Vasubandhu the Unified

Eli Franco

Representing various Buddhist traditions, the doctrinal discrepancies among the many works attributed to Vasubandhu have been subject to different interpretations. I can think of at least six, not mutually exclusive.

(1) Proponents of the perhaps most prevalent stance consider the traditional accounts by Parama¯rtha, etc., despite their obviously hagiographical nature, as historically reliable. Following Parama¯rtha, the disparity between Vasubandhu’s works is understood to be due to his “conversion” from Sarva¯stiva¯da to Sautra¯ntika, and more importantly from Sautra¯ntika to Yoga¯ca¯ra. (2) The skeptical stance is best illustrated by Lamotte in the introduction to his edition and translation of the Karmasiddhiprakaraṇa (p. 179). Lamotte calls Vasubandhu a “personage e ´nigmatique” and continues: “Ses biographes, depuis Parama¯rtha jusqu’aux auteurs des me ´langes Lanman, exigent trop de notre cre ´dulite ´ : qui donc croira que Vasubandhu, sans parler de ses accointances avec le Sa¯m ˙ khya, fut Vaibha¯s ˙ ika dans sa jeunesse, Sautra¯ntika dans son a ˆge mu ˆr, Vijn ˜a¯nava¯din dans sa vieillesse et Amidiste a ` sa mort ? A ` notre avis, Vasubandhu a trop ve ´cu, trop pense ´, trop e ´crit; et avant de se prononcer sur sa & Eli Franco franco@uni-leipzig.de

1 Institute of Indology and Central Asian Studies, Leipzig University, Leipzig, Germany

J Indian Philos DOI 10.1007/s10781-017-9330-6

personnalite ´, il faudrait avoir lu, critique ´ et compare ´ toutes ses œuvres. Nous sommes loin de pouvoir le faire.”1 (3) A somewhat related stance is to consider Vasubandhu a great commentator and systematizer, a kind of Buddhist sarvatantrasvatantra, and to refrain from attributing to him the different doctrines propounded in his works. One encounters this position in various forms here and there, but it is not associated, to my knowledge, with any specific scholar. (4) Frauwallner hypothesized the existence of two Vasubandhus, a theory that was supported and modified by Schmithausen. A few words on this hypothesis may prove useful because Gold misunderstands it. In his On the Date of the Buddhist Master of Law (published in 1951, p. 56), Frauwallner states that he is unable to

decide whether the Vimśatikā and Triṃśikā are to be attributed to the Older or Younger Vasubandhu. However, in Die Philosophie des Buddhismus (1st edition 1956) he says (p. 351), somewhat tentatively, that in his opinion, the two works are to be attributed to the Younger Vasubandhu.2 Thus, Schmithausen’s thesis on the Sautra¯ntika presuppositions in these two works was not meant “to counter Frauwallner’s thesis” (Gold, p. 7). In the additional notes to the third edition (1969, p. 425), Frauwallner refers to Schmithausen’s paper published in 19673 as confirming his attribution of the two works to the Younger Vasubandhu. In 1992,

Schmithausen published “A Note on Vasubandhu and the Lan ˙ka¯vata¯rasu ¯tra” (AS/E ´A XLVI, pp. 392–397), where he pointed out that the Triṃśikā is quoted in the Laṅkāvatārasūtra (in the part already translated into Chinese in 443). This was a serious blow to Frauwallner’s theory, for it made the two Vasubandhus, if two they ever were, almost contemporaries. However, in spite of various acknowledged problems, Schmithausen has not given up the assumption of the existence of two Vasubandhus, not only out of deference to Frauwallner, but also because it is essential to his theory of Sautra¯ntika presuppositions in the Viṃśatikāand Triṃśikā. According to Schmithausen, the brother of Asan ˙ga was a full-fledged Yoga¯ca¯ra, whereas the Younger Vasubandhu only gradually gravitated towards the Yoga¯ca¯ra (Kośa, Karmasiddhi, Viṃśatikā,4 Triṃśikā) in

1 Lamotte (1936): “His biographers, since Parama¯rtha and up to the authors of the Lanman felicitation volume, demand too much of our credulity: Who would believe that Vasubandhu, not to mention his acquaintances with the Sa¯m ˙ khya, was Vaibha¯s ˙ ika in his youth, Sautra¯ntika in his ripe age, Vijn ˜a¯navadin in his old age, and Amidist (follower of Amida Buddha) before his death? In our opinion, Vasubandhu lived too much, thought too much, wrote too much; and before one gives an opinion on his personality, one would have to have read, criticized and compared all his works. We are far from being able to do that.” 2 This statement is a bit puzzling in its context because these two works are the only ones that appear under the heading of The Older Vasubandhu (pp. 350ff.). I am not sure

whether this indicates that at some earlier stage Frauwallner held the opposite opinion. 3 Schmithausen (1967). 4 Reading Schmithausen’s 1967 paper, one certainly gets the impression that he argues that in the Viṃśatikā, Vasubandhu did not accept the existence of the ālayavijñāna and the multilayered series of cognition (see especially pp. 129–130). However, when I pointed out that in this case one would have to assume that Vasubandhu accepted the ālayavijñāna and multilayered series in the Karmasiddhi, rejected them in the Viṃśatikā and reaccepted them in the Triṃśikā, Schmithausen kindly informed me that he did not mean to say that Vasubandhu did not accept them, but only that he did not use them. Under this

a processthat never came toan end, foreven inhis lastwork,theTriṃśikā,certain Sautra¯ntika presuppositions are still apparent.5 (5) After a close comparison of the AKBh with the Yoga¯ca¯rabhu ¯mi, Robert Kritzer came up with the hypothesis that Vasubandhu was already a covert Yoga¯ca¯ra when he wrote the AKBh: “Vasubandhu was inserting Yoga¯ca¯ra ideas into the AKBh under the guise of the Sautra¯ntika.”6 Several other scholars had already pointed out that Vasubandu was on his way to accepting Maha¯ya¯na when writing the AKBh (Jaini, Hakamaya, Yamabe, Miyashita, Harada, summarized by Kritzer, p. xxvii). (6) Finally, there is Gold’s new interpretation, which I would like to consider here in some detail.

Unlike all previous approaches, Gold argues that Vasubandhu’s works—and he takes into consideration not only his Abhidharma and Yoga¯ca¯ra works, but also the Vyākhyāyukti and the fragments from the Vādavidhi—make a meaningful and coherent whole with a distinctive purpose. In his own words (pp. 4–5): “I will argue throughout this book, that Vasubandhu repeatedly displays a preference for certain specific patterns of argument, certain methodologies in the application of reason and scripture, and remarkably—given the diversity of the topics covered—certain epistemological and ontological claims. These patterns within Vasubandhu’s works, I will argue, allow us a glimpse into what drives him as a thinker. Thus, whereas it is difficult to pin down Vasubandhu when we look at him from a doctrinal perspective, he is nonetheless readily identified by the light of his distinctive arguments. … The present study emphasizes the evident continuity of interest and purpose across diverse works attributed to Vasubandhu, and paints a picture of a great thinker’s central concerns and philosophical trajectory. It renews our understanding of Vasubandhu as one of the most important Buddhist philosophers of all time.”7

Footnote 4 continued interpretation, it would be difficult to maintain that a real development has occurred in Vasubandhu thought at the time that separates “the system of the Viṃśatikā” from “the system of the Triṃśikā.”

There is evidence to suggest that when writing his 1967 paper, Schmithausen maintained that the ālayavijñāna is not accepted in the Viṃśatikā; he later changed his opinion on this point. 5 To add my two cents worth to the discussion, I find it odd that when discussing the conglomerations of atoms, the author of the Viṃśatikā does not take the position of the Kośa into consideration, at least not in its specificity, namely, that atoms are not in contact with each other, but also do not have any intervals between them (see AKBh p. 33.2–3 on 1.43). From this, admittedly narrow perspective, it looks as if the Kos ´aka¯ra responds to

the argument of the Viṃśatikā. This observation adds to von Rospatt’s comment (The Buddhist Doctrine of Momentariness: A Survey of the Origins and Early Phase of this Doctrine up to Vasubandhu. Stuttgart 1995: 187, n. 408) that the author of the Mahāyānasūtrālaṅkārabhāṣya does not seem to have known the AKBh. However, unlike von Rospatt’s observation, which was put forward in support of Schmithausen’s hypothesis, the relationship between the Kośa and the Viṃśatikā seems to

indicate that the latter has to be attributed to The Older Vasubandhu, or at least that it predates the former. For the most recent defence of the two Vasubanshus theory cf. Buescher (2013). Note also that according to the Tibetan tradition as reported by Bu ston (Obermiller’s translation p. 144), the AK and AKBh were composed after Vasubandhu had embraced the Maha¯ya¯na doctrines. Perhaps this tradition should not be discarded off hand, as has been done by modern scholarship. 6 Kritzer (2005, p. xxviii). 7 In his enthusiasm for Vasubandhu, Gold calls him (p. 218) “the greatest mainstream Buddhist philosopher after the Buddha himself.”

This sounds like an exciting new thesis and one can understand the enthusiasm of the reviewers quoted on the back cover of the book, who go well beyond the usual compliments.“ThisbookwillforeverchangethewaywereadVasubandhu”,saysone. “A model for how to treat the work of a classical Buddhist thinker”, says another. “A rare example of a sustained and subtle engagement with the whole career of one of history’s greatest Buddhist philosophers” says a third. “A masterpiece of

philosophical exposition, synthesis, and creative commentary”, says a fourth. However, God and the devil are in the details, and reading the book carefully is bound to produce a sobering effect. On the whole, Gold’s work contains numerous gross misrepresentations of Vasubandhu’s thought and arguments. There are problems with the Sanskrit vocabulary and grammar and gaps in the knowledge of the basics of Indian philosophy. The many mistranslations and misinterpretations suggest that although he based his book mainly on the Abhidharmakośa, Gold is not sufficiently knowledgeable in Abhidharma scholastics. Further, his general thesis remains unproved, even if we take it in a weaker form. This may sound like a harsh judgment, so let me illustrate and substantiate the above claims in

some detail. First, a few examples that make it clear that Gold makes surprising mistakes when it comes to Sanskrit grammar on a very basic level of morphology and syntax. One encounters, for instance, confusions between the accusative and nominative cases, and between singular and plural personal endings (p. 70, translating AKBh 30.4): yadi cakṣuṣ paśyed8 anyavijñānasamaṅgino ’pi paśyet. na vai sarvaṃ cakṣuṣ paśyati. “If the eye sees, then so also the other sufficient conditions forconsciousness should see. Certainly not every eye sees.” (my emphasis, EF; see also below, p. 6) Some confusions in vocabulary have far-reaching

consequences. For instance, Gold translates (p. 89) tantu as “cloth” and paṭa as “thread” with the implication that the thread as a whole (avayavin) inheres in many cloths as its parts (avayava). One also encounters lack of familiarity with what one would think are relatively well-known and unproblematic terms. For instance, the gerund kṛtvā is sometimes used idiomatically in the sense of “having considered,” “having in mind” (short for manasi kṛtvā). This usage is clear in set phrases such as X kṛtvā,Yucyate (having X in mind, he [the Buddha] said Y). Gold, who seems to be unfamiliar with this usage, translates (pp. 40 and 95)

“insofar as this is the case”. To give another example, far-reaching analysis is conducted on the basis of a misunderstanding of the very common term asmadādi (jana). As is well known, the compound refers to common people, people like you and me, in contradistinction to yogis, persons of authority and other “religious virtuosos.” To take this expression personally is similar to the reaction, “Oh, I didn’t know the author knows us,” when being told that “the author has written his book for people like you and me.” Yet this is exactly what Gold does. When Vasubandhu says of Asan ˙ga vaktāraṃ cāsmadādibhyaḥ (Gold: -ādibho), Gold explains (p. 159): “He [[[Vasubandhu]]] calls Asan ˙ga ‘the vocalizer to me, etc:’ –meaning ‘to myself, among others.’ The commentary thus claims for itself an extremely short lineage: I heard the text, it says, from its initial speaker, …”. Perhaps Gold himself felt uncomfortable with this 8 Gold: paṣyed—this is not the only typo in this book.

interpretation and added in a note: “This is not to say that Vasubandhu was among the first to hear the text. The commentary allows the interpretation that its author was one among several people to whom Asan ˙ga spoke the text, successively, at different times.” The misunderstanding of common words and expressions inevitably leads to misinterpretation of philosophical arguments. Consider for instance AKBh p. 189.18f., where Vasubandhu employs a well-known argument: when seeing just the fringe of a piece of cloth, one would have to see the entire cloth; if not, one would never see it, “because the middle and farther parts (i.e., the parts within the cloth and those on the opposite side from the observer) are not in contact with the sense faculty” (madhyaparabhāgānām

indriyeṇāsaṃnikarṣāt). Gold translates nonsensically (p. 88): “For the sensory organ does not come into contact with an intermediary power,” thoughtlessly reproducing the translation of parabhāga in MW. Furthermore, one has to note some insensitivity to philosophical terminology. The translation of lakṣaṇa (mark, sign, characteristic) as “quality” when discussing doctrines of Abhidharma schools which reject the existence of substances and thus the distinction between qualities and substances is infelicitous, especially in the context of the theory of time of Ghos ˙ aka, who is dubbed by Gold “a qualitydifferentialist”.

Similarly, the repeated reference with substantives to the three times (past, present and future) in context of the Sarva¯stiva¯da doctrine is disturbing because according to the Sarva¯stiva¯da time and the three times do not exist as substantive entities. According to Sarva¯stiva¯da, there are past and future things (dharmas), but the past and the future, or time itself, are not considered to exist as such (they are not dharmas). Thus, e.g., p. 37: “If the past and the future exist substantially just like the present does” should read “if past and future [[[dharmas]]] exist substantially like the present [[[dharmas]]].” However, this

loose usage also appears sometimes in the Kośa itself. The lack of solid philological background in Sanskrit is compounded by the fact that Gold does not seem to have the necessary background in Abhidharma scholastics in general that are essential for reading the Abhidharmakośa. The following is a remarkable, but not untypical example of a translated passage where something went wrong in every single sentence. The passage occurs in the context of the question “what sees?”, i.e., what is the agent of the act of visual perception, the alternatives being the sense faculty or cognition itself. AKBh 30.12-18: naiva hy āvṛte

cakṣurvijñānam utpadyata ity anutpannaṃ kathaṃ drakṣyati? kiṃ khalu notpadyate? yasya tu cakṣuṣ paśyati tasya cakṣuṣaḥ sapratighatvād vyavahite vṛttyabhāva iti vijñānasyāpy anutpattir āśrayeṇaikaviṣayapravṛttatvāt yujyate. kiṃ nu vai cakṣuṣ prāptaviṣayaṃ kāyendriyavat yata āvṛtaṃ na paśyet sapratighatvāt? kācābhrapaṭalasphaṭikāmbubhiś cāntaritaṃ kathaṃ dṛśyate? tasmān na sapratighatvāc cakṣuṣa āvṛtasya rūpasyādarśanam. kiṃ tarhi? yatrālokasyāpratibandha āvṛte ‘pi rūpe tatropapadyata eva cakṣurvijñānam. yatra tu pratibandhas tatra notpadyata ity anutpannatvād āvṛtaṃ nekṣate. Gold (p. 72) translates as follows (the line breaks are Gold’s; the numbers at the end of the lines are mine and refer to the notes below):

“No, the eye consciousness does not arise with respect to something covered, so how will it see what does not arise? (1) How then does it not arise? Since the eye is subject to resistance, the state of seeing something does not come about with respect to what is covered. (2) So for the consciousness, too, it does not arise; it operates by means of a single sensory object, which joins it as support [[[āśraya]]]. (3) Why then would you say that the eye, like the bodily sense organ which meets its object, does not see it to the extent it is covered? (4) Because it is subject to resistance. (5) And how is something seen which has an

interposition by glass, fog, veil, or water? (6) That is not a case where, because it is subject to resistance, an eye fails to see a covered form. (7) What is it then? The eye consciousness does arise in the case where sight has no impediment even with respect to a covered form. But where there is an impediment, it does not arise. In that case, because it does not arise, the covered thing is not seen.” (8)

(1) Gold confuses nominative and accusative; rather than “how will it see what does not arise?” One should translate: “inasmuch as it has not arisen, how will it see?” (2) Gold fails to understand that yasya and tasya refer to a person: “for the one who professes that the ‘eye’ sees, the non-arising … is correct.” In addition, and throughout the book, Gold confuses the sense faculties, such as the senses of vision (cakṣus), hearing (śrotra), smell (ghrāṇa) and so on, with the parts of the body such as the eye (cakṣurgolaka), the ear (karṇa) and the nose (nāsā, etc.), where the sense faculties are located. For the Buddhists, as in Indian philosophy in general, the senses are distinct from the physical organs, which are only their support (āśraya, adhiṣṭhāna). They are made of subtle

invisible matter. Thus, when Vasubandhu says that the senses are invisible and have to be inferred, he obviously does not mean to say that the eye, the ear, the nose, the tongue and the “body” are invisible. Similarly, when he says, referring to the opponent’s theory (!) that the sense comes into contact with its object, he does not attribute to his opponent the absurd opinion that, say, the eye leaves its socket to touch the moon, or the ear its cavity in the skull, to touch the point where the stick hits the drum.9 (3) The translation does not convey the reason: consciousness does not arise because it operates (becomes active) on the same object as the sense faculty on

9 It is only in a single instance that Gold shows a dim understanding that “eye” would simply not do as a translation for cakṣur(indriya), but instead of investigating the matter further, he makes up his own original solution (p. 71): “the eye organ (made of a kind of subtle physicality, not just the eyeball).” Needless to say, the eyeball is not at all the sense of vision, but only the physical basis on which the latter rests. We will see a similar kind of ad hoc creativity when he discusses the theory of pākajaguṇa

which it rests. Gold also misinterprets yujyate, here “is correct,” “logically possible,” as “joins”. (4) Gold misunderstands the syntax and thus the question. As already noted above, he does not seem to know that cakṣus and so on do not refer to organs like “eye,” “ear,” “nose” etc., but to sense faculties or capacities. Here he also shows that he is unaware of the centuries-long discussion among Buddhists and Brahmin philosophers whether the senses of vision and hearing have to touch their object in order to operate on them (prāpyakāritva). The Buddhists, Vasubandhu included, deny this.10 Being unaware of all this, Gold does not understand the dismay of the proponent: kiṃ nu vai cakṣuḥ prāptaviṣayaṃ kāyendriyavat …? “Do you really mean that the sense of vision has an attained

object (i.e. that it really comes in contact with its object) like the tactile sense?” Clearly, prāptaviṣayaṃ cannot be construed with kāyendriyavat. (5) This is followed by another misunderstanding of the syntax. The words yata āvṛtaṃ na paśyet sapratighatvāt are part of the previous sentence (4): “Do you really mean that the sense of vision has an attained object due to which / by reason of which it would not see something covered because it is subject to resistance?” (6) Here Gold mistakes the proponent for an opponent. The proponent asks how the sense of vision is subject to resistance since it can see something separated by

glass and so on. Therefore, it is not because it is subject to resistance that the sense of vision does not perceive a covered object. (7) This statement continues the argument of the previous speaker, i.e., the proponent. Gold’s translation wrongly implies that the proponent holds that the sense of vision is

subject to resistance. (8) The word āloka (light) is misunderstood as “sight” and as a consequence the translation is beside the point. Rather, the point is this: Where light is not obstructed, visual consciousness is possible (here he also mixes upapadyate and utpadyate) even with respect to a covered object. In other words, we do not see a covered object not because the sense of vision is unable to penetrate the cover, but because the necessary light is unable to penetrate it.

The above misunderstandings are all the more surprising when one considers that de La Valle ´e Poussin translated the passage correctly, albeit not literally since he translated from the Chinese. Yas ´omitra’s commentary is also extremely helpful for understanding Vasubandhu’s text, but for some reason Gold does not utilize it. Obviously, to understand Vasubandhu’s arguments against his opponents, one has to have a basic familiarity with their doctrines. A typical example of Gold’s creativity is the way he deals with the Vais ´es ˙ ika theory of pākaja-guṇa. According to the Vais ´es ˙ ika, the eternal elementary atoms have eternal elementary qualities such as color, taste, touch and so on. However, there is an exception to this rule in

10 Oddly, Gold p. 266, n. 27 refers the reader to AKBh 1.43d (trayam anyathā) for the opinion that the senses come in contact with their object, not realizing that this passage speaks only of three senses (of smell, taste and touch), and overlooking 43c where Vasubandhu says that the senses of vision and hearing, and the mental sense do not come in contact with their objects: cakṣuḥśrotramano ’prāptaviṣayam.

the case of earthen atoms and composite earthen substances. When they are “baked,” i.e., heated, their color changes. According to later literature, this tenet was interpreted differently by Vais ´es ˙ ikas and Naiya¯yikas. The former claimed that the atoms change their color, which, because of certain assumptions about the formation of wholes, involved the inelegant assumption that a whole substance (avayavin) like a pot has to be broken down to its atoms by fire, which then change their color due to the heat and subsequently reconstitute as a new pot step by step. The Naiya¯yikas, on the other hand, preferred to assume that the whole changes its color due to the heat; therefore one need not assume that it disintegrates and then reintegrates as a new substance.11 Interestingly,

Vasubandhu’s opponent holds the so-called Nya¯ya position.12 He gives the example of a pot that changes its color when “baked” or fired in a kiln. Although the old color is destroyed and a new color “arises from the baking” (pākaja), one recognizes that it is the same pot. Vasubandhu responds with the usual argument that the recognition of a pot as being the same before and after its baking is due to its parts having the same arrangement as before. Gold, however, who seems not to have heard of this issue, explains (p. 91): “The last example, of the ‘cooked’ food in a pot, …”.13 Further problems concerning the theory of the whole (avayavin) are encountered in connection with AKBh 189.21f., which says that the parts cannot create a whole “because when the threads are of different color,

have different universals (e.g., “cottonhood” and “silkhood” when they are made of different materials, such as cotton and silk) and different actions (e.g., keeping a person warm to different degrees), the cloth could not have a [unified] color and so on” (bhinnarūpajātikriyeṣu tantuṣu pa ṭasya rūpādyasaṃbhavāt). Gold translates (p. 89): “The form of a thread does not appear in cloths that are manufactured [jātikriyā] out of different forms (colors), etc.” Gold has missed the point here and in his translation constructs strange new meanings that do not exist in the philosophical tradition under discussion. Thus, jāti, “universal,” and kriyā, “action,” become “manufactured”, “cloth” becomes “thread” and “threads” become “cloths.” As a result, one thread is assumed to reside in many cloths. A similar case is Gold’s understanding of the following argument raised by Vasubandhu against the Sa¯m ˙ khya (AKBh 159.18-22): kathaṃ ca sāṃkhyapariṇāmaḥ. avasthitasya dravyasya dharmāntaranivṛttau dharmāntaraprādurbhāva iti. kaś cātra doṣaḥ. sa eva hi dharmī na saṃvidyate yasyāvasthitasya dharmāṇāṃ pariṇāmaḥ kalpyeta. kaś caivam āha dharmebhyo ‘nyo dharmīti. tasyaiva tu dravyasyānyathībhāvamātraṃ pariṇāmaḥ. evam apy ayuktam.

11 These two tenets led to the well-known nicknames of the Vais ´es ˙ ikas and Naiya¯yikas, Pı¯lupa¯kava¯dins

and Pit ˙ harapa¯kava¯dins respectively. The pākaja doctrine is explained by practically all introductions to Nya¯ya-Vais ´es ˙ ika atomism; see for instance Bhaduri (1975, pp. 90–102). 12 That is, what is known as the Nya¯ya position in later literature. We are not sure whether Nya¯ya as a full-fledged independent philosophical tradition already existed in Vasubandhu’s time. Elsewhere (p. 96), Gold assumes that Vasubandhu knew Va¯tsya¯yana. In fact, even Digna¯ga makes no reference to Va¯tsya¯yana, who probably has to be dated in the second half of the fifth century. 13 See also the translation of pākaja in p. 85. There Gold misunderstands the argument because he is not aware of the Nya¯ya theory of variegated color (citrarūpa).

Gold translates (p. 30):

“And how is change for the Sa¯m ˙ khyas? Where, possessed by a stable substance, one dharma disappears and another dharma arises. And what is the fault here? That the dharma-possessor is not known which is stable, but whose dharmas are made to change. What about saying that the dharma-possessor is other than the dharmas? But the substance itself has a change, simply by becoming possessed of difference [anyathībhāva]. This is what is illogical.”

Gold is unfamiliar with the philosophical presuppositions of the Sa¯m ˙ khya opponent and as a result confuses opponent and proponent, thus missing the point of the argument. To understand the argument, one would have to know that for the Sa¯m ˙ khya dharmin and dharma are relative terms. One and the same thing is considered dharma in respect to a previous state of transformation and dharmin in respect to a later state of transformation.14 Thus, when the Buddhist charges that the dharmin is not known, the Sa¯m ˙ khya responds that it is very well known because the dharmin and the dharmas are the same. Put differently, even if the Buddhist were right that only dharmas are seen, the dharmin now could well be seen when it was a dharma before, for the terms dharmin and dharma refer to the same thing in different states of transformation. Thus, the Sa¯m ˙ khya retorts: “Who says that the dharmin is different from the dharmas? As for the

transformation [it does not imply that they are different, but] only [means that] the very same substance changes.” Gold, however, misunderstands the rhetorical question of the Sa¯m ˙ khya and splits his statement between opponent and proponent: “What about saying that the dharma-possessor is other than the dharmas? But the substance itself has a change, simply by becoming possessed of difference [anyathībhāva]. This is what is illogical.”

Thus, the dialectical structure gets lost, and the dialogue becomes incomprehensible. One could go on and on like this, but one may also ask to what extent the many misunderstandings and misinterpretations actually affect Gold’s general thesis in this book. One may be wrong on the details, but get the general idea right. Unfortunately, this is not at all the case. Gold’s general thesis has problems of a different, mainly methodological order. In order for it to be convincing, Gold would have to prove both Vasubandhu’s originality and intention. So the question that arises next is: how much of the philosophy that appears in Vasubandhu’s works is original? Strictly speaking, the question cannot be answered, for to know what is new, one has to know everything that is old, and our knowledge of Vasubandhu’s predecessors is incomplete, not to say fragmentary. Even so, one could still proceed pragmatically and tentatively assume that all ideas that appear in Vasubandhu’s work and do not appear in the known works of his predecessors are 14 See for instance Chakravarti (21975, pp. 198–200).

Vasubandhu’s own. This would hardly be a perfect criterion for originality, but we live in an imperfect world. Yet, Gold does not attempt to demonstrate Vasubandhu’s originality in this way. As is well known, the AKBh, which forms the textual foundation of Gold’s study, relies heavily on the Mahāvibhāṣā, of which it is considered to be a summary. Yet Gold does not trouble to compare the AKBh with any of the Vibhāṣās. Perhaps even more surprising is the fact that he does not take the Yogācārabhūmi into consideration. Robert Kritzer has presented a fundamental study in which he compares the AKBh with YBh and demonstrates that many of the opinions attributed in the Kośa to the Sautra¯ntika actually appear in the YBh. Gold knows, of course, of Kritzer’s work, but unexplainably does not use it for the present purpose. What does Gold claim to be Vasubandhu’s original philosophy? On p. 94 he summarizes “some of the central methods and motives of

Vasubandhu’s philosophy”: “In addition to presenting a consistent view, the ‘Sautra¯ntika’ critique of Vaibha¯s ˙ ika Abhidharma has a number of characteristic philosophical strategies. Above all, the goal, in argument after argument, seems to have been to dismantle the edifice of the opponent’s system that supports one or another false entity (a past/ present/future reality, a self, a cloth). In each case, Vasubandhu targets the causal logic of the entity—its contribution to the Vaibha¯s ˙ ika or non-Buddhist system.” But surely, the rejection of a self and a cloth is common to most if not all Conservative Buddhists. As for the

rejection of the existence of past and future dharmas, Kritzer has already laid the ground for the research necessary to define Vasubandhu’s originality (p. xxxiv): “In chapter five of the Abhidharmakośabhāṣya, Vasubandhu criticizes at length the basic Sarva¯stiva¯din principle that past, present and future dharmas all really exist. Many of his arguments are similar to those found in the refutations of sarvāstivāda found in the Viniścayasaṅgrahaṇī on the

Pañcavijñānakāyamanobhūmi and the Savitarkādi-bhūmi (items 5.12-18 and 5.21-27 in Chapter Three [i.e. Five?] of this book).” Gold may retort that the way Vasubandhu rejects these entities is new, but how can we know that without an extensive comparison with the relevant passages? Similarly, in the chapter on Vasubandhu’s epistemology (which I will not discuss here, in spite of a number of unfounded affirmations), Gold concludes (p. 113): “As soon as we attend to it, we see that one of Vasubandhu’s principal philosophical concerns was to combat dogmatism and literalism in the interpretation of scripture.” I would like to see a single Buddhist philosopher who would not claim such a concern. Or put more cautiously, every Buddhist philosophical tradition claims that its rival traditions are misrepresenting the Buddha’s word, taking it literally and misunderstanding its true and deeper intention. Certainly, this hardly distinguishes Vasubandhu

from numerous fellow Buddhists. Coming to the Yoga¯ca¯ra chapter, one has to note first that Gold relies heavily on the Trisvabhāvanirdeśa, a treatise whose attribution to Vasubandhu is doubtful. Even if one grants, at least for the sake of argument, that this attribution is correct, the same questions as above immediately arise. Does this short treatise present an original position by Vasubandhu, or does it merely restate the traditional Yoga¯ca¯ra tenets succinctly? Gold obviously opts for the former alternative, but I fail to see that he substantiates his position. Certainly, the mode of argument he claims to be so

typical of Vasubandhu is absent in this work. On the whole, Gold’s interpretation of Vasubandhu’s idealism will raise a few eyebrows (pp. 223–224): “Far from being a counterintuitive, solipsistic idealism, Vasubandhu’s worldview resonates in a commonsense way with the unavoidable recognition that we are medium-sized beings living in a world beholden to the physics of subatomic particles. Buddhist Causal Framing places sensible conditions on the kinds of affirmation such a reality permits: we are subject to our own, necessarily limited, frames of reference, but within those frames we are responsible to the inputs of objectivity-testing, which can call the frames themselves into question.” This position no longer needs to be refuted. It is akin to what one may call the epistemological

interpretation of Vasubandhu’s argument in the Viṃśatikā, as propounded for instance by Claus Oetke,15 which has been thoroughly refuted by Birgit Kellner and John Taber.16 Oetke’s interpretation was obviously improbable, but it was not at all obvious to show where exactly and why it was wrong, and this has now been convincingly demonstrated by Kellner and Taber. Now, coming to the question of intention, it is, as every lawyer will tell you, very difficult to prove when not stated explicitly, and Vasubandhu certainly does not say what Gold says he intends. The upshot of Gold’s thesis, which is expressed in the title of the book, is that Vasubandhu’s various works are part of a grand design (p. 134): “It [the Yoga¯ca¯ra view] is difficult to understand let alone to prove. Yet Vasubandhu’s

works provide an excellent guide: a gradual development from AKBh to systematic Yoga¯ca¯ra, in manageable stages. Vasubandhu’s works pave the way.” What are these “manageable stages”? If I understand correctly, they are centered on Vasubandhu’s rejection of entities that are not causally effective (p. 139): “As we follow the argument in the Vim ˙ s ´, we will see that Vasubandhu continues to pursue his relentless AKBh practice of rejecting perspectives and entities that show no causal contribution to the arising of experience. By providing a defense of the ‘appearance only’ doctrine in terms continuous with the causal logic of

the AKBh, Vasubandhu’s Vim ˙ s ´ paves the way for the Maha¯ya¯na critique of all reification of entities—and the scriptures in which these views arise—to appear both sensible and quintessentially Buddhist. It is the first stage in the path to a Yoga¯ca¯ra synthesis.” But again we ask: Is such a rejection unique to Vasubandhu or is it a commonplace for Buddhist philosophers? And even if Gold would be able to prove that it is unique to Vasubandhu, the question of intention

will remain, at best, open, for intention is difficult to prove, all the more so as the first and arguably most important step of the grand design would be nothing less than bewildering, or indeed counterproductive: Would it really have been necessary, or advisable, to produce a work as vast and comprehensive as the AKBh as a “manageable stage” on the path towards the Yoga¯ca¯ra philosophy? Couldn’t Vasubandhu have found a more efficient way of “paving the way” by concentrating on a few salient features of Abhidharma

15 Oetke (1992). 16 Kellner and Taber (2014). Note that this paper was published at almost the same time as Gold’s book and thus could not be taken into consideration by him. However, several other relevant studies which deal with this question, such as Schmithausen’s criticism of Dan Lusthaus’s phenomenological interpretation of Vasubandhu (Schmithausen 2005), are also not taken into serious consideration. Vasubandhu the Unified

philosophy? Such questions are not addressed in Gold’s work. What we get instead are statements like the following (p. 165): “We could hardly hope for a closer connection between Vasubandhu-as-commentator’s obsession with the ‘graspergrasped’ gloss of ‘duality’ and Vasubandhu-as-AKBh-author’s obsession with the linear, causal sequencing of ‘view.’” I, for one, failed to see this connection, or the “obsession” for that matter. So Gold’s thesis remains to be proved. Stripped to its basics, his claim is a modern version of the traditional Maha¯ya¯na view that one has to understand the pudgalanairātmya before being able to understand the dharmanairātmya, and that the Buddha taught the pudgalanairātmya for this purpose, applied to Vasubandhu and his supposedly unified thought. If you were not already convinced by the Maha¯ya¯na scriptures, you are not likely to be convinced by Gold’s book.

Acknowledgements As always, I am greatly indebted to my wife, Karin Preisendanz, for her meticulous reading of this review and for extremely helpful comments. I am also indebted to the Academy of Korean Studies KSPS) for a very generous grant funded by the Korean Government (MOE) (AKS-2012-AAZ104).

References

Bhaduri,S.(1975).StudiesinNyāya-VaiśeṣikaMetaphysics.Poona:BhandarkarOriental ResearchInstitute. Buescher, H. (2013). Distinguishing the Two Vasubandhus, the Bha¯s ˙ yaka¯ra and the Kos ´aka¯ra, as Yoga¯ca¯ra-Vijn ˜a¯nava¯da Authors. In U. Kragh (Ed.), The Foundation for Yoga Practitioners: The Buddhist Yogācārabhūmi Treatise and Its Adaptation in India, East Asia, and Tibet (pp. 368–396). Cambridge, MA: Harvard University Department of South Asian Studies and Harvard University Press. Chakravarti, P. (21975) Origin and Development of Sāṃkhya System of Thought. Delhi: Munshiram Manoharlal Publishers. Kellner, B., & Taber, J. (2014). Studies in Yoga¯ca¯ra-Vijn ˜a¯nava¯da idealism I: The interpretation of Vasubandhu’s Vim ˙ s ´ika ¯. ASIA, 68(3), 709–756. Kritzer, R. (2005). Vasubandhu and the Yogācārabhūmi: Yogācāra Elements in the Abhidharmakośabhāṣya. Tokyo: International Institute for Buddhist Studies. Lamotte, E. (1936). Le traité de

l’acte de Vasubandhu. Karmasiddhiprakaraṇa. Mélanges Chinois et Bouddhiques, 4, 151–263, in p. 179. Oetke, C. (1992). Doctrine and argument in Vijn ˜a¯nava¯da-Buddhism. Wiener Zeitschrift für die Kunde Südasiens, 36, 217–225. Schmithausen, L. (1967). Sautra¯ntika-Voraussetzungen in Vim ˙ s ´atika¯ und Trim ˙ s ´ika ¯. Wiener Zeitschrift für die Kunde Süd- und Ostasiens, 11, 109–136. Schmithausen, L. (2005). On the problem of the external world in the Ch’eng wei shih lun. Tokyo: The International Institute for Buddhist Studies