On the Problem of the External World in the Ch’eng wei shih lun

On the Problem of the External World in the Ch’eng wei shih lun

STUDIA PHILOLOGICA BUDDHICA

Lambert Schmithausen

Tokyo The International Institute for Buddhist Studies of The International College for Postgraduate Buddhist Studies

Published by the International Institute for Buddhist Studies of the ICPBS:

2-8-9 Kasuga, Bunkyo-ku, Tokyo 112-0003, Japan

Contents

Preface …………………………………………………… 7

I. A recent trend in interpreting the Ch'eng wei shih lun … 9

II. Discussion of the basic evidence ……………………. 13

III. The spiritual context of vijñaptimātratā ………………49

IV. Appendix:

Discussion of four Ch'eng wei shih lun passages … 57

V. Abbreviations .................................................................65

Preface

An earlier version of this paper was presented at a symposium on Yogācāra Buddhism in China, organized by Prof. Chen-kuo Lin in June 2000 at the International Institute for Asian Studies, Leiden. I take the opportunity to express my gratitude to the participants of the symposium, as well as to Dr. Anne MacDonald and Prof.s Florin Deleanu, Tilmann Vetter and Nobuyoshi Yamabe who were so kind as to read revised versions of this paper, for corrections and stimulating critical remarks which have led to a number of modifications and additions in the present version, which on account of my own (temporal as well as other) limitations is nevertheless still preliminary. To Japanese scholars working in the field, the result may not come as a surprise, but the motive for taking up the issue again will become clear from the following introductory remarks.

1 From a somewhat different angle (focussing on the problem of the plurality of sentient beings as purely mental continua and their mutual interaction) but with similar results, the issue has also been dealt with in an excellent paper by Nobuyoshi YAMABE (“Self and Other in the Yogācāra Tradition”, in: 北畠典生博士古稀記念論文集 (Kitabatake Tensei hakushi koki-kinen rombunshū), Kyoto 1998: 15–41). Japanese readers may miss references to secondary literature in Japanese. It is, however, not only because of the constraints of time but also on account of the special purpose of this paper (as delineated in ch. I) that I shall base my argument on primary sources only.

I. A Recent Trend in Interpreting the Ch'eng wei shih lun

Yogācāra thought has traditionally been understood as advocating the epistemological position that mind, or consciousness,2 does not — at least not directly — perceive or cognize anything outside itself, but rather cognizes only its own image3 of an object, and as propounding the ontological position that there are no entities, especially no material entities, apart from consciousness, or, more precisely, apart from the various kinds of mind (citta) and mental factors or mind-associates (caitta) (see § II.2). This understanding was not invented by modern scholars but is in line with the works of medieval Indian (and Tibetan) authors, both non-Buddhist and [p.10] Buddhist In recent times, some 2 In the present paper, I mechanically use “mind” for 心 (citta) and, after considerable hesitation (in view of the subliminal character of the ālayavijñāna), “consciousness” for 識 (vijñāna, vijñapti), although at least citta and vijñāna are substantially interchangeable. The term vijñapti is, more specifically, used for the function of vijñāna, i.e. in the sense of “making known”, “cognizing”, and is then rendered by 了 or 了別); it may, however, also qualify the object of consciousness as being nothing but an image in consciousness. For jñāna (智) I have chosen “knowledge” in the case of the Buddha, but “insight” in the context of the Path. I beg the reader's pardon if my choice is not the most felicitous one, but after all English is not my mother tongue.

3 The use of the word “image” to render Ch. 相 (when it is equivalent to 相分, i.e. the object part or aspect of a consciousness, probably corresponding to Skt. nimitta) is for want of something better. It is not intended to imply, necessarily, the existence of an original of which the “image” is a reflection (as would often, though perhaps not always, seem to hold good when the more specific term 影像 is used).

4 Cf., e.g., Śaṅkara's commentary on Brahmasūtra 2.2.18: vijñānâstitva- mātra-vādinaḥ and 2.2.28: vijñānaîka-skandha-vādaḥ (the latter term being, of course, hardly justified); or Yuktidīpikā (ed. A. WEZLER and Sh. MOTEGI, Stuttgart 1998) p. 219,5-6: asattvaṁ bhāvānām (in a

10 On the Problem of the External World in the CWSL scholars, mainly from the Anglo-Saxon cultural sphere, have challenged the traditional understanding, and especially its ontological aspect. There is no doubt that in some early Yogācāra works (especially the Yogācārabhūmi, but also others like the Abhidharmasamuccaya) 6 the above-mentioned views are not found at all or at best only sporadically. However, the issue of the critics is not just this but the interpretation of the very principle of ‘nothing but consciousness’ (vijñaptimātra(tā)), which is usually taken to express the epistemological and ontological position the Yogācāras are credited with. Even full-fledged vipassage alluding to Vasubandhu's Viṁśatikā). Cf. also Udayana, Ātmatattvaviveka (Bibl. Indica) p. 429: vijñānavādini jāgaruke bāhyam eva nāsti kuta ātmā (quoted in Ikkō ARAI, "Critiques of the Vijñānavāda by Udayana: A Study of the Chapter on bāhyārthabhaṅga of the Ātmatattvaviveka (1), in: Komazawa Daigaku Bukkyōgaku-bu Ronshū 34

5 Cf., e.g., Madhyamakāvatāra (ed. DE LA VALLÉE POUSSIN, St. Petersburg 1907–12) 182,2-18 and 185,6-20, where Candrakīrti refutes the Vijñānavādin’s interpretation of the famous Daśabhūmikasūtra passage cittamātram idaṁ yad idaṁ traidhātukam. According to Candrakīrti, the purport of this passage is merely to negate that there is a permanent Self as an agent and that matter (rūpa), etc., are of primary importance [for the formation of the traidhātuka]; its purport is not (as the Vijñānavādin asserts: cf. 181,8-12) to negate the existence of matter (rūpa) or external [[[objects]]] (phyi rol: 185,8), in the sense that only mind (cittamātra) exists whereas matter does not, at any rate not apart from mind and mind-associates (sems tsam zhig kho na yod kyi gzugs ni med do: 185,19). Cf. also Blo gsal grub mtha’, ed. K. MIMAKI, 104– 105: ... gzugs ... thams cad kyang sems dang sems las byung ba las gud na med do.

6 Cf. L. SCHMITHAUSEN, “Zur Literaturgeschichte der älteren Yogācāra-Schule”, in: Zeitschrift der Deutschen Morgenländischen Gesellschaft Suppl. I.3 (1969): 820–821; id., “Spirituelle Praxis und philosophische Theorie im Buddhismus”, in: Zeitschrift für Missionswissenschaft und Religionswissenschaft 57.3 (1973): 161–186, esp. 165–167; id., “On the Problem of the Relation of Spiritual Practice and Philosophical Theory in Buddhism”, in: German Scholars on India, ed. by the Cultural Department, Embassy of the Federal Republic of Germany, vol. II, Bombay 1976: 235–250, esp. 238–240; YAMABE 1998 (s. fn. 1): 17.

A Recent Trend in Interpreting the CWSL 11 jñaptimātra texts like Vasubandhu's Viṁśatikā and Triṁśikā, and the commentary on the latter compiled, on the basis of Indian materials, by Hsüan-tsang (玄奘, 602?–664), viz. the Ch'eng wei shih lun (成[p.11]唯識論, henceforward CWSL), are stated not to deny an independent existence of material things (rūpa), or of the so-called external (i.e. the physical) world.

A recent contribution in this vein is Dan LUSTHAUS's monograph Buddhist Phenomenology. 7 According to LUSTHAUS (L 536), “to the extent that epistemological idealists can also be critical realists, Yogācāra may be deemed a type of epistemological idealism, with the proviso that the purpose of its arguments was not to engender an improved ontological theory or commitment.” I agree with his view that the teaching of vijñaptimātra is basically not a theoretical aim in itself but a “therapeutical device”, a soteric strategy, directed against attachment and appropriation.8 Still, Hsüan-tsang was also a great scholar of Abhidharma, and in a sense the CWSL can also be understood as an attempt to re-formulate conservative Abhidharma in terms of vijñaptimātra. This may well imply a – preliminary – ontological commitment. LUSTHAUS seems to admit that Yogācāras reject

the externality of objects (e.g. L 484), but he insists on some kind of ‘real’ existence of matter (rūpa, 色), which is independent from mind in the same way that other persons' or sentient beings' minds exist independently from one's own mind (L 492 and 512). Not being a philosopher, I am not going to discuss LUSTHAUS's philosophical interpretation of the CWSL, let alone of the whole Yogācāra tradition. I shall rather try to re-examine the main passage on which LUSTHAUS grounds his thesis of the independent existence of matter from a philological point of 7 Dan LUSTHAUS, Buddhist Phenomenology. A Philosophical Investigation of Yogācāra Buddhism and the Ch'eng Wei-shih lun. London: RoutledgeCurzon 2002 (henceforward: L). The original version of the present paper had referred to LUSTHAUS's PhD dissertation which is stated by him to be the “distant ancestor” of this book (see L p. xi). 8 E.g. L 537. Cf. also YAMABE 1998 (s. fn. 1): 35–37. 12 On the Problem of the External World in the CWSL view and on the basis of what I would like to call an internal, or ‘emic’, interpretation of pertinent textual materials. [p.13]

II. Discussion of the Basic Evidence

[1.1] If I understand LUSTHAUS correctly, an ‘independent existence’ of matter would involve that matter is not entirely reducible to images in some form of mind or other. Fortunately, LUSTHAUS produces at least one passage9 from the CWSL which he considers capable of proving that matter (rūpa) “exists independently, though not separate from my mind” (L 491), in the

same way that the mind continuum of another person does. In the context of the question of knowledge cognizing another person's mind (paracittajñāna)10,11 Hsüan-tsang points out that such a cognition is, to be sure, not possible in a direct way but occurs only through the mediation of an image into which one's own mind itself has transformed or developed (自所變)12. After quot- 9 Two more are adduced at L 512. For a discussion of these passages, see appendix, §§ 1–2.

10 This is one of the supra-normal knowledges or powers (abhijñā), accessible only to Buddhas and advanced yogis or Bodhisattvas. 11 CWSL 39c9-16 / P 430; C 239; S 320.

12 The Sanskrit term at the basis of 變 is pariṇāma, which means “change, alteration, transformation; development; ripeness, maturity” (MONIER WILLIAMS). In Vasubandhu's works (cf. my article “Sautrāntika- Voraussetzungen in Viṁśatikā und Triṁśikā”, Wiener Zeitschrift für die Kunde Süd- und Ostasiens 11/1967: 109–136) it is used as an action noun describing a process taking place in the continuum (saṁtāna, saṁtati) of a person or in the consciousness continuum or its latent stratum. It may also refer to the culmination of this process or to its result (the actual kinds of vijñāna). In the CWSL, however, it refers to a de-temporalized ‘transformation’ or ‘development’ within a single moment of a vijñāna or mental factor, i.e. to the fact that each moment arises in such a way that it has ‘changed’ or ‘developed’, from the outset, into an image (相) of an object cognized (or into a duality of image 相 and vision 見; cf. SNSṬ vol. Thi 122b8: lta ba dang rgyu mtshan gyi rnam par yongs su gyur pa). This image (or duality of image and vision) is called 識所變 (“what vijñāna has changed, or de14 On the Problem of the External World in the CWSL

ing the Saṁdhinirmocanasūtra13 in support, Hsüan-tsang continues: [p.14]如緣他心, 色等亦爾｡ (CWSL 39c16) LUSTHAUS (L 491) translates the sentence as follows: “Other mind is this sort of condition; rūpa, etc. are the same case.”

For grammatical reasons, this translation must be discarded;14 in Hsüan-tsang's diction, 如 can hardly mean “this sort of”, and is veloped, into”, which is equivalent to a passive expression “developed by vijñāna”) or, by way of an ellipsis of 識, simply 所變 (also 所變現, e.g. 46c8). In syntactically unambiguous situations, this may even be reduced to mere 變. This holds good not only for ordinary consciousnesses but also for those of a Buddha as far as they are directed towards the conventional (cf. CWSL 57c3-4; 58c1-3 and 27-29; opposite view refuted at BBhU 317b19-29). The manifestation of an image in consciousness is described by expressions like “consciousness appears, or arises, developed/changed into an image looking like (or: into what looks like) X (e.g. visible matter)” (CWSL 4a27-28: 識 ... 變 似眼等色等相現; 5a6: 識變似色生), “consciousness develops/changes

into an image of (or: into what looks like) X” (10c16: 識變為此相; 4c22: 識變似聲), “consciousness develops/changes [in such a way as to] manifest X” (2a8: 識變現諸蘊), or simply “consciousness develops/ changes into X” (10a17-18: 識 ...變為 ...有根身). In another text translated by Hsüan-tsang (T vol. 31 no. 1598: 401a29), we even find this idea expressed by the formulation 阿賴耶識變作諸色, which translated literally would mean “ālayavijñāna, developing/changing, makes/creates material things”.

13 VIII.7 (p. 91,8-11) in E. LAMOTTE’s ed.; Skt. in Jñānaśrīmitranibandhāvalī (ed. A. THAKUR, Patna 1959) 478,3-4: na hi Maitreya tatra kaścid <dharmo ka ṁcid> dharmaṁ pratyavekṣate, api tu tath ā samutpannaṁ tac cittaṁ yat tathā khyāti.

14 There are many more cases where I fundamentally disagree with LUSTHAUS's translations, often for grammatical reasons, but this is not the right place to go into details (cf., however, fns. 26, 31, 74, 76 and 78, and appendix § 1). Generally, L. DE LA VALLÉE POUSSIN's translation is much more reliable, and LUSTHAUS's occasional mockery is entirely inappropriate. L 470, n. 18, DE LA VALLÉE POUSSIN's translation of CWSL 39a28 (by the way: of eight characters, not four), though containing explanatory additions, is perfectly correct, while

no [p.15]doubt a conjunction corresponding with 亦爾,15 and 緣 他心 has to be taken as verb + object, as it has been in the translations by DE LA VALLÉE POUSSIN and COOK: “Il en va de la pensée qui connaît le Rūpa, etc., comme de la pensée qui porte (ālambaka) sur la pensée d'autrui” (P 430).

“As with having the minds of others as objects, so with form, etc.” (C 239; cf. S 320). This syntactical analysis is also confirmed by a Tibetan rendering of the passage in the Tibetan translation of a commentary on the Saṁdhinirmocanasūtra by Hsüan-tsang's Korean student Wŏn-ch'ŭk (圓測, Ch. Yüan-ts'ê, Tib. Wen tshig [or tshegs]: 613–696)16:

“As that which has another's mind as its object, so also [that which has] rūpa, etc., [as its object]” (SNSṬ vol. Thi 119b3: gzhan gyi sems la dmigs pa ji lta ba bzhin du gzugs la sogs pa yang de bzhin no).

LUSTHAUS's is wrong; 二隨一故 clearly means “because it is one of the two” (隨一 = anyatara), just as at 39a25 五隨一故 means “because it is one of the five”. The reader of LUSTHAUS's book will also often be surprised at his quotations of, or remarks on, Sanskrit expressions. Cf., e.g., L 497, where the correct cvi-formation saṁmukhī-bhāva (which he seems to connect with saṁmukhin) is deliberately replaced by the non-existing word saṁmukhā-bhāva. Incidentally, Ch. 現行, as the antonym of ‘seeds’ (bīja), corresponds to samudācāra (“full, actual emergence”) or saṁmukhībhāva (“becoming face to face”, “becoming actually present”), not adhyācāra (which refers to committing an offence), as can be gleaned from Abhidharmasamuccaya (ed. P. PRADHAN, Santiniketan 1950) 35,3 and 35,15-16.

15 Cf. CWSL 12b1-2; c5-6; 7-8; 9-10; 21a17; 39c5-6; 50a28-29; 58c4-5. 16 Cf. John POWERS, “Lost in China, Found in Tibet: How Wonch'uk Became the Author of the Great Chinese Commentary”, in: Journal of the International Association of Buddhist Studies 15.1 (1992): 95–103.

16 On the Problem of the External World in the CWSL [p.16]Taking into account the context of the passage, i.e. the sentence which immediately precedes the Saṁdhinirmocana quotation,17 a more explicit rendering would run as follows: “Just as [in the case of consciousness] having another's mind as [its] objective support (緣 X = X-ālambana) [what is directly cognized is not the other person’s mind itself but only an image of it developed by the cognizing mind itself], so also [in the case of] visible matter (色 = rūpa), etc.18 (i.e. in the case of a consciousness having visible matter, etc., as its objective support) [what is cognized directly is only an image developed by the cognizing mind itself].”19

[1.2] But is this small passage really strong enough and sufficiently unambiguous to carry the burden of a radical reinterpretation of the system of the CWSL? Does it really presuppose an existence of matter that is independent of the cognizing mind? And if it does, is DE LA VALLÉE POUSSIN's interpretation20, ac- 17 “[This knowlege] is called ‘knowledge of another [[[person]]]’s mind’ (paracittajñāna) only because the latter, like [an image in] a mirror, etc., appears as an external object (viz. as the mind of another person), but [this knowledge] is not capable of cognizing [another's mind] directly. What is cognized directly, is [only the image of the other person's mind] developed by [the paracittajñāna] itself.”(CWSL 39c13–14: 但如鏡等 似外境現 名了他心｡ 非親能了｡ 親所了者 謂自所 變｡ / P 430; C 239; S 320; cf. L 491)

18 I suppose “etc.” refers to the other kinds of material senseobjects (viz. sound, etc.), as at CWSL 39a26, b27 or c2 (cf. fn. 81), and not to the viprayukta-saṁskāras and the asaṁskṛtas, as at CWSL 4a7, 7a19 or, perhaps but not necessarily, 39c25. 19 In YAMABE 1998 (s. fn. 1): 31, whose rendering ( “Cognizing other people’s minds or matter is also [effected] in the same way.”) slighly differs from mine, the sentence has, probably by misprint, been included in the Saṁdhinirmocana quotation.

20 P 430: “Le Rūpa qui est le nimittabhāga de la pensée d'autrui (le corps d'autrui, développement du Vijñāna d'autrui), et aussi le Rūpa qui est le développement d'un autre Vijñāna de la même personne. [C'est-à-dire: le cakṣurvijñāna (darśanabhāga) a pour ālambana immé

cording to [p.17]which this independent matter is to be understood as the image of matter in vijñānas other than the cognizing one, indeed nothing but the imposition of his own idealist presupposition, as LUSTHAUS (L 492–493) asserts? I, for my part,

should rather prefer to understand the passage in the light of sufficiently explicit and unambiguous statements of the position of the CWSL in the CWSL itself. If the picture emerging from the CWSL itself is explicitly confirmed or organically supplemented by the earliest Chinese commentators or by Indian and Tibetan sources, I am inclined to regard this as corroborative evidence. [1.3] I am, of course, aware of LUSTHAUS's (L 382ff) distrust in the authenticity of the explanations of K'uei-chi (窺基: 632– 682)21, Hsüan-tsang's student and author of the only available direct commentary on the CWSL. It would certainly be desirable to systematically search for additional information on the issue under consideration in the Saṁdhinirmocana commentary of K'uei-chi's opponent Wŏn-ch'ŭk and in the Yogācārabhūmi

commentary by Tun-lun 遁倫 (or Tao-lun 道倫)22, a Korean collaborator of K'uei-chi [p.18]who, however, quite often quotes the interpretations of other exegetes as well, including Wŏnch'ŭk, but this is beyond the limits of my time. At any rate, LUSTHAUS's scepticism regarding K'uei-chi's ascriptions of diat son propre nimitta, qui est une reproduction du Rūpa développé de l'Ālayavijñāna].”

21 I use the “traditional name” (cf. HôbFA p. 264, also for other names). As Prof. N. YAMABE kindly pointed out to me, the problems about this name are discussed in FUKAURA Seibun 深浦正文,『唯識 学研究, 上巻 : 教史論』, 東京 : 永田文昌堂 [1954]1972: 256, n. 2, and in Stanley WEINSTEIN, “A Biographical Study of Tz’u-en”, in: Monumenta Nipponica 15.1–2 (1959): 119–149 (esp. 129 ff). 22 HôbFA p. 284 (s.v. Tonrin). The problem of his name is discussed in YŪKI Reimon 結城令聞,『唯識学典籍史』, Tokyo 1962: 264 ff, and, as Prof. N. YAMABE kindly informed me, also in YŪKI Reimon,「『瑜伽論記』の著者名に対する疑義」, repr. in:『結城令 聞著作撰集, 第一巻: 唯識思想』東京, 春秋社 1999: 145–155, and in: YANG Pai-i 楊白衣,「新羅の学僧道(遁)倫の『瑜伽師地論記』の研 究」, in: 東洋学術研究 23.1 (1984): 292–305.

18 On the Problem of the External World in the CWSL CWSL materials to Indian authors, especially Dharmapāla, needs reconsideration in view of the fact that these ascriptions are, in significant cases, also confirmed by Wŏnch'ŭk. Even a very sporadic use of his Saṁdhinirmocana commentary23 yielded evidence to the effect that he too ascribes CWSL materials to

Dharmapāla, e.g., the explanation of the final verse of the fifth chapter of the Saṁdhinirmocanasūtra at CWSL 15c7-14.24 Still more important is Wŏnch'ŭk's quotation, from the CWSL (45c22- 26), of the two opinions on the question of whether all the eight kinds of consciousness, or only some of them, are false imagination (能遍計), because he explicitly ascribes these opinions to Sthiramati and Dharmapāla,25 respectively, just as K'uei-chi does. There is thus reason to assume that at least some of the ascriptions are not just K'uei-chi's forgery but go back to Hsüantsang himself (though admittedly this does not solve all the problems).

[2.] Let me start my re-examination of the passage quoted above with a closer look at its context. It concludes a paragraph which is introduced by the following objection to the Yogācāra point of view:

[p.19]“[Since according to your system] matter outside [the mind] is non-existent in reality, it may be admissible [for you to assert] that it is not an object of consciousness within. [But even in your system] another's mind exists 23 Unfortunately, I have no access to the Chinese original (as far as it is preserved) but have to rely on the Tibetan translation by Chos grub.

24 SNSṬ Ti 286a5(ff): bstan bcos rNam par rig pa tsam du grub pa'i nang nas slob dpon Chos skyong gi bshad pas ni … 25 SNSṬ Ti 290a6–b1: kha cig na re rnam par shes pa brgyad dang sems las byung ba zag pa dang bcas pas bsdus pa ni … zhes zer te / 'di ni Blo brtan gyi bshad pa'i don to // yang kha cig ni rnam par shes pa drug <pa?> dang bdun pa'i sems kyi rnam pa (text: pas) bdag dang chos su 'dzin pa ni ... zhes zer te / 'di ni slob dpon Chos skyong gi bshad pa'i gzhung ngo // In this case, the position ascribed to Sthiramati is in agreement with his Triṁśikābhāṣya (TrBh 35,14-17).

in reality. Why is it not an objective support of one’s own [[[consciousness]]]?” (CWSL 39c9-10: 外色實無 可非內 識境｡ 他心實有 寧非自所緣｡ / P 430; C 239; S 320.26) The opponent clearly presupposes that the Yogācāra system negates the existence of matter outside mind (ontologically) but admits the existence of another's mind, i.e. his ontological interpretation of vijñaptimātratā is that it excludes external matter

but not a plurality of mind continua. At the same time the opponent presupposes an epistemological interpretation of vijñaptimātratā in the sense of each consciousness being strictly confined to itself, i.e. unable to cognize anything outside itself. He also seems to correlate the epistemological postulate to the ontological one by taking the former to be based on the latter, i.e. by assuming that the Yogācāra rejects external entities as objects of consciousness because of their non-existence. But this would not hold good in the case of other mind continua which exist and hence ought to be cognized (at least by yogis, as the Buddhist tradition generally assumes). This, however, would break the epistemological principle of vijñaptimātratā27 and hence render it doubtful also in the case of matter.

The proponent rejoins by specifying the purport of vijñaptimātratā: As an epistemological principle it means, to be sure, that nothing outside the respective moment of consciousness itself can be its direct objective support (親所緣). But it does not exclude [p.20]entities outside the respective consciousness from being its object at all, i.e. from being cognized in some indirect way. In this context, the disputed sentence makes clear that this holds good not only for another's mind but also for matter. But there is no rejection of the opponent's presupposition that [for the Yogācāra] external matter, i.e. matter existing outside any form of mind, does not exist in reality. On the 26 L 490 gives the passage a strange twist which forces him to take 實無in a concessive sense, which in view of the absence of a concessive conjunction is improbable.

27 Cf. Madhyamakāvatāra (see fn. 5) 166,14-16.

20 On the Problem of the External World in the CWSL contrary, the following paragraph which resumes the problem of how the principle of vijñaptimātratā is compatible with ontological pluralism is, it is true, quite explicit in interpreting vijñaptimātratā to include a plurality of multi-layered mind continua along with their mind-associates (caitta) and the intramental dyad of image (*nimitta) and vision (*darśana) into which mind and mind-associates ‘transform’ or ‘develop’ (所變 相見), etc.:

“The word ‘consciousness’ (識: vijñapti) summarily indicates that in each of all the sentient beings there are (1.) eight [[[forms]] of] consciousness (識: vijñāna), (2.) six categories of mind-associates, (3.) image- and vision- [part] into which they develop, (4.) [their] different states (分位差別: avasthā-viśeṣa, some of which are wrongly hypostatized by the Sarvāstivādins as cittaviprayuktasaṁskāras), and (5.) true Suchness (真如: tathatā)28 mani-

28 As for the tathatā, I disagree with LUSTHAUS's (L 530f; cf. also 359 and 535) statement that it is merely a prajñapti. This statement is based on CWSL 6c10-20 where the unconditioned (asaṁskṛta) entities like space (ākāśa) are interpreted as denominations (prajñapti) of the transconceptual, ineffable true ultimate nature (dharmatā) of everything, viewed under certain aspects. In this context, it is then added that even [the term] ‘tathatā’ and the qualifications of tathatā as existent, empty, etc., are mere designations [of this dharmatā], used with the purport of removing wrong conceptualizations of the transconceptual ultimate nature. tathatā, the text adds, must not be conceived of as an unconditioned entity apart from the dharmas (matter, mind, etc.), as other Buddhist schools do. But at the same time this implies that tathatā as the transconceptual, ineffable true ultimate nature (dharmatā) of everything, which becomes manifest in transconceptual insight (nirvikalpaṁ jñānam : cf. CWSL 49c20(ff)), is not rejected.

Tathatā constitutes the primordial nirvāṇa of natural purity (55b7-8) and is the basis of all forms of actualized nirvāṇa (55b12-17). In its non-conceptualized, true nature, it can by no means be a mere prajñapti. In fact, elsewhere the CWSL explicitly states that the Perfect Nature (圓成實性 = pariniṣpannaḥ svabhāvaḥ), which is expressly equated with tathatā at CWSL 46b15-16, is exclusively truly existent

fested (所顯: pra[p.21]bhāvita) by the principle of their voidness; for these [five items] are [respectively] (1.) what has consciousness (or: [the function of] cognizing) (識: vijñapti) as its specific characteristic (自相: svalakṣaṇa), 29 (2.) what is associated (相應:30 saṁprayukta) with consciousness (識: vijñāna), (3.) what the two [preceding items] ‘develop’ into, (4.) specific states of the three [preceding items], and (5.) the true nature of the four [preceding items]. These (lit. such) dharmas, all of them not being separate from consciousness, are summarily designated as ‘consciousness’.” (CWSL 39c20-24: 識 言總顯 一切有情 各有八識．六位心所．所變相見．分 位差別 及彼空理所顯真如｡ 識自相故, 識相應故, 二所變 故, 三分位故, 四實性故｡ 如是諸法, 皆不離識, 總立識名｡ / P 431; C 240; S 320–321; cf. L 487.)

But the text is equally unambiguous in excluding matter as something really existing entirely apart from any form of mind as it is conceived of by ill- or untrained people: [p.22]“The word ‘-mātra’ merely excludes31 visible matbecause it is not constituted as a [mere] designation (prajñapti) on the basis of something else (CWSL 47c12-13: 圓成實性 唯是實有, 不依他 緣 而施設故｡).

29 Cf. Vasubandhu, Pañcaskandhaka (Peking Tanjur vol. Si) 16b8: rnam par shes pa gang zhe na / dmigs pa rnam par rig pa'o //, which probably renders Skt. *vijñānaṁ katamat / ālambanavijñaptiḥ / Hsüantsangs Chinese rendering of the passage (T vol. 31 no. 1612: 849c 27) is as follows: 云何識蘊｡ 謂 於所緣境 了別為性｡ This use of 性 is very close to that of 自相 in the passage translated above. 30 Chinese lit. “corresponding”, “in agreement”, which is in fact an aspect included in the term (cf., e.g., AKBh 62,6-10). 31 L 488 (“The word wei (mātra, only, nothing but) is only [used to reveal what is] concealed from the fools who are attached to …”) misunderstands 遮 which may mean “to cover” but is, in exegetical terminology, current in the sense of “to ward off”, “to exclude” (cf. A. HIRAKAWA, Buddhist Chinese-Sanskrit Dictionary, Tokyo: The Reiyukai 1997: 1166: pratiṣedha, nivārayati, prati-√kṣip, vyāvṛtti, vyudāsa, etc.). Moreover, 所 is normally a relative pronoun in the locative or 22 On the Problem of the External World in the CWSL ter, etc.,32 as they are conceived of by ill- or untrained people, i.e. as something really existing definitely apart from any form of consciousness.” (CWSL 39c24-25: 唯言 但遮 愚夫所執 定離諸識 實有色等｡ / P 431; C 240; S 321.)

There does not seem to be any room left for matter existing independently in the sense of being neither an image in some form of mind nor entirely separate from mind, as postulated by LUSTHAUS (see ch. I).

[3.] The CWSL contains quite a few statements concerning matter or entities outside the mind or consciousness. Many of them are negative, but it would seem to be advisable to distinguish between different categories, even though a clear-cut distinction may not always be possible.

[3.1] One class of statements negates entities or matter outside or separate from mental factors (i.e. mind and mind-associates) as the object (境, i.e. viṣaya, lit. domain) or objective support (所緣, i.e. ālambana) of the latter. To give just a few examples: “Therefore one must know that in reality there is no external object, but only the internal consciousness which arises [p.23][in such a way that it] resembles an external object.” (CWSL 7a22-23: 由此應知｡ 實無外境, 唯有內識 似外境生｡ / P 84; C 40–41; S 59. Cf. also CWSL 1b14-15 and 1b2-3.)

“An external object, because of being established arbitrarily33, does not exist [in the same way] as consciousaccusative, not in the nominative. Hence, 愚夫所執X can hardly mean “the fools who are attached to X” but only “the X to which the fools are attached”, which is equivalent to a passive expression, viz. “the X clung to by the fools”. 32 Cf. fn. 18. In the present passage, the alternative “matter, etc.” (with “etc.” referring to viprayukta-saṁskāras and asaṁskṛtas) would also make good sense.

33隨情 is used in the sense of 隨妄情: cf. CWSL 1b8 and Shu-chi 243b20+22; CWSL 3c13 and Shu-chi 265a29. In Hsüan-tsang's transla - Discussion of the Basic Evidence 23 ness [[[exists]]].” (CWSL 1b10-11: 外境隨情而施設故 非有 如識｡ / P 10; C 10; S 13. Cf. also CWSL 1a12-13.) “In order to dispel the wrong conception that an object exists in reality outside mind and mind-associates, it is taught that there is only consciousness (vijñaptimātra).” (CWSL 6c24-25: 為遣妄執 心心所外 實有境故, 說唯有識 ｡ / P 80; C 39; S 57; cf. L 531.) “[When] they have thoroughly understood that there is no objective support separate from consciousness, then they are taught that the image part [of the respective consciousness itself] is the objective support.” (CWSL 10b5-6: 達無離識所緣境者, 則說相分是所緣｡ / P 128; C 62; S 79.)

These sorts of statements may well intend a wholesale denial of the existence of entities existing outside any form of mind, or independently, but if we want to be cautious we should suspend judgement and rather interpret these statements epistemologically, i.e. as rejecting extra-mental entities merely as objects of consciousness (which would leave them the possibility of existing as non-objects, i.e. without being cognized, or at least without being cognized in a way which justifies their being termed ‘objects’ of consciousness).

[3.2] However, according to CWSL 7a17-19 (P 82; C 40; S 59) the reason why mind and mind-associates definitely do not have [p.24]external matter, etc., as their objective support (ālambana) is that such extra-mental entities do not really exist: “Thus, Self and dharmas apart from consciousness, as they are conceived of by [respectively] the non-Buddhists and the [followers of the] other Vehicles, are all non-existent in reality. Therefore, mind and mindassociates certainly do not use external matter, etc., as tion of the Yogācārabhūmi (T vol. 31 no.1579: 639a8-9: 隨情造作身語 意業三種惡行) the expression 隨情 corresponds to Tibetan ci 'dod dgur “as he likes” (Peking Tanjur vol. Zi: 160b2). 24 On the Problem of the External World in the CWSL their objective support (ālambana).” (如是外道餘乘所執 離識我法 皆非實有｡ 故心心所 決定不用 外色等法 為所 緣緣｡)

And there are other passages as well which flatly assert, in obviously ontological terms, the non-existence of external matter, 34 as, e.g., CWSL 5a5-6 (P 53; C 28; S 35): “Thus one should know that in reality there is no external matter, and only internal consciousness arising [in such a way that it] develops (or, more literally: transforms itself into) [an image] resembling matter.” (由此應 知｡ 實無外色, 唯有內識 變似色生｡)

Though what precedes is a refutation of the Abhidharmic conception of matter, there is no indication that any other way of main[p.25]taining the existence of extra-mental or independently existing matter would be exempted from criticism. Actually, the text argues that real matter is logically impossible because as a divisible whole it would not be real and as indivisible atoms it would be without any shape or physical substance (無質礙 =

• amūrta) and hence, in the last analysis, immaterial (CWSL 4a11-

34 Cf. also TrBh 15, 25–16,1: “Dharmas and a Self do not exist outside a transformation of consciousness” ( dharmāṇām ātmanaś ca vijñāna-pariṇāmād bahir abhāvāt ); 17, 2: bāhyārthābhāvāt; 16, 22-23: vinaîva bāhyenârthena (cf. also 16, 6-7 and 18). There seems to be a tendency to read more into the term artha than what may have been

intended by the Indian writers. According to Sanskrit lexicographers (e.g. Amarakośa 3.3.86; Śrīdharasena, Abhidhānaviśvalocana, ed. L. JAMSPAL, Naritasan 1992: 942–943), it may not only mean, among other things, “purpose” (prayojana), “meaning” (abhidheya), or “object” (viṣaya, not in Amara), but may also simply mean “thing” (vastu). This suggests that it was felt to cover both the epistemological and the ontological aspect. Cf. also Sthiramati, comm. on Mahāyānasūtrālaṅ- kāra XI.47: gzugs la sogs pa phyi’i dngos po ni med kyi ... (Tanjur, Peking vol. Mi 213a1; Derge vol. Mi 192a7-8; O. HAYASHIMA, “Dharmaparyeṣṭi”, in: Bulletin of Faculty of Education, Nagasaki University 27/1978: 116).

13 / P 39–40; C 24; S 30).35 Summing up its arguments, the text concludes: “Hence one must know that all obstructive (sapratigha, i.e. ordinary) matter is [merely] a manifestation [of an image] developed by consciousness, and does not consist of atoms (i.e. is not really material).” (CWSL 4c4-5: 由此 應知: 諸有對色 皆識變現 非極微成｡ / P 47; C 27; S 33.)

And: “Since even obstructive matter, which [at least] looks material, turns out, on investigation by logical arguments, not to exist apart from consciousness, how much less can unobstructive (apratigha) matter, which does not even look like matter, be called a real material entity.” (CWSL 4c6-8: 諸有對色 現有色相 以理推究 離識尚無｡ 況無對 色 現無色相 而可說為 真實色法｡/ P 47; C 27; S 33.)36 [4.1] However, there are also some passages which affirm the existence of “external dharmas”, albeit in a specific sense. CWSL [p.26]7a12-13 (P 81; C 40; S 59), e.g., may, at least at first glance, be taken to state that in the case of the various forms of “clinging to (or: hypostatizing) entities” (法執: dharma-grāha) described in the immediately preceding portion of the text,37 the actually existing factors distorted by the hypostatizing mental 35 This is, of course, the same argumentation as in Vasubandhu's Viṁśatikā (ed. S. LÉVI) p. 6,22 ff.

36 Similarly, at CWSL 39c24-25 what is stated to be excluded by the element mātra in vijñaptimātra is (visible) matter, etc., insofar as it is regarded as really existing apart from mind (唯言 但遮 愚夫所執 定離 諸識 實有色等). According to other passages it is real entities (實物) apart from mind (but not entities not separate from mind like the mind-associates) (CWSL 38c24-25: 唯言 為遮 離識實物, 非不離識心所 法等｡), or just the external (外, or extra-mental, cf. 59a16: 除識性), which is entirely non-existent (都無), in contrast to the “internal object” (內境) (59a8-9: 唯言遣外, 不遮內境; 59a14: 非謂 內境 如外都無). 37 I.e. CWSL 6c26–7a12.

26 On the Problem of the External World in the CWSL activity are not only dharmas inside one’s own mind (自心內法) but, in certain cases, also dharmas outside one’s own mind (自 心外法; my translation of 自心 is preliminary and will be revised in § 4.3):

“As for all [these different forms of] ‘clinging to entities’ (dharma-grāha) that have been described thus (i.e. in the preceding lines of the text), dharmas outside [one’s] own mind are partly present, partly absent, whereas dharmas inside [one’s] own mind are invariably present.” (如是所 說 一切法執, 自心外法 或有或無｡ 自心內法 一切皆有｡) [4.2] Still, we have to ask the question what, precisely, is meant by these dharmas, which I shall, for the sake of convenience, call ‘external’ and ‘internal’ dharmas. A first possibility to be considered is understanding these ‘external’ and ‘internal’ dharmas in a general sense, i.e. as the constituents of the complex of conditions which in the case of the two forms of dharmagrāha based on theoretical reflection (分別: vikalpita) is stated to include ‘external’ conditions (外緣)38,39 whereas the inborn (俱生: sahaja) dharmagrāha arises from previous impregnations (熏習: vāsanā), i.e. an ‘internal’ cause (內因),

only.40 But the text contains a passage (sc. CWSL 2a9-29) where the various forms of clinging to a Self (我執: ātmagrāha) are described in a way which is exactly parallel in [p.27]structure to the aforementioned description of dharmagrāha. In the corresponding part of this passage, we are told that in the case of the various forms of clinging to a Self ‘external’ skandhas (自心外 蘊) are partly present, partly absent, whereas ‘internal’ skandhas 38 K’uei-chi does not specify what, precisely, these ‘external’ conditions are, but perhaps the expression refers to the wrong teachings and wrong reflections mentioned subsequently; cf. also YoBhū 162,11- 12.

39 CWSL 7a5; cf. 2a17. 40 CWSL 6c27-28; cf. 2a10-11.

(自心內蘊) are invariably present.41 In view of the canonical statement, actually quoted in the CWSL (2a27-29), that all views of a Self have the five upādāna-skandhas, or at least one or another of them, as their object,42 the reference to the skandhas in the context of clinging to a Self unambiguously refers to its objective support.43 Hence, in the case of clinging to dharmas, too, the reference to ‘external’ and ‘internal’ dharmas is most naturally understood as pointing to its objective support.44 Actually, the immediately following sentence makes clear that the ‘internal’ dharmas consist in the image (相)45 that appears like dharmas (似法), which is manifested within [one’s] own mind (自心所現)46 and which [p.28]must be its objective support (所 41 CWSL 2a24-25 (P 19): 如是所說一切我執, 自心外蘊 或有或無｡ 自心內蘊 一切皆有｡

42 Saṁyuttanikāya III 46 (no. 22.47; cf. Tilmann VETTER, The ‘Khandha Passages’ in the Vinayapiṭaka and the four main Nikāyas, Wien: Verlag der Österreichischen Akademie der Wissenschaften 2000: 193–194) T vol. 2 no. 99: 16b15-16. Cf. also T vol. 27 no. 1545: 38a14-15 (quotation) and T vol. 30 no. 1579: 788a6 ff (commentary on the Sūtra). In the Saṁyuttanikāya passage, the fact that the upādānakkhandhas are the object of the conception of a Self is expressed by their being the direct object of the verb samanupassanti (cf. also YoBhū 162,12-13). In the Saṁyuktāgama parallel, they are marked by the preposition 於. Hsüan-tsang uses 緣 instead, which may have been chosen in order to explicitly point out their functioning as the objective support.

43 Implicitly confirmed by Shu-chi 250c9-10 and c20. 44 Cf. Shu-chi 292c22-23: “如是所說”下 (i.e. CWSL 7a12-13), 顯執 所緣或無或有｡

45 CWSL 7a14 (似法相). Cf. 2a25-26 (五取蘊相) and 2a26 (諸蘊相). 46 CWSL 7a13-14 (P 81–82): “Therefore, the [various forms of] clinging to entities (dharma-grāha) invariably take as their objective support (緣 x = x-ālambana) an [image] appearing like dharmas [that is] manifested within one’s own mind (cf., however, § 4.3!) [but] falsely apprehend it as really existing [apart].” (是故法執, 皆緣自心所 現似法, 執為實有｡) In the case of ātmagrāha (CWSL 2a25-26), only the term 相 = nimitta is used (see fn. 45).

28 On the Problem of the External World in the CWSL 緣[[[緣]]]: ālambana[[[pratyaya]]]), in analogy to, e.g., the sense consciousnesses, in the case of which the objective support is defined as the image (相) appearing like visible matter, etc. (似色 等), this image being developed by one’s own consciousness (自 識所變: *sva-vijñāna-pariṇāma).47

But what are the ‘external’ skandhas or dharmas which are stated to be, in some cases, involved in the arising of the clinging to a Self or to dharmas? It is self-evident that they cannot be the hypostatized dharmas, i.e. those qualified, a few lines later (CWSL 7a18), as “apart from consciousness” (離識) and as “not really existent” (非實有), like the hypostatized Self (ātman) in the case of ātmagrāha. On the contrary, while in the cases of purely fictional conceptions of Self (like the ātman of the Vaiśe- ṣikas) or dharmas (like the Sāṅkhya concept of ‘primary matter’ (prakṛti or pradhāna) or the Vaiśeṣika concept of ‘substance’ (dravya-padārtha)) ‘external’ skandhas or dharmas seem to be absent,48 the CWSL itself makes clear that their presence is invariably required precisely in those conceptual activities that are inborn (俱生: sahaja).49 What [p.29]kind of factors, then, are referred to by these ‘external’ skandhas or dharmas? Do they in- 47 CWSL 4b24-25:自識所變 似色等相 為所緣緣｡ Cf. also 4b3-4

(P 42; C 25; S 31): “Since in the case of these [[[sense-consciousnesses]], i.e.] visual consciousness etc.[,] an external objective support (here clearly referring to matter existing outside any form of mind), [can]not reasonably [be taken to] exist, one must necessarily admit that [an image] developed by one’s own consciousness functions as [their] objective support.” (此眼等識, 外所緣緣 理非有故, 決定應許 自識所變 為所緣緣｡) 48 Cf. Shu-chi 250c16-17 (cf. also c18-19) and 293c25-26. 49 Cf. CWSL 40c27-29 (P 447; C 247; S 327): 第七心品 ..., 是俱生 故, 必仗外質; Shu-chi 249c24: “... because it is not [possible] that [a conception] might be inborn without having [as its ‘remote objective support’] an ‘original’ (bimba) [of its own image]” (... 非無本質 是俱 生故｡).

clude the independently existing matter (rūpa) postulated by LUSTHAUS? [4.3] Let me start with a comparatively easy issue. One of the two inborn (sahaja) conceptions of Self, and also of dharmas, is based on the ‘seventh consciousness’, manas, and is stated to be continuous. The inborn conception of Self of the ‘seventh consciousness’ is defined as follows (once again, the translation is preliminary, for the time being): “The first [inborn conception of Self] is continuous and occurs in the ‘seventh consciousness’. Taking the ‘eighth consciousness’ (i.e. ālayavijñāna) as its objective support, it produces an image (nimitta) in [one’s] own mind (自心 相) and conceives of it as a real Self.” (CWSL 2a12-14: 一, 常相續, 在第七識｡ 緣第八識, 起自心相, 執為實我｡ / P 17; C 13; S 17–18.)50

Being inborn (俱生: sahaja), the conception of a Self based on the ‘seventh consciousness’ must have, as its objective support, not only ‘internal’ skandhas but also ‘external’ ones (see § 4.2 with fn. 49). Since the ‘internal’ skandhas are, in analogy to the ‘internal’ dharmas, to be understood as an image (nimitta) in one's own mind (自心相), they are easily identifiable in our definition because the expression used here is the same as there.51 As for the ‘external’ skandha(s),52 the only candidate in 50 The continuous conception of dharmas based on the ‘seventh consciousness’ is defined analogously (CWSL 6c29–7a2: 一, 常相續, 在第七識｡ 緣第八識, 起自心相, 執為實法｡ / P 80; C 39; S 58). 51 Cf. Shu-chi 250c20 (ad CWSL 2a25), expressly equating the ‘internal’ skandhas with the ‘close objective support’: 論: 自心內蘊 一 切皆有｡ 述曰: 親所緣也｡

52 Actually, according to the position which seems to be favoured by Hsüan-tsang (cf. CWSL 21c17–22a13, esp. 22a7-8) the clinging to a Self of the ‘seventh consciousness’ may be taken to have only the

• darśanabhāga of the ālayavijñāna (應知 此意 但緣藏識見分), hence

only the skandha vijñāna, as its objective support. Cf. also Shu-chi 30 On the Problem of the External World in the CWSL our present definition [p.30]is the ‘eighth consciousness’, the ālayavijñāna.53 This is confirmed by the CWSL itself. In its systematic treatment of the notion of ‘objective support’ the

CWSL54 distinguishes between a ‘close objective support’ (親所 緣緣) and a ‘remote objective support’ (疎所緣緣). The former is invariably present in all mental factors, whereas the latter is present in certain forms of mind but not in others (CWSL 40c19- 21). It is obvious that this dichotomy corresponds to the distinction between ‘internal’ and ‘external’ skandhas or dharmas at CWSL 2a24-25 and 7a12-13. Actually, the ‘close objective support’ is defined as not separated from the mental factor that takes its form or image (相), whereas the ‘remote objective support’ or ‘original’ (/‘model’ /‘prototype’?) ((本)質: *bimba)55 is separated from it and expressly characterized as ‘external’ (外)56. [p.31]The CWSL is quite explicit in stating that the ‘seventh

249c17 (第七識中唯緣別識蘊｡) and SNSṬ Ti 257b1 (... nyon mongs pa can gyi yid ni / kun gzhi rnam par shes pa'i lta (Peking wrongly lte) ba'i rnam pa la dmigs shing bdag dang chos su 'dzin par byed do //). 53 That ālayavijñāna is included in the scheme of the five skandhas is clear from, e.g., CWSL 15a23-27, where the ālayavijñāna is equated with the *āsaṁsārika-skandha(s) of the Mahīśāsakas; cf. also Mahāyānasaṁgraha (ed. É. LAMOTTE, Louvain 1938) I.11.3. 54 CWSL 40c14-19 / P 445–446; C 246–247; S 326–327. 55 Cf. A. HIRAKAWA (ed.), Index to the Abhidharmakośabhāṣya, pt. 1: Tokyo 1973, s.v. bimba; Saṁdhinirmocanasūtra (ed. E. LAMOTTE) VIII.7; H. NAKAMURA, Bukkyō-go Daijiten: Tokyo 1975: 1264c: ... 影 像 (相分) のよりどころとなるもの; F.S. COUVREUR, Dictionnaire classique de la langue chinoise, repr. Kuangchi Press 1966: 883f (s.v. 質): “matière, substance, ... ; base, fondement; ... témoignage, preuve, garantie”. Cf. also L 501, who opts for the rendering “hyle”, which to my mind suggests something too amorphous. My impression is that in Hsüan-tsang's use of the term (本)質the aspects of “basis” and “original” are more relevant than the aspect “stuff”.

56 CWSL 40c21; cf. 41a2. Cf. also Shu-chi 606a26 expressly referring ‘externality’ to the ‘remote objective support’ (又 疎所緣緣 亦是 外), i.e. the ‘original’ (本質: T vol. 67 no. 2266: 915b16-17), and distinguishing this use from ‘externality’ in the sense of ‘not being a dharma developed by mind’ (606a24-25: 以非心所變法 說之為外). Discussion of the Basic Evidence 31 consciousness’, being inborn, is, necessarily, based on such an “external ‘original’” (外質): “As long as it is not yet fundamentally restructured [by the Bodhisattva path], the cluster57 of the ‘seventh consciousness’, because of being inborn (sahaja, i.e. spontaneous), necessarily relies on an external ‘original’ (*bimba); hence, it invariably also has a ‘remote objective support'.” (CWSL 40c27-29: 第七心品 未轉依位 是俱 生故 必仗外質｡ 故亦定有 疎所緣緣｡ / P 447; C 247; S 327.)

Since the only objective support of the ‘seventh consciousness’ is the ‘eighth consciousness’,58 this external ‘original’ can only consist in the ‘eighth consciousness’ or ālayavijñāna, as K’ueichi rightly explains.59 It is thus clear that the ‘external’ skandhas of CWSL 2a24-25 and, in the same way, the ‘external’ dharmas of 7a12-13 are not necessarily extra-mental entities but may well be mental factors outside the specific mental cluster of the respective ātma- or dharmagrāha, just as in the case of the ātmagrāha (and, analogously, the dharmagrāha) of the ‘seventh consciousness’. Accordingly, the ex[p.32]pressions 自心內 and 自心外as well should not be understood as “inside/outside one’s own mind” in the comprehensive sense of the whole, multi-layered mental continuum of a person or sentient being. They should rather be 57 I.e. the respective mind (心, citta) together with its mindassociates (心所, caitta).

58 CWSL 21c17 ff (cf. fn. 52), especially 22a13 (P 252; C 130; S 173): “As long as it has not yet been fundamentally restructured [by the Bodhisattva path], [the ‘seventh consciousness'] exclusively takes the ‘eighth consciousness’ as its objective support.” (未轉依位 唯緣藏 識｡) Cf. also 42c11-14. 59 Shu-chi 501c8-10 (ad CWSL 40c28): “As for the cluster of the ‘seventh consciousness' ..., [my] commentary says that this mind ... necessarily relies on the ‘eighth consciousness' as its external ‘original’.” (第七心品 ...｡ 述曰｡此識 ... 必仗第八識 以為外質｡) 32 On the Problem of the External World in the CWSL taken as referring to skandhas or dharmas inside/outside the specific form or moment of mind with which the respective ātma- or dharmagrāha is associated.60 A more precise rendering of the passage CWSL 2a24-25 (translated at the beginning of § 4.1) would thus be:

“As for all [these different forms of] ‘clinging to entities’ (dharma-grāha) that have been described thus (i.e. in the preceding lines of the text), dharmas outside [its = the respective dharmagrāha’s] own mind61 are partly present, partly absent, whereas dharmas inside [its] own mind are invariably present.” (如是所說 一切法執, 自心外法 或有或無｡ 自心內法 一切皆有｡)

The same holds good for the expression 自心相, which therefore should, in this context, rather be rendered as “image in [its = the respective dharma- or ātmagrāha’s] own mind”, or even as “a mental image of its own”.62 This is confirmed by K'uei-chi who, in [p.33]the case of the conception of Self of the ‘seventh consciousness’, specifies the “image in [its] own mind” of CWSL 2a13-14 as “the image in [its, i.e. the ātmagrāha’s] own 60 Cf. Shu-chi 501a15-17, where the “condition consisting in the remote objective support” (疎所緣緣), characterized as being “outside the [respective] mind” (心外: 501a28-29), is stated to comprise both “[images, or appearances] developed by consciousnesses of others and [images, or appearances,] developed by separate [[[forms]] of] consciousness in one’s own continuum” (疎所緣緣, 與能緣心相離法, 是｡ 謂 即他識所變, 及自身中別識所變｡ 杖為質者, 是｡). Similarly, Shuchi 250c14-15.

61 Or could one understand: “outside the [respective] mind itself” and, analogously, in the next line: “inside the [respective] mind itself”? 62 As far as I can see, the passages under discussion do not distinguish the ātma- or dharmagrāha as a caitta from the citta itself but rather seem to take it as the function of the citta itself. If such a distinction is made, the rendering “a mental image of its own” would definitely be preferable. For even though the image-parts of all mental factors (citta and caittas) of a given ‘cluster’ (品) are alike, each factor develops an image-part of its own.

mind, [i.e. in] the ‘seventh consciousness’” (Shu-chi 249b27-28: 是第七識自心之相).63 Therefore, the passage translated in the beginning of § 4.3 (viz. CWSL 2a13-14) should rather be rendered as follows:

“The first [inborn conception of Self] is continuous and occurs in the ‘seventh consciousness’. Taking the ‘eighth consciousness’ (i.e. ālayavijñāna) as its [remote] objective support, it produces an image (nimitta) [of this remote objective support] in [its] own mind (or: a [[[corresponding]]] mental image of its own) (自心相) and conceives of it as a real Self.” (一, 常相續, 在第七識｡ 緣第 八識, 起自心相, 執為實我｡)

[4.4] Let me now turn to the inborn (but intermittent) clinging to a Self and to dharmas that is associated with the ‘sixth consciousness’, i.e. manovijñāna. Its definition runs thus: “The second [kind of inborn clinging to a Self] is intermittent and occurs in the ‘sixth consciousness’. Taking as its objective support an image (相) of the five appropriated skandhas (or, in the case of dharmagrāha, of skandhas, āyatanas and dhātus) — together or separately — that is developed by consciousness (識所變), it produces an image in [its] own mind (or: a mental image of its own) (自心相) and conceives of it as a real Self (or dharmas).” (CWSL 2a14-15: [p.34]二, 有間斷, 在第六識｡ 緣識所變五取蘊相, 或總或別, 起自心相, 執為實我｡ / P 17; C 13–14; S 18.– CWSL 7a2-3: 二, 有間斷, 在第六識 ｡ 緣識所變蘊處界相, 或總或別, 起自心相, 執為實法｡ / 63 In the case of the inborn clinging to a Self of the ‘sixth consciousness’ (i.e. manovijñāna) it is the sub-commentator Chih-chou 智 周 who makes clear that the expression “image in the [[[Wikipedia:Cognition|cognizing]]] mind itself (or: in [one’s] own mind)” in the CWSL refers to “what is developed by the sixth consciousness [and is hence its] own, immediate image-part” (Yen-pi 825a23-24: 第六所變自親相分, 下文別說: 起自 心相 ...｡).

34 On the Problem of the External World in the CWSL P 80–81; C 39; S 58.) Here too it is perfectly clear that the skandhas or dharmas outside the cognizing mind that are involved are mental phenomena, more precisely: images or appearances (相) developed by consciousness (識所變, vijñāna-pariṇāma).

To be sure, there is a problem here, viz. that the ‘external’ elements that are reproduced (and then misconceived) by the ātma- or dharmagrāha of the ‘sixth consciousness’ are defined as mental images of sets of factors (skandhas, etc.) that include mind and mind-associates. Why is it not mind and mindassociates themselves that are taken to be the ‘external’ objective support, just as the ālayavijñāna in the case of the ‘seventh consciousness’? The problem has been seen and discussed by later commentators.64 But I need not go into details since this is not crucial for my main issue, i.e. independently existing matter. With regard to matter (色: rūpa), the definition does not pose any problem,65 and it would indeed [p.35]seem that it was in the first place matter, corporeal or other, that Hsüan-tsang had in mind when coining it. At any rate, the passage clearly shows that the material entities that function as the ‘external’ objective support of the inborn conception of Self or dharmas of the ‘sixth 64 Yen-pi 825a25–b1; T vol. 67 no. 2266 (by Tan-e 湛慧, 1675– 1747): 117b10-22 (quoting the I-yün; see fn. 65).

65 Cf. T vol. 67 no. 2266: 117b9-10 and b14-15 (quoting the I-yün 義 蘊), where it is first stated in connection with the inborn conception of Self of the ‘sixth consciousness’ that in this case the ‘sixth consciousness’ apprehends the nimitta (相) of the five upādāna-skandhas developed by the ‘eighth consciousness’ but that this expression is unproblematic only with regard to matter, whereas the four immaterial skandhas (i.e. mind and mind associates), not being part of the objective support of the ‘eighth consciousness’, cannot be called ‘developed [by mind]’ in the usual sense, and, if they are called so in another sense (cf. Shu-chi 317a18–b7), cannot be called 相, at least not in the sense of ‘image’ (... 第六意識, 緣第八識所變五取蘊相也｡ 問｡ 若爾, 第八所變 色可爾｡ 餘之四蘊, 第八不緣, 如何稱變｡ ... 問｡ 變義可爾｡ 四蘊既非 第八影像, 何名蘊相｡ ...).

consciousness’ are understood by Hsüan-tsang exclusively as images (相) of material entities in consciousness (識, vijñāna). But in which form of consciousness? Hsüan-tsang does not specify it, and the commentators disagree.66 At any rate, images of material things are, intermittently, available in the sense consciousnesses. 67 Besides, an image of one's own body and material sense-faculties and a — more or less complete — image of the surrounding material world and also of the bodies of other sentient beings68 is continuously produced by the ālayavijñāna of every sentient being:69

“When the ālayavijñāna itself arises due to its causes and conditions70 it develops internally into … the body possessed of sense-faculties, and externally into the surrounding [[[world]]] (bhājana), and it takes these very [images] into which it has developed as its object (ālam-bana ).” (CWSL [p.36]10a17-19: 阿賴耶識 因緣力故 自體 生時 內變為 ... 有根身, 外變為器｡ 即以所變 為自所緣 / P 125; C 60–61; S 78.)71 “Due to the ripening of common seeds (*sādhāraṇa-bīja), 66 Cf., for ātmagrāha, Yen-pi 825a22-24; I-yen 24a6-12; for dharmagrāha, Yen-pi 854a1-10.

67 Cf., e.g., CWSL 4b3-4 (see fn. 47). 68 That the bodies of others belong, properly speaking, to the surrounding world is explicitly stated in Tun- (/Tao-) lun’s commentary on the Yogācārabhūmi (T vol. 42 no. 1828: 602b23: 然變為他身, 是即 外器所攝｡).

69 According to T vol. 67 no. 2266: 117b10 (see fn. 65), it is this image of corporeal matter in the ālayavijñāna that is the ‘original’, i.e. the ‘external’ objective support, of the image in the natural conception of Self of the ‘sixth consciousness’.

70 I.e., according to Shu-chi 317a9-10, the direct seeds as hetu(pratyaya) (親因種) and the karmic seeds as (adhipati)pratyaya (業緣種). 71 There is a Tibetan translation of this passage at SNSṬ Ti 269a3-4 in which 變為 X (“develops or changes into”) is translated by X lta bur snang (“appears as X”).

36 On the Problem of the External World in the CWSL the consciousness which is [the result of karmic] maturation (vipākavijñāna)72 develops into an image appearing as the surrounding world (lit. ‘container world’: bhājanaloka) [consisting] of visible matter, etc., i.e. the external gross elements (mahābhūta) and secondary matter (upādāyarūpa). Although the [images of the surrounding world] developed by each sentient being are [numerically] different,73 their appearance is similar and their location without distinction; just as the lights of many candles [in a room], each filling the whole [room], look as if they were one [and the same light, so the images of the surrounding world in the vipākavijñānas of different beings, each filling the same space, look as if they were one and the same].” (CWSL 10c13-16: 異熟識 由共相種成熟 力故 變似色等器世間相｡ 即外大種 及所造色｡ 雖諸有情 所變各別, 而相相似 處所無異｡ 如眾燈明 各遍似一｡ / P 135–136; C 64; S 81.)

“Due to the ripening of special seeds (*asādhāraṇa-bīja), the consciousness which is [the result of karmic] maturation (vipākavijñāna) develops into [an image] appearing as [one's own subtle] material sense-faculties (rūpīndriya) 74 [p.37]and [one's own gross body which is] the sup- 72 Another designation of the ‘eighth consciousness’ with a wider range of application than ‘ālayavijñāna’ (cf. CWSL 13c13-19). 73 Each sentient being has its own ālayavijñāna (or at least its own ‘eighth consciousness’): see CWSL 2b5-6; YAMABE (s. fn. 1): 21. 74 Cf. also CWSL 20a25-26 (P 235; C 119; S 162): “The ālayavijñāna develops into what appears as the material sense-faculties, the support of the sense-faculties, the surrounding world, etc.” (阿賴耶識 變似色 根 及根依處 器世間等); 41a10-11 (P 449; C 248; S 328): “The first five [i.e. the] material [[[sense]]-]faculties consist in (…為性 = -svabhāva) the clear matter (rūpaprasāda) of [the faculty of] sight (cakṣus), etc., [this clear matter being an image] developed by the `root consciousness’ (mūlavijñāna, i.e. the ‘eighth consciousness’), etc.” (前五色根 以本識 等所變眼等淨色為性). I am sure that LUSTHAUS's (L 495 n. 25) interpretation of the passage is untenable. He renders the sentence as fol

port of [these] [p.38]sense-faculties (indriyādhiṣṭhāna), i.e. the internal gross elements and secondary matter. There are [other] common seeds due to the ripening of which [the vipākavijñāna develops into yet another image:] where there are bodies of others,75 it also develops lows: “The five rūpa organs, which alterations (so-pien) in the mūlavijñāna, etc., take in through the eyes, etc. Pure rūpa [(]淨色 ching-se) is regarded as their nature.” According to LUSTHAUS, the purport of the passage is that “visible experience of ‘pure rūpa’ is registered as ‘alterations in the mūla-vijñāna, etc.’” and that “far from implying a causative idealist theory, the ālayavijñāna … and other consciousnesses are passive recorders (so-pien) of the activities of the visual organ and its corresponding object.” Apart from contradicting the more explicit passages adduced above, LUSTHAUS's translation (somewhat

odd even in English; should one perhaps read the passage as one sentence by introducing a colon: “… etc.: Pure …”?) is, as far as I can see, entirely incompatible with the rules of Chinese syntax since he seems to take 以 as the main verb (“take in”), followed by 本識等所 變 (“alterations in the mūla-vijñāna, etc.”) as the subject, which in its turn would be followed by an unmarked instrumental element 眼等 (“through the eyes”). The initial expression 五色根 (not “rūpa organs” but the subtle material sense-faculties, rūpīndriya = rūpin + indriya) which is, at least ad sensum, (rightly) connected with the final portion of the passage, is at the same time also construed as the object of the verb 以 (“take in”), which leads to the consequence that organs or sense-faculties are both the object and the means of the taking-in or registering. Actually, there can be little doubt that the passage is construed on the common pattern A 以 B 為性 (“A, taking B, makes it its

nature”, “A takes/has B as its nature”, “A consists in B”). Hence, 本識 等所變眼等淨色 must be the definiens (B), constituted by the traditional 眼等淨色 (= cakṣurādi-rūpaprasāda) qualified by the attribute 本識等所變 (“into which the ‘root consciousness’, etc., develops, or: has developed”, i.e., “developed by the ‘root consciousness’, etc.”; cf. also CWSL 56b26) which is specifically Yogācāra. I admit that the 等 (“etc.”) in this expression requires explanation; one may think of the mind-associates of the eighth consciousness or of the sixth consciousness (cf. T vol. 67 no. 2266: 160a18-20), but I am not sure whether this is the right track.

75 I am not sure whether my understanding of 於 … 處 (disregarded by P and C) is correct. I presume Hsüan-tsang wants to say: where the bodies of other sentient beings are located in the ‘imagepart’ of the vipākavijñānas of the beings whose bodies they are. A dif38 On the Problem of the External World in the CWSL into [an image] appearing as these; otherwise it would not be possible to experience [the physical presence of] others.” (CWSL 11a8-11: 異熟識 不共相種成熟力故 變似 色根 及根依處｡ 即內大種 及所造色｡ 有共相種成熟力故 於他身處 亦變似彼｡ 不爾應無受用他義｡ / P 138; C 66; S 83.)

[5.] But what about sense consciousnesses or the ālayavijñāna? Are they, or the internal images of material entities which they develop, in their turn also based on a ‘remote objective support’ or external ‘original’? And if so, couldn't at least this external ‘original’ be independently existing matter? [5.1] As for the first question, the CWSL is unambiguous in the case of the sense consciousnesses: “The clusters of the first five [[[forms]] of] mind (i.e. the sense consciousnesses), as long as they are not yet fundamentally restructured, ... are necessarily based on76 an external ‘original’ and hence invariably also have a remote objective [p.39]support.” (CWSL 41a3-4: 前五心品, 未轉依位, ... 必仗外質, 故亦定有疎所緣緣｡/ P 448; C 248; S 328.)

[5.2] In the case of the ālayavijñāna, the CWSL reports disagreement. According to one opinion, it has no remote objective support but is merely determined by karmic forces (CWSL 40c21-22). But according to others, at least the image of those ferent understanding of 處 has been suggested by Koichi TAKAHASHI in his review of the present essay (in: International Journal of South Asian Studies [Manohar] 1/2008: 173‒177). 76 LUSTHAUS (L 504) translates 仗 by “confronted with”, which seems to be based on the meaning “weapons of war; to fight”, but this appears far-fetched. 仗 as well as 杖 (the reading preferred in the Shuchi, e.g. 501c15) also have the meaning “to rely on”. That this is the meaning in which the word is used by Hsüan-tsang is evident from CWSL 20c4, stating that ‘basis’ (依) refers to the fact that in order to arise and stay on all impermanent dharmas rely on causes and are dependent on conditions (仗因託緣). Cf. also Shu-chi 317a14: 杖謂杖託.

parts of the material world which are experienced alike by many sentient beings — i.e. the surrounding world as well as gross bodies — is based on an ‘original’ (40c22-27: two positions, the second, more restrictive one being authoritative). But this ‘original’ (and this answers the second question) is explicitly defined as consisting in the [images] developed by others, i.e. others' minds:

“There are [some who] assert that [the ālaya- or vipākavijñāna] invariably has also a remote objective support because it must rely on an ‘original’ (質) [consisting in an image] developed by [the consciousnesses of] others (他變)77: only then it develops its own [image].78 There are [others [p.40]who] assert that ... one's own [[[body]]] and others' bodies as well as the earth (i.e. the surrounding world) can be mutually experienced [only] because the [[[corresponding]] image(s)] developed by [the minds of] others function as the original of one's own [[[mind]], i.e.

77 In the terminology of the CWSL (cf. fn. 12), 變, for 所變 (cf. CWSL 40c24-25: 他所變) = pariṇāma, is an abbreviation for 識所變 (vijñāna-pariṇāma) and inevitably suggests mental images (or at least mental aspects, if we take the ‘aspect of vision’, the 見分, into account). Thus, 他(所)變is a shorthand for 他識所變. Actually, in his commentary on the definition of the ‘remote objective support’ (see fn. 60) K'uei-chi expressly states that it consists not only in [images, or appearances,] into which other consciousnesses of the same person have developed, but also in [images] into which the consciousnesses of other persons have developed (他識所變). No further kind of ‘remote objective support’, existing independently of any form of mind, is mentioned.

78 LUSTHAUS (L 502) translates: “... since it is the influence of others' changes (他變 …) that is the hyle directing one's own changes (自方變 ...)”. Does this mean that he takes 方 as a verb (“to direct”) and 變 as its object (together with 自, which seems odd)? Actually,方 is rather a conjunction here (“then only”), as is supported by K'ueichi's paraphrase of the sentence: “This eighth consciousness must rely on [images] developed by others[' consciousnesses] as its ‘original’, then only (方) it is able to develop [one] itself” (Shu-chi 501b9-10: 此 第八識 要杖他變為本質 方能自變). 40 On the Problem of the External World in the CWSL ālayavijñāna, and vice-versa]. …” (CWSL 40c22-25: 有義: 亦定有疎所緣緣, 要仗他變質 自方變故｡ 有義: ... 自他身 土可互受用, 他所變者為自質故｡ ... / P 447; C 247; S 327.)

It is hardly conceivable that these images developed by others’ minds could be anything other than the corresponding images in their ālayavijñānas; for it is only these images that are similar to the image in one's own ālayavijñāna. A fully explicit statement to this extent is found in the commentary on the Yogācārabhūmi compiled by K'uei-chi's Korean collaborator Tun-lun 遁倫 (or Tao-lun 道倫)79: “One should not raise the objection that since the ‘eighth consciousness’ has no ‘original’ (本 for 本質), [one cannot see] how [in its case mind and mind-associates] could have the same object (if this means that they are based on the same ‘original’: a12-13); for the ‘eighth consciousness’ has images developed by the ‘eighth consciousnesses’ of others as its ‘original’.” (T vol. 42 no. 1828: 317a16-17): 不應難言: 第八無本, 如何說云同一所 緣｡ 用他第八所變諸相為本質故｡)

[5.3] Since even the image of the material world in the ālayavijñāna is not based on an independent material world but on other mental images (viz. those in other beings' ālayavijñānas), it is highly improbable that the situation is essentially different in the [p.41]case of the sense consciousnesses: In view of the fact that in their case an external object (in the strict sense) is categorically rejected,80 their remote objective support, too, can 79 See fn. 22.

80 CWSL 4b3-4 (see fn. 47). Cf. Shu-chi 269b22-23 and 270a8-9: “An objective support [completely] outside mind, however, is definitely non-existent. ... If we do not rule out dharmas outside [the respective cognizing] mind as functioning as a remote objective support, this is just [in the sense] that the eighth [[[consciousness]]] functions as the ‘original’, resting on which the remaining consciousnesses develop

hardly be independent matter, and is most likely to be understood as images of matter in some other, deeper form of mind. What suggests itself as ‘original’, at least in the case of ordinary consciousnesses, is, of course, the image of the material world (surrounding world as well as bodies) in one's own ālayavijñāna. This is confirmed by an unambiguous statement which the CWSL (42c11-15) makes in the context of a systematic clarification of the causal relations between the different forms of consciousness within one and the same continuum. With regard to the objective support (所緣緣: ālambanapratyaya) we are told that only the eighth consciousness (viz. ālaya- or vipākavijñāna), and neither the seventh consciousness (viz. manas) nor the sixth (viz. manovijñāna), can function as objective support for the five sense-consciousnesses (nor, of course, can they be an objective support for one another),

“... for the five [[[sense]]-]consciousnesses base themselves exclusively on the image[-part] of the eighth [[[consciousness]]]” (CWSL 42c14-15: 五識唯託第八相故｡ / P 469; C 259; S 349; cf. L 505).

[p.42]K'uei-chi paraphrases this passage as follows: “... for the five sense-consciousnesses take [an image] developed by the ‘root consciousness’ (本識: mūlavijñāna, i.e. the eighth consciousness) as their object;

they do not presuppose visible matter (色: rūpa), etc.81, as developed by the sixth consciousness (i.e. manovijñāna) [into a corresponding image of their own].” (然心外所緣緣 決定非有 ｡ ... 若不遮心外法為疎所緣緣, 即是第八為質, 餘識託之而變｡) Cf. the explanation of this passage at I-yen 38c6-8 = T vol. 67 no. 2266: 166a25-28: “... Even though there are real (實體) dharmas which function as a remote objective support, still these ‘original’ dharmas are merely [images] into which the ‘eighth consciousness’ develops, and they are not [something] really existing (實有) apart from consciousness.” (... 雖然有實體法 為疎所緣緣, 然此本質法 但是第八識所變, 亦不是離識實有｡)

81 In this passage, “etc.” clearly refers to objects of the other objects of sense consciousnesses (viz. sound, etc.). 42 On the Problem of the External World in the CWSL as their own object.” (Shu-chi 512b24-25: 五識 緣本識所 變為境 不待第六識所變色等為自境故｡)82 The same idea is expressed by K’uei-chi also in other places: “... because functioning as their ‘original’ (本質), visible matter, etc., [as developed in] the image[-part] of the eighth [[[consciousness]]] generate visible matter, etc., as the image-part of the five [[[sense]]-]consciousnesses.” (Shu-chi 512b10-11: 以 第八相色等 為其本質 生五識相分 色等 故｡)

Or, still more explicitly: “... The eighth consciousness develops [an image] appearing like the five sense faculties and the five sense objects. The five sense-consciousnesses – visual [[[consciousness]]], etc. –, basing themselves on those sense faculties developed by the [[[eighth consciousness]]], take those ‘original’ sense objects (i.e. those developed by the eighth consciousness) as [p.43]their objective support. ... In reality (i.e.: to be more precise), on the basis of the [image of] the material sense objects in the ‘root consciousness’ (本識: mūlavijñāna), [the sense-consciousnesses] appear [in such a way as to] develop [their own] image of the five sense objects. ... If the five [[[sense]]] consciousnesses did not base themselves on [the image of

the sense objects] developed by the ‘eighth consciousness’, then they would not have a [complete] objective 82 Shu-chi 501c16-19, on the other hand, though stressing that the five sense consciousnesses must be based on an external ‘original’, takes this ‘original’ to be an image either in the ‘eighth consciousness’ (i.e. the ālayavijñāna) or in the ‘sixth consciousness’ (i.e. in the manovijñāna) (眼等五識, ... 必杖第八或第六所變外質 方起｡). Yet, this specification seems to refer to non-ordinary consciousnesses of matter, as in meditation (cf. I-yen 261d1 ff = T vol. 67 no. 2266: 697c1 ff), which may be disregarded in the present context (cf. also the discussion in K'uei-chi's Ch'u-yao 樞要 [T vol. 43 no. 1831] 648c19 ff, and especially in T vol. 66 no. 2263 [by Ryōzan 良算, around 1200]: 449b13 ff).

support; for [in the case of the sense consciousnesses] the objective support includes a close one and a remote one.” (Shu-chi 268b7-13: 第八識變似五根．五塵｡ 眼等五識, 依 彼所變根, 緣彼本質塵｡ ... 實於本識色塵之上 變作五塵 相現｡ ... 五識若不託第八所變, 便無所緣｡ 所緣之中有親 疎故｡)

A succinct statement to the same effect is also found in Tun- (or Tao-) lun's commentary on the Yogācārabhūmi: “Visual consciousness and its mind-associates are stated to have the same objective support [because] they arise [supported] by an ‘original’ sense-object [consisting in an] image part developed by the ālaya[[[vijñāna]]] that is similar [to their own image part].” (T vol. 42 no. 1828: 317a12-13: 眼識及心所, 同類賴耶所變相分本質塵起, 名 同一所緣｡)

[5.4] That this explanation is not an idiosyncratic view of K'uei-chi and his school is testified by a passage from Tsongkha- pa's treatise on manas and ālayavijñāna (Yid dang kun gzhi'i dka' 'grel):83 [p.44]“As for the object part of the ālayavijñāna, it is apprehended by the five [[[sense]]-]consciousnesses” (kun gzhi'i gzung cha dag la ni // rnam shes lnga yis 'dzin byed ...).

83 Tshultrim KELSANG and Nobuchiyo ODANI, Tsonkapa-cho Araya-shiki to mana-shiki no kenkyū, Kyoto 1986: 151, fol. 7b5. — Tsong-kha-pa was acquainted with the Tibetan version of Wŏn-ch'ŭk's commentary on the Saṁdhinirmocana-sūtra (cf. POWERS, op.cit. [see fn. 16]: 96), but not with K'uei-chi’s writings. Unfortunately, I had no time for more than a very superficial glance at Wŏn-ch'ŭk's commentary, but one can find at least the distinction between rūpa as developed (*-pariṇāma) from the ālayavijñāna (kun gzhi rnam par shes pa las gyur pa'i gzugs) and rūpa as an objective image (nimitta) developed from visual consciousness (mig gi rnam par shes pa las gyur pa'i rgyu mtshan gyi rnam pa) (SNSṬ Thi 122a6-7).

44 On the Problem of the External World in the CWSL That the image of such things or objects as are common to all in the ālayavijñāna participates in the production of ordinary consciousnesses is also confirmed by the Indian Yogācāra master Sthiramati (ca. 510–570)84, though he takes it to function only as a controlling or directing condition (adhipati-pratyaya), not as an object (viṣaya):

“As for the appearance of objects common [to all sentient beings] in the ālayavijñāna, it, too, is called ‘external basis’ on account of functioning as a ‘controlling condition’ (adhipati-pratyaya) for the arising of the actual consciousness (pravṛtti-vijñāna) appearing as the objectof- consciousness (grāhya-pratibhāsa), but not on account of being its (i.e. the actual consciousness’s) object[ive support] (viṣaya, ⋲ā lambana).”85

On the other hand, the Vivṛta-gūḍhārtha-piṇḍa-vyākhyā (an obviously comparatively late fragmentary commentary on the Mahā[p.45]yāna-saṁgraha) refers to the distinction between the image of the object in consciousness itself as its direct objective support and the ālayavijñāna as its indirect objective support. “[The objective support] is twofold in another way [than the one the other discussant had presupposed, viz.] the directly [[[perceived]]] objective support and [what is an objective support] indirectly. Among these, the direct ob- 84 Date according to Erich FRAUWALLNER, Kleine Schriften (ed. G. OBERHAMMER and E. STEINKELLNER, Wiesbaden: Franz Steiner

1982): 859; cf. also Yuichi KAJIYAMA, Studies in Buddhist Philosophy (ed. K. MIMAKI et al., Kyoto: Rinsen 1989): 200. 85 Madhyāntavibhāga-ṭīkā (ed. S. YAMAGUCHI) 146,5-9: <ya ālayavijñānasya sādhāraṇârtha-prati>bhāso ..., so 'pi grāhya-...pratibhāsasya pravṛtti-vijñānasyôtpattāv adhipati-pratyayatvād bāhyam āyatanam ucyate, na tu tad-viṣayatvād iti. See also ibid. 17,13-14 and 16-17: tatrârtha- sattva-pratibhāsam ālayavijñānaṁ. ... vijñapti-pratibhāsaṁ ... cakṣur-vijñānâdi-ṣaṭkam. Cf. L. SCHMITHAUSEN, Ālayavijñāna, Tokyo: The International Institute for Buddhist Studies 1987, II: 414 and 416 (ns. 763 and 769).

jective support is the form (i.e. image) of what is apprehended (grāhyākāra). The indirect objective support is the ālayavijñāna; for by virtue of the [[[latter]] a consciousness] appears in the form of what is apprehended.” 86 Apart from this, the Vivṛta-gūḍhārtha-piṇḍa-vyākhyā also explicitly states that entities not belonging to (i.e. other than) mind and mind-associates do not exist.87

[6.] Returning now to the disputed passage (CWSL 39c16), it would seem to me extremely unlikely that the words “visible matter, etc.” refer to any kind of matter existing independently, i.e. as something that is not just an image in some form of mind. As I have tried to show, there are a sufficient number of passages rejecting the existence of extra-mental material (or other) entities not only as objects of consciousness but also as such, without any qualification, and there is sufficient internal evidence in the CWSL, supported by the earliest commentaries and partly even by independent Indian sources, indicating that wherever consciousnesses have an external ‘original’ or ‘remote objective support’ this is understood as consisting either in other consciousnesses or mental [p.46]factors or, in the case of matter, in a mental image of matter contained in other consciousnesses, either in one's own mind continuum or in that of other sentient beings. Hence, DE LA VALLÉE POUSSIN's explanation of the disputed passage is, at least basically, in accordance with the doc-

86 Peking Tanjur, Sems-tsam, vol. Li: 395b5-7 (Derge vol. Ri: 329b3-4): ... mngon sum du dmigs pa dang brgyud pa'i sgo nas gzhan du rnam pa gnyis te / de la mngon sum gyi dmigs pa ni gzung ba'i rnam pa gang [6] yin pa'o // brgyud pa'i dmigs pa ni kun gzhi'i rnam par shes pa ste / de'i dbang gis gzung (P: bzung) ba'i rnam par snang ba'i phyir ro // 87 Ibid. 395a7: sems dang sems las byung ba la ma gtogs pa'i dngos po med pa'i phyir .../. According to Susumu ŌTAKE, the Vivṛtagūḍārthapiṇḍavyākhyā might, however, have been produced in sixth century Central Asia (ŌTAKE in: Imre HAMAR [ed.], Reflecting Mirrors, Wiesbaden: Harrassowitz 2007: 95).

46 On the Problem of the External World in the CWSL trinal position of the CWSL88 – all the more so since it is not at all merely based on his own conjecture but is in fact more or less a rendering of K'uei-chi's commentary on the passage. For K'uei-chi explains:

“[[[Consciousnesses]]] taking as their objective support matter which is the image part of another['s mind] or matter which is developed by a different consciousness in one's own person[al continuum], etc., are also [to be explained] thus (i.e. as doing so by means of a corresponding image within themselves, just like consciousness taking another's mind as its objective support).” (Shu-chi 494a8-9: 緣他相分色 自身別識所變色等 亦爾｡)89

I regard it as highly improbable – at any rate as long as no evidence to the contrary is produced from the extant works of Wŏn-ch'ŭk – that K'uei-chi, as one of Hsüan-tsang's foremost pupils, [p.47]misunderstood or even distorted the view of his teacher on such a crucial point as the independent existence of matter, the more so since his explanation is fully in accordance with the internal evidence of the CWSL itself. 88 Cf. also YAMABE 1998 (s. fn. 1): 31.

89 I-yen 243c7-10 ⋲ Shu-ch’ao 330b9-12 explains: “One’s own vijñāna, when apprehending matter (*rūpa) which is the image-part of another[’s, i.e.] somebody else['s mind], makes [this image] its remote objective support; it also develops its own image-part and makes it its close objective support. And [when] the six vijñānas, etc.a, of one’s own continuum apprehend matter which is the image-part developed by [one’s own] ‘eighth vijñāna’, they take it as their remote objective support, and these same six vijñānas, etc.a, [also] develop their own image-part which they take as their close objective support.” (自識, 若 1 緣他 (2 餘人2) 相分色, 以為本質｡ 亦自變起3 相分 (4 為親所緣｡4) 及5 自身前六識等, 緣第八識所6 變相分色7, 以為本質｡ 即六識等, 自變起 相分,8 為親所緣｡)

a This may be meant to include the mental factors associated with them. 1 I-yen: 別. (2...2) I-yen: 人餘. 3 I-yen om. (4...4) I-yen: 為親新緣; Shu-ch’ao: 之. 5 I-yen om. 6 I-yen om. 7 Shu-ch’ao om. 8 Shu-ch’ao add.:六識等. Discussion of the Basic Evidence 47 48 On the Problem of the External World in the CWSL [p.49]

III. The Spiritual Context of vijñaptimātratā

[1.] The result of my investigation of the disputed passage in its closer and broader context as well as in the light of the earliest commentaries and other evidence may not be welcomed by the critics of the traditional interpretation of Yogācāra thought. But I have to admit my amazement at the emotional vehemence of their criticism. Is it merely because Yogācāra thought as traditionally understood seems so counter-intuitive to modern

Western common-sense that some scholars think they must ‘defend’ the Yogācāras against such an understanding? But isn't this the same mode of procedure that scholars who worked when idealism was the dominant strand in Western philosophy are criticized for, viz. reading the presuppositions of one's own time and milieu into the old texts? It may be difficult to avoid doing this completely, but one can at least try one's best to understand the texts from within (and surely there are various degrees of approximation) and to make sense of them on their own premises.90 It is one of the merits of LUSTHAUS's study that he indeed tries to take into account, in his interpretation of Yogācāra thought, central concerns of the Buddhist tradition, especially karma (e.g. L 485 and 536) and attachment (e.g. L 537).

However, I find that, viewed from Buddhist premises, Yogācāra in general, and the system of the CWSL in particular, makes equally good sense if understood in the way I have shown as being closer to what the texts themselves say, i.e. in terms of a negation or at least a non-occurrence of external as well as independently existing [p.50]material things and their reduction to images in some form of mind or other, even when they function as a ‘remote objective support’.

90 It is by no means my intention to deny the legitimacy of creative philosophical interpretation, but if such an interpretation claims to be in tune with what the old texts themselves wanted to communicate, it has to put up with being checked against the actual wording of the sources in its intratextual as well as wider historical context. 50 On the Problem of the External World in the CWSL

[2.] At least for most forms of Buddhism, it would seem appropriate to state that their central concern is with sentient beings, more precisely: with their suffering (perpetuated through rebirth) and its causes, viz. spiritual defilements (kleśa, especially attachment and misconceptions) and karma, as well as with their liberation from suffering and attainment of spiritual peace or true happiness through the elimination or replacement of those causes. In Mahāyāna Buddhism, the concern is, more explicitly, not only with one's own suffering but also, and even mainly, with the suffering of other sentient beings. In understanding the CWSL, we also cannot ignore the Abhidharma

tradition which aims at designing a coherent structure of the processes of entanglement and liberation and of the factors involved. Yogācāra has been affiliated with the Abhidharmic tradition more or less from the outset.91 In a sense, the CWSL may be regarded as a kind of revision of, or supplement to, earlier Abhidharmic Yogācāra works in terms of ‘nothing but consciousness’ (vijñaptimātra).

[3.] In Buddhist thought, karma is not normally the basic cause of saṁsāra and hence of sentient beings' entanglement in suffering as such, but rather the cause of their specific situation (/form of rebirth) on which the actual quantity and quality of suffering or happiness is largely dependent, bad karma entailing an unfavourable situation (/rebirth), good karma a favourable one. Now, in Buddhism karmically productive action is defined as consisting in either intention (cetanā) itself or intentional (cetayitvā) acts,92 [p.51]which means that in any case intention, hence a mental factor, is decisive. If the experience of suffering or happiness is the final result of karma (no matter whether direct or indirect), the process starts and ends in the 91 Cf., e.g., the Abhidharmasamuccaya, but much of the Yogācārabhūmi, too, is more or less Abhidharmic.

92 E.g. Abhidharmasamuccaya (see fn. 14) 53,3 (reconstructed); AKBh 192,10; even intentional bodily and vocal karma reduced to intention: ib. 195,18-21.

The Spiritual Context of vijñaptimātratā 51

mind, and the production of a body and sense-faculties as the direct result of karma is merely an intermediate step. The significance of these material factors is further reduced in early Yogācāra by the introduction of ālayavijñāna as the only direct, primary result of karmic maturation.93 Still less important for the process of karma and its retribution is the external world. As indicated by its designation as bhājana-loka (Ch. 器世間), its function is that of a mere ‘container’ for the bodies of sentient beings. Already in the earliest Yogācāra sources, its formation and disappearance as well as its quality is regarded as a kind of byproduct of the karma of sentient beings,94 and already in the Viniścayasaṁgrahaṇī section of the Yogācārabhūmi it is the ālayavijñāna

that, probably because of carrying the karmic seeds, is stated to produce (skyed par byed pa , 生起) the ‘container world’, and at the same time is also taken to cognize it.95 Hence, not much was lost when the nexus between karma and the experience of its effect was able to be explained without recourse to material entities apart from or independent of the mind. Such a theory may even have been welcomed as more ‘economic’96 since already in earlier Yogācāra works like [p.52]the Abhidha- 93 Cf. L. SCHMITHAUSEN, (see fn. 85): 57–62. 94 YoBhū 30,21–31,1; 36,19-20; 184,6-9; Abhidharmasamuccaya (T vol. 31 no. 1605: 679b26-27; Tanjur (Peking) vol. Li: 102b6-7: las thun mong ba … gang snod kyi 'jig rten rnam par 'byed pa'o. Cf. also AKBh 94,22; 95,15; 158,1-2 (vāyumaṇḍalam abhinirvṛttaṁ sarvasattv ānāṁ karmādhipatyena); 158,6-11; 192,3-5.

95 Tanjur (Peking) vol. Zi: 8a6 and 4b1; T vol. 30 no. 1579: 581a28-29 and 580a2-4; cf. L. SCHMITHAUSEN, Ālayavijñāna (see fn. 85): 64; 90; 203. 96 On the rôle of vijñaptimātratā in Vasubandhu’s solution of the problem of karma and its fruition see J. BRONKHORST, Karma and Teleology, Tokyo: The International Institute for Buddhist Studies 2000: 67–75. According to BRONKHORST (op.cit. 77–93), reflections on this problem were even the decisive stimulus for the original intoduction of vijñaptimātratā. This is, however, not my concern in the present paper.

52 On the Problem of the External World in the CWSL rmasamuccaya there is indeed a tendency to dispense with theoretically superfluous entities.97 [4.] On the other hand, the fact that Buddhism, as a soteriology or religion, is, essentially, concerned with the welfare of all sentient beings has guarded the system against solipsism.98 In the position that is given preference in the CWSL, the causal interaction of a plurality of mental continua asserted in the Viṁ- śatikā (9,20-23) was developed into an indirect intersubjectivity also on the epistemic level, through the introduction of the notion of the ‘remote objective support’. This notion is also employed in the context of the experience of the material world, with the result that a better explanation of its being presupposed99 and intersubjective can be given even on a vijñaptimātratā basis.

[5.] The main cause of suffering and entanglement is attachment, or desire, which in its turn is based on misconception. In order to put an end to misconception and attachment, the early Bud[p.53]dhist traditions stress the impermanence, unsatisfactoriness and not-Self-ness (or lack of Self) of all phenomena. In many Mahāyāna-sūtras, the things which we believe we experience are more radically characterized as illusory or unreal, or as 97 Cf. the reduction of the cittaviprayukta-saṁskāras to mere denominations (Abhidharmasamuccaya [see fn. 14] 10,15–11,24), or of many traditional mind-associates to specific forms or combinations of a few basic ones (ibid. 8,13–9,19; 10,5-12).

98 This is rightly emphasized by LUSTHAUS (L 486–487). On this problem, see YAMABE 1998 (s. fn. 1): 15–41, and Yūichi KAJIYAMA, “Do Other People's Minds Exist?”, in: Annual Report of the International Research Institute for Advanced Buddhology at Soka University 3 (1999): 3–35 (in Japanese; dealing especially with Vasubandhu’s Viṁ- śatikā, Dharmakīrti’s Santānāntarasiddhi and Ratnakīrti’s Santānāntaradūṣaṇa); see also Y. KAJIYAMA, “Buddhist Solipsism. A free translation of Ratnakīrti's Saṁtānāntaradūṣaṇa”, in: Y. KAJIYAMA, Studies in Buddhist Philosophy (see fn. 84): 401–416. 99 I.e. experienced as not being created at will by one's own perceptions. The Spiritual Context of vijñaptimātratā 53

mere fictions, or as nothing but mind (cittamātra). LUSTHAUS (L 534) is right in pointing out that like these predicates, the Yogācāra term “nothing but consciousness” (vijñaptimātra) is, primarily, depreciating. In many early Yogācāra texts, realizing in meditative concentration100 that the objects we experience are nothing but mind or consciousness is only a preliminary step on the way to the awareness of the highest truth, or ‘true reality’ (tathatā), attained on the supra-mundane Path of Vision (darśanamārga).101 Even so, the level of ‘nothing but consciousness’ (vijñaptimātra) is superior to that of ordinary experience

or of conservative Abhidharma, and therefore appropriate for the construction of a revised, higher Abhidharmic system. As is rightly stressed by LUSTHAUS, one should not become attached to consciousness as an entity either, and the CWSL itself — in this point, too, basically in agreement with the Viṁśatikā (6,13-22 and 10,28–11,5) — explicitly situates the elaborate system of the eight vijñānas on the level of “provisional truth [based on, or clarified by] arguments” (理世俗 = yukti-saṁvṛti).102 On the ultimate [p.54]level of the highest truth (真勝義), thought and language are cut off (心言絕).103 This is, no doubt, the level of tathatā, not as it may be conceptualized, but as it is experienced 100 E.g. Mahāyānasūtrālaṁkāra (ed. S. LÉVI, Paris 1907) XIV.23– 24 (samāhitaḥ!). Cf., in this connection, also YAMABE 1998 (s. fn. 1): 35–38, suggesting spiritual practice as the basic motif behind the Yogācāra system. This was also the result of my own previous investigation of the matter in my articles “Spirituelle Praxis …” and “On the Problem of the Relation …” (see fn. 6).

101 E.g. Mahāyānasūtrālaṁkāra (see fn. 100) VI.6–8; XIV.23–28; Dharmadharmatāvibhāga (ed. Klaus-Dieter MATHES, Unterscheidung der Gegebenheiten von ihrem Wesen, Swisttal-Odendorf 1996): 64,98- 100 and 66,140-147; (Vṛtti:) 84,416-427 = 102,82-103,94 (Sanskrit) and 91,582-92,608; Mahāyānasaṁgraha (ed. E. LAMOTTE ) II.14b.c and f = Abhidharmasamuccaya-bhāṣya (ed. N. TATIA ) 42, 8-9 and 14-15. Cf. L 539, quoting Madhyāntavibhāgabhāṣya and Trisvabhāvanirdeśa. 102 CWSL 38c8-9 (P 414; C 231; S 314); cf. also 37a9. 103 CWSL 38c9; cf. 6c11, and also 6c23-26, 37a9-11 and 55b11. 54 On the Problem of the External World in the CWSL

in the transconceptual insight (nirvikalpaṁ jñānam). I do not see any indication for assuming that in this insight there is any direct experience of independent matter, or any multiplicity at all. Sthiramati's quotation from the Nirvikalpapraveśadhāraṇī according to which in transconceptual insight one experiences all dharmas to be like the surface of the empty sky104 would rather point to the opposite. If the vijñaptimātra experience and the Abhidharmic system built on its level is indeed the decisive intermediate step leading to such a radical dissolution of any experience of multiplicity in what one may justly call a mystical experience, the assumption of an independent material world behind the curtain, so to speak, would even appear soterically counter-productive.

[6.] The level of vijñaptimātra is, however, not only a preliminary step to be surmounted, but also a plane to which Buddhas and Bodhisattvas return or re-descend, for the sake of other sentient beings. After the first transconceptual insight of a Bodhisattva, this return is probably more like a fall back into an experience of the re-emerging world of multiplicity. But in this socalled ‘subsequent insight’ (tatpṛṣṭhalabdhaṁ jñānam) the dependent world of multiplicity is no longer misconceived, but is adequately experienced as [p.55]illusory or as nothing but mind (and mind-associates).105 The ‘subsequent insight’ is, how- 104 TrBh 40,29–41,1: nirvikalpena jñānenâkāśasamatalān sarvadharmān paśyati. The reading °samatalān (LÉVI: °samatāyāṁ; ms. °samatayā tāṁ) is taken from Kamalaśīla's 3rd Bhāvanākrama (ed. G. TUCCI, Rome 1971, p. 11,7-8) and the Nirvikalpapraveśadhāraṇī (ed. K. MATSUDA), in: Bulletin of the Research Institute of Bukkyo University 3/1996: 96,7 and 109, fol. 2a8). My rendering of the compound follows the Tibetan (nam mkha'i dkyil dang mtshungs par); the Sanskrit compound may mean “forming a [blank] surface like the empty sky”. Cf. also TrBh 40,29: pariniṣpannaś câkāśavad ekarasaḥ (with ms.), jñānaṁ ca.

105 CWSL 46b28–c6; TrBh 40,26-28 (quotation from the Nirvikalpapraveśadhāraṇī): tatpṛṣṭhalabdhena jñānena māyā-marīci-svapna-pratiśrutkôdakacandra- nirmita-samān sarvadharm ān pratyeti ; Sthiramati, The Spiritual Context of vijñaptimātratā

ever, also the mode of consciousness of a Buddha when he turns towards the world of multiplicity. Now, for the position preferred in the CWSL, the ‘subsequent insight’ as well as a Buddha's analogous knowledge of multiplicity is, epistemologically, essentially structured in the same way as ordinary, mundane consciousnesses.106 To be sure, both the CWSL and the *Buddhabhūmy- upadeśa mention a view according to which ‘subsequent insight’, at least in the case of a Buddha, is able to cognize its objects directly, without the mediation of an image.107 But this view is discarded, among other things, with the epistemological argument that if the appearance of a thing does not emerge in a consciousness this thing simply cannot be the object of that consciousness. 108 Hence, according to the view given preference in the CWSL, vijñaptimātratā as an epistemological principle excluding the direct cognizing of anything outside a given moment of consciousness is valid even in the case of a Buddha. In

this context, vijñaptimātratā does not seem to have the negative connotation of a level to be surmounted or of a limitation to be transcended. It rather appears as the essential structure of any experience whatsoever in the sphere of multiplicity, i.e. saṁvṛti (俗)109, as far as it is describable at all. This is all the more so in view of the fact that the images manifested by a Buddha's knowledge are not limited to reproduc[p.56]tions of the ordinary, impure world but include glorious, pure Buddha fields and pure bodies.110 Nothing suggests that these fields and bodies are considered to have an independent material existence111 any comm. on the Mahāyānasūtrālaṅkāra, Tanjur (Peking) vol. Mi 306a 5- 6: dag pa 'jig rten pa'i ye shes kyis ni khams gsum pa'i 'dus byas thams cad sems dang sems las byung ba tsam du zad par mthong ngo //. 106 CWSL 50b20; BBhU 317b17-18. 107 CWSL 50b18-19; BBhU 317b19-21. 108 BBhU 317b28-29: 若諸境相 非心上現, 彼雖有力 生心心所, 如 五根等 不名所緣｡

109 Cf. CWSL 56c10 and c16.

110 CWSL 58c1-3 and 8-10.

111 Cf. also appendix § 1.

56 On the Problem of the External World in the CWSL more than the ordinary, sullied surrounding world and bodies do. Especially in the case of the surrounding world, it is precisely on account of its being only an image in mind that its transformation from a impure world into a pure, sublime one through individual spiritual practise becomes plausible. This would be much more difficult with an adamant, unwieldy material world independent of mind. [p.57]

IV. Appendix:

Discussion of four CWSL passages

By way of an appendix, I shall discuss four more passages from the CWSL. Two of them are adduced by LUSTHAUS (L 512) in favour of the independent existence of matter (rūpa), whereas the others were indicated as possible counter-evidence to my position during the symposium at which this paper was first presented. [1.] The first passage is part of the characterization of the mirror- like knowledge (ādarśajñāna) of the Buddha:

大圓鏡智相應心品, ... 能現能生 身土智影, ... 如大圓鏡 現眾色像｡ (CWSL 56a12-16 / P 681; C 347; S 452.) LUSTHAUS translates:

“The Great Perfect Mirror Cognition, … associated with the mind (saṁprayukta citta-varga [sic!]) … […is] able to project/perceive and able to produce lived-bodies and perceptual-fields, knowing their reflections (ying 影, pratibimba) … like a great mirror projecting/perceiving rūpas and pratibimbas (material things and their reflections, se hsiang 色像).”

But 智 is not normally used as a verb. Actually, 智影 is a term taken from the Buddhabhūmisūtra, corresponding to Skt. jñānapratibimba (cf. Mahāyānasūtrālaṁkāra IX.69d112 and the Tibetan 112 “And because of the arising of ‘knowledge-images’, it is [called] the [[[mirror-like knowledge]]]” (jñānapratibimbodayāc ca tat; cf. Bhāṣya: jñānapratibimbodayāc ca tad ‘ādarśajñānam’ ity ucyate). 58 On the Problem of the External World in the CWSL [p.58]rendering ye shes kyi gzugs brnyan113). On the one hand, this expression may simply be intended to characterize the images appearing in the mirror-like knowledge as (adequate) mental reflections in contrast to the images reflected in a physical mirror. On the other, it is also understood as “images consisting in [the other three Buddha-]knowledge[s]”.114

For the interpretation of the somewhat ambiguous expression 能現能生身土智影, it would seem advisable to follow a passage in the *Buddhabhūmy-upadeśa where precisely the same expression (BBhU 311a27-28) is explained as follows:115 “When it encounters the [proper] ‘external’ conditions, just then [this pure consciousness (淨識: 311a27) associated with the mirror-like knowledge] develops [in such a way as to] manifest manifold images of bodies and [[[Buddha]]-]fields as [its] object (境界, viṣaya),116 and produces the groups of mind [and mind-associates] associated with ‘equality-knowledge’ (samatā-jñāna) and the other [[[knowledges]] of a Buddha as] different modes [of mental activity] (行相, ākāra)117” (BBhU 311a27–b1: 若遇 113 Kyoo NISHIO (ed.), The Buddhabhūmi-sūtra and the Buddhabhūmi- vyākhyāna of Çīlabhadra, [reprint] Tokyo: Kokusho Kankokai 1982, p. 9,10, etc.

114 Cf. Sthiramati, comm. on the Mahāyānasūtrālaṅkāra, Tanjur (Peking) vol. Mi: 157a3-5. 115 One of K'uei-chi's explanations (viz. Shu-chi 598c28–599a1) is similar.

116 Cf. also BBhU 302c24-25: 現身土等一切影像, 312c23: 能變現身 and CWSL 58c1-3: 圓鏡智相應淨識 … 變為純淨佛土, showing that 現 X, 變現 X and 變為 X are interchangeable. 117 This refers to the Buddha's own continuum (cf. also BBhU 312c21-22). Moreover, as adhipati-pratyaya, the mind-cluster associated with the ‘mirror-like knowledge’ generates ‘images’ consisting in knowledge, etc., (智等影) in the continua of other sentient beings (BBhU 312c22-24). Cf. also BBhU 317c18-21: the pratyavekṣā-jñāna of a Buddha, either by means of arousing an image of the teaching of the Dharma (教法影像) in the mind associated with his ‘mirror-like Appendix: Discussion of Four CWSL Passages 59 外緣, 即便變現 身土境[p.59]界 種種影像, 及能生起 平 等智等相應心品 行相差別｡)118

As for the expression 色像, there is no reason to take it as a dvandva. In connection with the verb 現, it is much more natural to understand it as a tatpuruṣa in the sense of “images of visible things”.119 Moreover, this part of the definition is obviously merely a comparison, 120 comparing the images manifesting themselves in the mirror-like knowledge with the appearance of images of visible things in a mirror, and hence anyway inconclusive for the doctrinal question of matter in the system of the CWSL.

My translation of the passage (not substantially differing from C 347) would be as follows: “The group of mind [and mind-associates] that is associated with the ‘great perfect mirror[-like] knowledge’ … knowledge’ or just through its own manifestion of such an image, is the condition for sufficiently mature sentient beings to develop a similar image in their own mind (自心變現). For this issue, cf. also Paul DEMIEVILLE, "Les versions chinoises du Milindapañha", in: Bulletin de lʼ École Française dʼ Extrême-Orient 24 (1924): 52‒57. 118 Slightly different translation in John KEENAN (transl.), The Interpretation of the Buddha Land by Bandhuprabha, Berkeley: Numata Center for Buddhist Translation and Research 2002 (BDK English Tripiṭaka 46-II): 134.

119 Cf. BBhU 309c2-3, stating that in the ‘mirror-like knowledge’, etc., the images of all objects to be known manifest themselves clearly. As is evident from the context, these objects include visible matter, etc., as well as the material sense-faculties (BBhU 309a27-29). Cf. also Buddhabhūmi-vyākhyāna (see fn. 113) 82,27-29, according to which these objects become manifest in the ‘mirror-like knowledge’ insofar as, taking it as their basis, they arise in the form of mere images (pratibimba) (me long lta bu'i ye shes ... de la brten nas gzugs brnyan tsam gyi tshul du 'byung ba de la snang bar 'gyur ba'i phyir ro). 120 This interpretation, supported by SAEKI (S 452) and COOK (C 347), is no doubt preferable to that of DE LA VALLÉE POUSSIN (P 682), but even the latter takes 色像as a tatpuruṣa (“les images de tous les Rūpa”).

60 On the Problem of the External World in the CWSL mani[p.60]fests [images of]121 bodies and fields122 (or: a body and a field?)123 and generates images [consisting in other forms] of knowledge, … , just as a great round mirror manifests a host of images of visible things.” [2.] In the second passage adduced by LUSTHAUS, Hsüantsang states that – in contrast to merely nominally existent (假有 = prajñaptisat) things like combinations and continuities – mind, mind-associates and matter are taught to be really existent (實有 = dravyasat) because they arise from conditions (CWSL 47c10-11: 心心所色 從緣生故 說為實有｡ / P 554; C 293–294; S 380).

Since the question of really existing matter (rūpa) is discussed in detail in the beginning of the CWSL (4a8–5a6), it is, in a fairly systematic work like the CWSL, methodically inadmissible to take the present passage in isolation. Rather, one has to understand it against the background of the more explicit statements of the detailed discussion. There we read that matter outside or apart from consciousness does not really exist. But this does not mean that matter does not exist at all: on the contrary, it does, though only as an image of matter into which consciousness develops. 124 For the position which is promi[p.61]nent in the 121 影 may belong to both sets of objects.

122 Cf. BBhU 317a3-9, stating that the other buddha-jñānas function as an adhipati-pratyaya which impels the consciousness associated with the ‘mirror-like knowledge’ (cf. also BBhU 317c18-19 (see fn. 117)) to develop (現, 現變) into [an image of] a saṁbhoga- or nirmāṇa- kāya. As far as I can see, the CWSL (57a1-4; 57c28–58a4) relates these bodies to the other jñānas only. 123 I.e. if the text refers, in particular, to the peculiar material body and Buddha-field of the svasaṁbhoga-kāya level (CWSL 57a1-2; c26-27 and 58c1-4).

124 Cf., e.g., CWSL 4a27: “Although it is not the case that matter does not exist, it is after all [only] a development of mind (vijñānapariṇāma)” (雖非無色 而是識變); 4b1-2: “Since resistent matter [[[existing]]] outside is unproved (asiddha, or: logically impossible, ayukta), it can therefore only be [something] developed and manifested by Appendix: Discussion of Four CWSL Passages 61 CWSL, this image, having arisen from conditions, is as real as consciousness itself.125 Hence, the passage under discussion, too, has to be understood in this sense. It uses matter as a short-hand for the image of matter in consciousness. It cannot be used as evidence for a real existence of matter entirely independent of all forms of consciousness – an idea of which there is no trace in the more explicit discussion.

[3.] CWSL 11a1-2 (P 137; C 65; S 83): “Therefore, when the surrounding world is going to be destroyed and when it has just started to re-arise, although there are no living beings [in it], it is yet actually existent.” (故器世界 將壞初成, 雖無有情, 而亦現有｡) This refers to the state immediately before the periodical dissolution (saṁvarta) of the lower spheres of the world when the sentient beings inhabiting them have, due to their favourable karma, already moved on to the higher realms (cf. YoBhū 34,9 ff, esp. 35,5-6), and to the period when the favourable karma of sentient beings in the higher realms is on the point of becoming exhausted and the lower spheres re-arise but no sentient being has yet descended into them (cf. ibid. 36,19 ff, esp. 41,17-18). If, as in the preceding (= second) theory, only the vipākavijñānas of those sentient beings who actually dwell in a certain sphere of the surrounding world or are bound to be reborn there would develop an image of that sphere (CWSL 10c21-22), one might ask whose vipākavijñāna develops an image of, e.g., the kāmadhātu before its dissolution when no sentient beings who actually inhabit it are left or are bound to be reborn there (CWSL 10c23-24). The solution proposed by the third theory is that the vipākavijñāna of a sentient being develops only [an image of] [p.62]that level of the surrounding world which this being actually inhabits, but does not confine itself to the world where this consciousness within” (外有對色 理既不成, 故應但是 內識變現); 5a6 (see § 3.2).

125 CWSL 59a8: 識相見 等從緣生 俱依他起 虛實如識. See also § 4 of this appendix.

62 On the Problem of the External World in the CWSL being actually lives; rather, it includes the corresponding level of other world systems as well. The rule is that “if [a sphere or level of the surrounding world] could have the function of support with regard to the body [of a certain sentient being], then [the vipākavijñāna of this being] develops into [an image of] that [[[sphere]] or level]. Therefore, even if [a sentient being] is born at one's own level in another region (i.e. world system), its [[[vipāka]]-]- vijñāna still comes to develop into [an image of] this earth (i.e. one's own level of this world).” (CWSL 10c28–11a1: 故若於身可有持用, 便變為彼｡ 由是設生他

方自地, 彼識亦得變為此土｡ / P 137; C 65; S 82–83) Since different world systems do not dissolve and re-arise at the same time, the kāmadhātu of our world exists as an image in the vipākavijñānas of sentient beings of other world systems even when it is already, or still, empty of sentient beings. Both from the point of view of the context and from the point of view of terminology (識 ... 變: vijñānapariṇāma) it is evident that what the text wants to make certain is that even during periods when no sentient beings are present the surrounding world is still there because it continues to exist as an image in the minds of other sentient beings. No such somewhat far-fetched theory would be required if an independent ‘real’ existence of the world were presupposed.

[4.] At the end of the CWSL, Hsüan-tsang resumes the issue of what the principle of vijñaptimātratā actually implies. Matter (rūpa) is not explicitly mentioned, but the problem of its status is involved. The first problem discussed is merely whether °mātra excludes even the ‘parts’ (*bhāga or *aṁśa) distinguished within one and the same moment of consciousness, especially the ‘image part’ (相分). The first opinion (59a5-7) states that it does and that the ‘parts’ [p.63]are merely imagined (which would seem to imply that forming images, or the opposition of object and subject, is invariably connected with ‘externalization’). The second opinion, however, which corresponds

Appendix: Discussion of Four CWSL Passages 63 to the position preferred by the CWSL, takes the ‘parts’ of consciousness to be included and to be on the same level of reality as consciousness itself, both being paratantra, not parikalpita (59a7-8). Hence °mātra merely excludes the external object but not the internal one (i.e., the ‘image part’) (59a8-9: 唯言遣外, 不 遮內境｡). To this, an opponent (perhaps the representative of the first opinion) raises an objection:126 “Since the internal (!) object and consciousness are both not false (i.e. on an equal level of reality), why do [the texts, in order to indicate their non-separateness,] merely speak of ‘nothing but consciousness’, [and] not of ‘[nothing but] object’127?” (CWSL 59a9-10: 內境與識 既 並非虛, 如何但言唯識非境｡ / P 718; C 369; S 470.)

[p.64]The first argument of the answer (59a10-11) points out that ‘object’ would be ambiguous because it might be misunderstood to include also the external object (here obviously in an ontological sense of extra-mental entities). The second argument (59a11-14) suggests a spiritual motive for the preference of 126 As already indicated by DE LA VALLÉE POUSSIN (P 718; cf. also T vol. 67, no. 2266: 915b2-6), this objection resembles an argument advanced, likewise from the point of view of a position according to which the ‘image-part’ in consciousness is unreal, at BBhU 317c6-9 (cf. KEENAN, op. cit. [see fn. 118]: 169, differring, however, considerably from my translation of the passage): “Although the image [of an object] developed by the mind looks as if it were existent, in reality it has no substance. Otherwise, visible matter, etc., would exist just like mind and mental factors, [and thus your position would] not come up to [the principle of] ‘nothing but consciousness’. If [you think that one can] speak of ‘nothing but consciousness’ because the [[[object]]] is really existent but not separate from consciousness, one should as well speak of ‘nothing but object’ because mind and mental factors, [in their turn,] are not separate from those images of visible matter, etc., either. [This, however,] would amount to a serious flaw [of your position].” (心所變相 雖相似有 而實無體｡ 若不爾者, 應有色 等 如心心法 不成唯識｡ 若彼實有 但不離識 名唯識者, 心及心法 亦不 離彼色等諸相 應名唯境, 便成大過｡)

127 Cf. Shu-chi 606a11; 27; 28 and BBhU 317c9 (see fn. 126): 唯境. 64 On the Problem of the External World in the CWSL ‘nothing but consciousness’ but stresses that it excludes only the utterly non-existent (都無) external object (外[境]) but not the internal one. Finally, the last argument (59a14-15) is unambiguous in explaining the preference for ‘nothing but consciousness’ by the fact that the ‘image part’ (相分, i.e. the internal object, which is not excluded), etc. (i.e. the other ‘parts’ of consciousness), all have consciousness as their nature (皆識為性) [whereas the ‘object-nature’ would be confined to the image part]. The text supplements this argument by showing that the tathatā and the mind-associates are also not excluded by the principle of vijñaptimātratā.

Thus, in this passage, too, there is neither explicit reference to nor any room for an independent existence of matter. Matter can be taken to be included by the principle of vijñaptimātratā only in the form of mental images of matter. Any form of matter which is not the image of any form of mind or mind-associates would inevitably be ‘external’ (in the strict use of the word) and would hence be excluded by the word °mātra. The assumption of matter that is neither internal nor external would be entirely arbitrary since no such possibility is mentioned in the text. P.S. A couple of corrections and additions as well as the pagination of the printed version have been added in red colour by the author.

January 2015

Abbreviations

AKBh = Vasubandhu, Abhidharmakośabhāṣya, ed. P. PRADHAN, Patna 1967.

BBhU = Buddhabhūmy-upadeśa: T vol. 26, no.1530. C = Francis H. COOK, Three Texts on Consciousness Only by Hsüantsang. BDK English Tripiṭaka 60-I, II, III. Berkeley: Numata Center for Buddhist Translation and Research 1999. CWSL = Ch’eng wei shih lun (成唯識論): T vol. 31, no. 1585. HôbFA = Hôbôgirin, Fasc. annexe, Paris & Tokyo 21978. I-yen = Ju-li (如理), Ch’eng wei shih lun shu i yen (成唯識論疏義 演): 大日本續藏經 1.79.1–5.

L = Dan LUSTHAUS, Buddhist Phenomenology (see fn. 7). P = L. DE LA VALLEE POUSSIN, Vijñaptimātratāsiddhi. La Siddhi de Hiuan-tsang. Paris 1928–29. S = SAEKI Jōin (ed.), Jōyuishikiron (成唯識論), 5th ed. Nara: Hōryūji 1984 (I am very grateful to Prof. N. Yamabe for a copy of this standard edition).

Shu-ch’ao = Ling-t’ai (靈泰), Ch’eng wei shih lun shu ch’ao (成唯識 論疏抄): 大日本續藏經 1.80.2–5.

Shu-chi = K’uei-chi (窺基), Ch’eng wei shih lun shu chi (成唯識論 述記): T vol. 43, no. 1830.

SNSṬ = Wŏn-ch'ŭk (圓測, Ch. Yüan-ts'ê, Tib. Ven tshig), Saṁ- dhinirmocanasūtra-ṭīkā: Peking-Tanjur vols. Ti, Thi and Di. T = Taishō Shinshū Daiz ōkyō (The Tripiṭaka in Chinese) , ed. by J. TAKAKUSU and K. WATANABE, 100 vols., Tokyo 1924 ff. TrBh = Sthiramati, Triṁśikābhāṣya, in: Vijñaptimātratāsiddhi, ed. Sylvain LÉVI , Paris 1925.

Yen-pi = Chih-chou (智周), Ch’eng wei shih lun yen pi (成唯識論 演祕): T vol. 43, no. 1833. YoBhū = Yogācārabhūmi, ed. V. BHATTACHARYA, Univ. of Calcutta 1957.

66 On the Problem of the External World in the CWSL Chinese texts have mostly been copied from the CBETA Chinese Electronic Tripitaka Collection (April 2004, obtained through the kindness of the Ven. Anālayo), but punctuation is mine.