is interpreted by Uc to mean:
"......the external object of mind (+lad alambanam) is merely manifested by internal cognition; this means: the object has mind cognition as its nature." The Sanskrit original may be reconstructed as follows: (2a) tatha hi tadalambanan (U,: tad alambanair) vijnaptimdtraprabha
vitam (2b) tatsvabhavan ily arthah.
This sentence can be accepted to support (A) only if it is understood as a paraphrase of the pratika, at least of its first part, but preferably of the whole. In the latter case, ity arthah would, against Ue and probably also U. have to be taken to govern the whole sentence, and tat- (U: 1) in (2b) would have to be interpreted as vijana.. In this latter case, (2) would say that the Satra sentence "means that this object, being constituted by mere cognition, has the nature of vijnana", which would be an excellent parpahrase of (A). But both this interpretation and the interpretation of (2) as a paraphrase of only the first part of the pratika are to be discarded because (2) is not likely to be a para phrase of the Sotra wording at all; for such a paraphrase is given in (3). 'di ltar tatha hi in the beginning of (2a) rather suggests an explanation or a reason Moreover, tat- (+) in (26) may equally well represent vijnapiif, as is much more likely, the regimen of ity arthah is, with Ue and probably also U., confined to (2b), i. e. if (2b) is understood as an additional explanation of (vijnaptimatra prabhavita in (2a). The sentence would then mean:
"I. e. (or: For) that object (or, definitely better, with U.: its object, i. e. the object of vijnana") is constituted by mere cognition, i.e. has that
(viz. mere cognition) as its nature." Understood in this way, (2) would, similar to (2) in the Bhasya (see 13. 3. 1), explain (1) by showing that vijnana can be called 'alambanavi jaaptima. traprabhavita because (or: in the sense that) its object is vijnlaptimatraprabhavita (consists of or is manifested by cognition only), i.e. because the compound is to be understood in the sense of (S3) or (S4), (3), on the other
hand, sums up by paraphrasing (1) in other words while retaining its construc tion. In this way, Asvabhava's commentary on our sentence proves to be a consistent whole supporting (S).
15. Jnanagarbha on Samdh VIII. 754) (1) rigs pa bstan pa'i phyir/ rnam par ses pa ni.....rab tu phye bayin no
tes nas bfad doll) tes guns so/ (2) dmigs pa ni yul gyi rnam par sems snart ba yin la/ (3) de yani rnam par rig pa dan tha dad pa ma yin tel (4) cig car dmigs pa'i phyir roll (2) "The objective support (alambana) is the appearance of mind (citta) in
the form of an object (visayakarena), (3) and this (appearance of mind in the form of an object)) is not
different from the fact' of) cognition (vijnapti), (1) because they are (by necessity*) perceived simultaneously."
If in this text, which tries to interpret the sentence under discussion in the light of the epistemology of Dharmakirti and his followers (-sahopalambha niyama argument"), (2) and (3) are actually a paraphrase of the basic text
--but I am not sure they are a paraphrase moreover which, free though it is, will still have substantially preserved the syntactical structure of the latter, Jhanagarbha would seem to confirm [A] because dmigs pa Calambana) func tions as the grammatical subject, as in [A 1). Moreover, he would have split the sentence into two, again as in [A 1). Perhaps Jhanagarbha has understood the sentence under the discussion as follows: The object Calambana) is mind (appearing as an object] (vijnana), [this mind-appearing-as-an-object being] characterized by [being] nothing but not different from) [the act of cognition (vijnaptimatraprabhavita). In any case, Jhanagarbha can hardly be adduced in support of [S].
16. Byan chub rdzu 'phrul on Samdh VIII. 758) (1) 'o na sems dan grugs brian tes gdags su yan ji ltar run sham pa las Prnam par ses pa ni tes nas bsad dol tes bya ba gsuis le/
51) See n. 19. 52) Halantsang's in (2b) is in fact rendered by mam par rig p in the Tibetan translation
of Ytlan-ts'e's Sandh commentary (Tj thi 117 b 8). 53) Cp also the discussion of Bhasya (2a) in $ 13. 3. 1.
54) JAnag. 22, 3-7; Nozawa, 193. (55...55) Or: "and C. being such.) the [objective support) ......." 56) Cp. Jnanag. 25, 7 f. 57) Cp Pramanavirttika III. 388; Pramanaviniscaya 1. 55ab; T. Iwata, Sahopalambhaniyama
(diss. Hamburg 1980). 58) SardhVy co 193 b 1-3; Nozawa, 197.

Page #11

Lambert Schmithausen
Samdhinr mocanastra VIII. 7
453
(2) rnam par ses pa tid gzugs brian gyi dmigs pa lta bur snart bas sems
dai gengs brian tes gdags sugar runt la (3) sems kyi no bo las tha mi dad pa'i gzugs brilan de ni rail rig pa'i ishul
gyis rnam par rig pa ni ram par ses pa'i mishan nid yin no tes bstan
to ll (2) "As mind itself (vijnanam eva) appears as the object (alambana) [in the
form of a mental) image (pratibimba), it can be called both 'mind' (citta) and 'image'. (3) The essential characteristic (laksana) of mind (vijnana) is that this image which is not different from mind itself (citta-(svarapa) is cognized (vijnanti) by way of autoperception (svasamtvitli) - this is what has been taught [by the Buddha) Cor: what is shown [by the Sotra passage.] The decisive sentence is (3) which in spite of interpreting it in the light of later developments (svasamvitti) is clearly a paraphrase of the Satra sentence under discussion. For vijnana and vijnapti are found in the text, and the other words of the Satra sentence are easily parelleled with other elements of (3): alambana is of course represented by gangs brian (pratibinba; cp. also (2) grugs brilan gyi dmigs pa, probablypratibimbalambana in the sense of a rapaka compound); mishan nid (laksana) corresponds to prabhavitam; and matra seems to be explained by rast rig pali tshul gyis (svasamvittiyo gena). Thus, there can be hardly any doubt that (3) is a paraphrase of the Satra sentence under discussion. And there can also be hardly any doubt that the text on which this paraphrase is based can only be (S), not (A); for clearly rnam par ses pavijnana is the definiendum, i.e. must have been taken as the subject of the whole Sotra sentence, whereas gengs bran (pratibim ba) which corresponds to alambana can only be construed as the grammatical object of nam par rig pavijnapti. This is precisely the construction of [S2]. Byan chub rdzu 'phrul's commentary is thus an unambiguous support of [S].
17. The result of the preceding investigation is that unambiguous evidence for [A] is, except for the somewhat evasive testimony of Jhanagarbha (15), restricted to Paramartha's and Dharmagupta's Chinese translations of both 59) The commentary has, at least in the Tibetan translation, changed the comtruction of the
Setra sentence, via "A.prabhavitam F" (which would correspond to "A-lakpapan B") into
the equivalent pattern "A is the lake of B". 60) Cp : 59
Mahayanasamgraha (8 12 2 and 3) and Mahayanasamgrahabhasya (13. 1 and 2). But as I have tried to show in my analysis of the Bhasya passage ($ 13.3), the interpretation of these Chinese translations is not likely to represent the original meaning of the Bhasya passage but rather is the effect of a misunder standing of the function and wording of its statements. It is quite probable that the reading alambanay in the Satra sentencei. e. (A), is somehow connected with this (mis)interpretation of the Bhasya passage. I even suppose that (A) is the result of this (mis)interpretation, for somebody who had difficulties with the rather unwieldy compound 'alambanavijAaptimatraprabhavita in the Sotra sentence and mistook the explanation of the Bhasya for a pratika may casily have inserted, into the Satra text, what he considered a missing anusvara. In a similar way, the initial tad- of the Bhasya may have come to intrude into the Satra sentence, too. It is of course difficult to say when this happened first, but at any rate both phenomena are conspicuous in Paramartha's and Dharmagupta's translations.
In the case of Halan-tsang, too, it is likely that his rendering of the Sotra sentence though not his reading which seems to have been [S] (see $ 12 1) is influenced by the explanation of the passage in the Bhasya ( 13. 1 and 13. 3. 1) and the Upanibandhana (8 14.3).
18. This does not mean that the reading (A) could not have arisen under the influence of this interpretation of the Sotra quotation in the Mahayanasamgraha or, in view of the graphic insignificance of the change, even indepen dently- also in the context of the Samdhinirmocanastra itself; for in the preceding sentence of the Satra (see $2), the grammatical subject qualified as vijfapti matra is the image (pratibimba, i. e. the objective support of mind in meditative concentration). Thus, it might have seemed natural to make the objective support, and not vijnana, the suliffect of the following sentence, too.
19. While it is thus intelligible and, in the case of MSg II. 7, even palpable how the reading (A) could arise from (S), I do not see how if we disregard the possibility of a mere scribal error which would hardly have become so wide spread [S] could be explained as having arisen from (A). For, as wé men tioned above ($ 17. 1), the predicate in [S], viz. 'alambanavijnaptimatraprabhavita, is, from the point of view of analysis, definitely unwieldy, especially if one wants to interpret it, as the context requires it, in an idealist sense. It is hard to imagine that anybody, even if he felt some change necessary, would have changed the wording of (A) into such a terse formulation.

Page #12
Samdhinirmocanasutra VIII. 7 Lambert Schmithausen 454 455 matical (or logical) subject of vijnaptimatra(ta)61): it is, in his terminology, not the object or image but cognition itself that is qualified to be cognition only (because it is devoid of an external object : MSg II. 6). On the other hand, the reading (S) of the sentence under discussion ...... -vijfaptimatraprabhavitam vijnanam), especially if understood in the sense of [S2], precisely corresponds to this pattern. Had [A] been its original form, it is difficult to see why Asanga should have preferred this sentence to the preceding one (as also to the occurrences of vijfaptimatra in Samdh VIII. 8 and 9. where too vijnaptimatra qualifies the images, not vijnana). 23. Thus, both a critical examination of the source material and the cumulative evidence of intrinsic probabilities of context and style show that the original form of the sentence under discussion was [S] (*alambanavi jnapti. matraprabhavitam vijnanam......), not [A](*alambanam vijnaptio). Therefore, the interpretation proposed in $ 10 being applicable, the sentence under discussion, even if pointing to a former statement (88 7-9), would not conflict the assumption that Samdh VIII. 7-9 is not only the oldest extant source for the doctrine of vijnaptimatra proper but also its first literary expression. 20. The originality of [S] is not only supported by the fact that the unwieldy structure of its predicate does not make sense as a secondary change but becomes perfectly intelligible if one gets to the bottom of its double entendre (see $ 10) which can be appreciated as purposeful only in the context of the introduction of a new idea on which its discoverer wanted to confer as much of a traditional garb as was available. One could also point to the terminology of the sentence which diverges significantly from that of its context: alambana instead of gocara or pratibimba, and vijnana instead cittaa divergency which would not be motivated if the sentence had originally been, as it would have to if [A] were authentic, merely a more explicit and emphatic restatement of the preceding sentence. But the change in terminology, too, is natural in the case of [S] in view of the double meaning alluding to an Abhidharmic (alambana !) etymological definition (vijnapti- vijnana !). 21. Another point of view is the syntactical structure of the sentence as a whole. In this regard, [S], following the pattern de finiens-de finiendum, is perfectly unobjectionable and unequivocal, whereas [A] (de finiendum- definiens 1-de finiens 2) is formally ambiguous and, to my feeling, somewhat clumsy, to say the least. This unsatisfactory character of [A] from a purely formal point of view is easily explained as an unintentional by-product of a transformation of [S] into [A]. On the other hand, had [A] been the original reading, one might, to be sure, easily have felt bound to improve its formal structure, but this would almost certainly have led to some expansion of the somewhat abrupt second predicate (vi jnanam-ato vijnanam eva, or the like), but never to [S]. 22. Finally, attention should be paid to the fact that MSg II. 7 quotes only the sentence under discussion but not the preceding one according to which the images (pratibimba) [perceived in meditative concentration] are cognition only (vijfaptimatra) (see $ 2). This is strange, for one would expect that Asanga, in the context of scriptural proofs, would not have omitted this sentere without some reason, for after all it is, except for the sentence under discussion, the only one in Samdh VIII. 7 that contains the term vijnaptimatra. As there does not seem to be any clue for regarding the sentence omitted in MSg II. 7 as a later interpolation in the Samdhinirmocanasutra (there are some more omissions in MSg II. 7 which is thus obviously an abbreviated quotation), I suppose that Asanga has omitted the sentence because it did not fit his own terminology; for in MSg II he almost invariably uses vijfapti (II. 8: vijnana) as the gram - ne 61) Cp., e. g., MSg II. 2 (last 8); II. 6; II. 7. 2; II. 9; II. 11.