English Section

      Buddhism Today 

Vietnamese Section


...... ... .  . .  .  .

Altruism and Reality: Studies in the Philosophy of the Bodhicaryaavataara. By Paul Williams. Richmond, Surrey: Curzon Press, 1998, ISBN 0700710310, cloth 40.00, $48.00.

Reviewed by
John W. Pettit
Ajita Institute
Woodstock, New York


Abstract: A critical review of four of the five articles by Williams collected in the book, with particular focus on Williams' intrepretation of Tibetan commentarial literature on the ninth chapter of the Bodhicaryaavataara, and his analysis of "Saantideva's arguments in favor of altruism.

In 1998 the Curzon Critical Studies in Buddhism Series published two volumes by veteran Buddhist scholar Paul Williams. The first volume, which will be the subject of a separate review, is The Reflexive Nature of Awareness, and concerns the exegetical controversy surrounding "Saantideva's refutation of apperceptive awareness (svasa.mvitti) in the ninth chapter of his Bodhicaryaavataara. The companion volume, Altruism and Reality(1) is a collection of five essays on passages from the eighth and ninth chapters of "Saantideva's text.(2)

The papers collected in the Altruism volume contain by far the most probing analyses ever attempted of the Bodhicaryaavataara, with reference to Tibetan and, to a lesser extent, Sanskrit commentarial literature on it. Williams' comparative research and text-critical analysis have plunged (or perhaps, ascended) several yojanas further into the arcana of Tibetan Maadhyamika interpretation than I have ever been able to venture myself. Consequently what I, as a researcher with similar interests, had expected to be a relatively straightforward review, has mushroomed into the present article, and I have no doubt only scratched the surface of the many important issues Williams has explored. Thus, though I find many of Williams' premises and conclusions provocative and am inclined to disagree on a number points, my objections (like many of his conclusions) should be understood as tentative, pending further investigation.

The first article, "On Prak.rtinirvaa.na/Prak.rtinirv.rta in the Bodhicaryaavataara," discusses verses 13-14, 34, 103, and 110(3) (in the Tibetan enumeration) of the ninth chapter, which relate more or less directly to the idea of innate or natural Nirvaa.na (prak.rtiparinirvaana).(4) In this article, as in most of the others, Williams uses a limited selection of passages to introduce important philosophical issues, and then references a variety of commentarial sources to illustrate how various commentators have brought their interpretations to bear upon those passages.

Of particular interest here is how Tibetan commentators have tended to interpret the ninth chapter of the Bodhicaryaavataara (BCA) according to their philosophical biases as either unreconstructed Maadhyamikas -- for whom "natural nirvana" and "emptiness" are more or less synonymous -- or as "Great Maadhyamikas," whom Williams implicitly (and in my opinion simplistically) equates with adherents of the "other-emptiness" (gzhan stong) style of Tibetan Maadhyamika interpretation.

Some, but not all, Tibetan commentators on the Bodhicaryaavataara invoke the principle of gnosis or ye shes (Skt. j~naana), or one of its associated terms such as "natural purity" (prak.rtivi"suddhi) in interpreting "Saantideva's statement at 103CD. Here, Williams suggests, we find a typically Cittamaatra concept, which is alien to the philosophical context of "Saantideva's text and his Indian commentators', being invoked by Tibetan scholars who were steeped in the lore of such texts as the Ratnagotravibhaaga. Without a doubt, as the heirs and curators of a variety of commentarial materials which were not always widely known or cross-referenced by commentators in the land of their origin (which incidentally appears to have been the case with Candrakirti's Maadhyamika commentaries), Tibetans were more likely than their Indian predecessors to forge creative interpretations. That is to say, they tended to interpret textual passages out of the precise philosophical-historical context of their original formulation. This tendency is something which has long fascinated Williams, and here he has done well in naming the usual suspects for this practice (such as my own virtual mentor, Mi pham(5) and in providing us with specific textual evidence. However, in using such words as "Cittamaatra" and "Great Madhyamaka " and "gZhan stong absolutism" (p. 15), Williams has sometimes gone out on a limb, perhaps knowingly.

Certainly one finds j~naana discussed frequently in the Ratnagotravibhaaga (RGV), where it is linked to the tathaagatagarbha concept, and in so-called "Cittamaatra" or "Yogaacaara" texts, while in Maadhyamika texts it is scarcely mentioned. Clearly it is the influence of the RGV and the gZhan stong traditions it inspired which underlie, for example, Sa bzang Mati Pa.nchen's interpretation of 9:103CD.(6) On the other hand, the RGV cannot be unproblematically classified either as a Maadhyamika or as a Cittamaatra text, and Tibetan commentators were hardly univocal here; in any case, it is interesting that that the Gelugpas, especially rGyal mtshab Dar ma rin chen, interpreted the RGV as a Prasaa"ngika text.

The exegetical use of a philosophical term outside of its normative philosophical context does not necessarily mean that a text is being interpreted out of context, nor does it necessarily mean that the author is bringing an innovative interpretive bias to bear on the text though one or the other of these possibilities may turn out to be the case. As far as the various commentarial traditions of the Bodhicaryaavatara are concerned, one might just as well assert that terms which the commentators understood to be non-contradictory or equivalent were invoked as needed to uncover nuances already understood to be implicit. In other words, the fact that RGV-esque language is used to interpret BCA 9:103CD does not necessarily mean that what BCA 9:103CD says is not, in fact, essentially the same as analogous passages in the RGV, or in the scriptural antecedents of the RGV. The use of variant terminology and the quotation of sources from historical contexts different from the text itself do not necessarily mean that, hermeneutically speaking, there is an out-of-context interpretation. For example, though "Saantideva does not use the language of the RGV, it is quite possible that he had read the text, and it is a fact that he studied many of the Suutras teaching the Buddha-nature concept (snying po'i mdo), which constitute scriptural antecedent for the RGV(7). What this means, in my opinion, is that non-Praasa"ngika trends of thought are effectively a part of "Saantideva's intellectual-historical context, whether he acknowledges them or not. Thus I would conclude that the apparent affinity of the prak.r.tinirvaa.na concept in the BCA and the tathaagatagarbha-concept of the RGV is probably more than accidental.

Whether a particular philosophical interpretation of a "Praasa"ngika" text such as "Saantideva's ninth chapter is "extra-contextual" is itself a matter of context. Does one hold terms like "luminosity" and "gnosis," on the one hand, and "emptiness" and "nirvana," on the other, to belong to separate and distinct realms of discourse, or to complementary or equivalent ones? If one assigns the use of these terms to strict doxographical categories, then any equivalence of such terms, implicit or otherwise, might seem controversial. Once again it should be noted that Indo-Tibetan doxographical categories are problematic; the primary sources themselves do not necessarily fit neatly into the doxographical pigeon-holes into which tradition has often wanted to stuff them. I am sure Williams is aware of all this. In mentioning these points I am simply following my own interpretive predilection, which is to try to understand the implicit hermeneutic of a particular author -- which more often than not falls into a doxographical gray area -- rather than playing the Tibetan doxography game.

Another place where Williams has begun to tread on thin ice is in implying that Mi pham's use of the expression "mere gnosis which is reflexive awareness" (so so rang rig pa'i ye shes tsam) in his commentary on 9:35(8) is practically the same as saying that gnosis is "really the ultimate." Williams then suggests (though not in so many words) that this is precisely the dBu ma chen po (Great Madhyamaka = gZhan stong), which stems historically from the Jonang tradition which is known to have maintained that the ultimate is a "really existing radiant gnosis" (p. 25). Historically speaking, this connection is not implausible -- Mi pham was the disciple of two prominent gZhan stong pas, 'Jam mgon Kong sprul and mKhyen brtse dBang po -- but "gZhan stong pa" is probably not a label with which Mi pham himself would have been comfortable. Though it is correct to characterize Mi pham's thought as "Great Madhyamaka," insofar as he does so himself, to suggest that his thought stems more or less directly from the Jonang gZhan stong tradition is problematic. For one thing, the term so so rang rig pa'i ye shes (pratisa.mvidj~naana) is generally understood to mean the gnosis (ye shes = j~naana) which realizes what can only be realized for oneself (so so rang gis rig pa = pratisa.mvid). This simply means that emptiness proper, as a non-conceptual ultimate (rnam grangs ma yin pa'i don dam = aparyaayaparamaartha), is not accessed by words or concepts, so Mi pham's use of the term so so rang rig pa'i ye shes is hardly incriminating. In spite of Mi pham's tendency to use terms like "coalescence" (zung 'jug = yuganaddha), "luminosity," and so forth in his Maadhyamika commentaries (especially his Sher 'grel nor bu ke ta ka, on the BCA's praj~naapariccheda), he was generally at pains to distance himself from the gZhan stong position (or positions). For example, he upholds "Prasaa.ngika" in one of his most important texts, the Nges shes rin po che'i sgron me; he promotes the "Yogachaara-Svaatantrika-Madhyamaka" in his commentary on "Saantarak.sita's Madhyamakaalamkaara; and he evidently devoted a great deal of effort to defending his Nor bu ke ta ka against its critics.

Though Mi pham did write a polemical defense of gZhan stong, the gZhan stong khas len seng ge'i nga ro -- which is said to have been written out of compassion for the weak philosophical acumen of the gZhan stong pas, or perhaps to fulfill the wishes of his teachers -- he does not argue there that "truly existent gnosis" is the ultimate, but rather that what is known by ultimate (that is, enlightened) wisdom is true, while what is taken for granted in conventional knowledge (true existence) is false and non-existent. This is not the quite same as saying "gnosis is the truly existent ultimate reality," which is the stereotyped position usually attributed, rightly or wrongly, to the gZhan stong pas by their opponents. Generally speaking, Mi pham accords equal ultimacy to reality per se (tattva, dharmadhaatu, "suunyataa, etc.), and to that which knows it (j~naana) -- but only because, from the perspective of gnosis, the knower and what is known cease to exist apart. This is standard Madhyamaka, and does make him either a Cittamaatrin or a gZhan stong pa.

Mi pham's position on gZhan stong in the gZhan stong khas len seng ge'i nga ro may not exactly reflect what earlier gZhan stong pas themselves had been saying, but even as such it would illustrate an important point. The term "gZhan stong" has meant many different things to many different people, and in using it one is very likely to obscure, or over-simplify, the philosophical position of one or another Tibetan author. Though Williams excels at comparative Tibetan philosophy, and is generally careful to balance his frank admiration for Gelugpa innovations in epistemology and philosophical hermeneutics with the acknowledgement these are just one set of "possible interpretations among many others" (p. 28), in the case of gZhan stong I think he has been too quick to accept the characterizations given of that trend of Maadhyamika interpretation by its opponents, for example, in calling it a "philosphical absolutism." What would make it an absolutism is a matter of hermeneutics, and in Tibet there was more than one way of assessing the philosophical Middle Way; it goes without saying that as self-described "Great Maadhyamikas" the gZhan stong pas did not consider themselves absolutists. The question is, on whose standard is "gZhan stong" (whatever that may be) a philosophical "absolutism"? In referring to a vaguely defined, inclusive "gZhan stong absolutism," Williams (perhaps following D. S. Ruegg) seems to have shown a rather uncritical acceptance of Gelugpa polemics. Be that as it may, the way Williams uses gZhan stong certainly points to the need for further research into the writings of authors assigned to this little-understood and much-maligned "school."

It should also be noted that it is not correct to equate "Great Madhyamaka" (dbu ma chen po) with gZhan stong, which seems to be an implicit assumption in the prak.rtinirvaa.na article, and is made more explicitly in the penultimate article (for example, on page 98). This term has been used by many "Praasa"ngika" authors as an epithet of their own systems, Atii"sa(9) and Tsong kha pa(10) included. Though gZhan stong pas tended to trumpet the same epithet in their anti-Praasa.ngika polemics, in the wider context of Tibetan Madhyamaka it merely would seem to indicate the gnostic dimension of Madhyamaka, as opposed to the philosophical discourse known as "Madhyamaka."

Rather less controversial is Williams' second essay, entitled "Altruism and Rebirth." The discussion centers on a single verse, 8:97:

Supposing one says that the suffering which happens to that (other) person does no harm to me, therefore (s)he should not be protected against (it)|

Then since future suffering (Skt. "the sufferings of future bodies") also is doing no harm (to you now) why is that to be protected against?(11)

As in the last article, Williams' analysis in large part concerns the rational coherence of the spiritual path. In this verse, what is at issue is not so much proving the rationality of the Mahaayaana Buddhist ethic of universal compassion, as it is disproving the opponent's position, which is that there is no compelling reason to be as concerned about others, in the same way as one is about oneself. As before, Williams reviews the various commentators' positions at length, and concludes that -- at least as far as the original Sanskrit, where "future bodies" are mentioned, is concerned -- one's future mind and body are indeed equally "other" to one's "self" (or causal psychosomatic continuum), as are the bodies and minds of beings other than oneself. However, the notion that future lives are "other" is problematic, insofar as it might tend to encourage one to discount the effects of one's actions in the future; this would be the "dreaded ucchedavada" or nihilism (p. 50). In the last analysis Williams is willing only to concede that "Saantideva has only succeeded in showing an incompatibility between the intention to act out of concern for one's own future lives and neglecting the welfare of sentient beings in the present, because one's "own" future lives and contemporary sentient beings are equally "other" to oneself at present. One detects here a dissatisfaction on the author's part with the rationality or efficacy of "Saantideva's arguments, which dissatisfaction becomes more apparent in his final essay.

The fourth chapter is "Identifying the Object of Negation." This refers to an aspect of Maadhyamika reasoning and meditation which is particularly emphasized in the Gelug school, namely, the requirement that in order to ascertain emptiness correctly, one must correctly identify that which emptiness is the negation of, namely inherent existence (rang bzhin = svabhava) or true existence (bden grub). The first half of verse 9:140 (Tib. 9:139) of the Bodhicaryaavataara is invoked by Tsong kha pa and his followers as a locus classicus:

Not having contacted a conceptually-constructed entity |
The negation of that is not apprehended |
Therefore, in the case of a delusory entity |
The negation of that is clearly delusory. | |(12)

Williams is not the first Western scholar to make note of the importance of this verse for Gelug Maadhyamika interpretation(13), but as far as I know he is the first to closely examine the verse's context, in terms of the BCA itself and its various commentators. His article raises some interesting points.

First of all, the Tibetan translation of the first paada of the immediately preceding verse (139/Tib. 138) does not accord with the BCA's Indian commentators.(14) The Tibetan translators apparently read bhaavanaa.m, taking this to be the accusative singular of bhaavanaa (meditation = sgom pa), while the Sanskrit reads bhaavaanaa.m, the genitive plural of bhaava (things or entities = dngos po). With the exception of Bu ston, Tibetan commentators did not notice the discrepancy between the Indian commentaries and the Tibetan translation, and thus took the preceding verse as the puurvapak.sa's refutation of the viability of meditation on emptiness, while the Sanskrit original and its Indian commentaries clearly mark the verse as a puurvapak.sa's objection to the authenticity of emptiness per se. Williams holds that in the final analysis the apparently erroneous Tibetan reading of this verse is not crucial to the overall structure of "Saantideva's argument. However, it is interesting that the translation of bhaavaanaa.m as "meditation" would tend to reinforce -- if it did not in fact incite -- the way in which Tibetan commentators (especially those whom Williams identifies as cleaving to the gnostically-oriented "Great Madhyamaka") have commented on the subsequent verse (9:139, Tib. 140).

According to most commentators, 9:139 is a more or less unconditional acceptance of the puurvapak.sa's premise in the previous verse, namely, that emptiness is delusory (m.r.saa) because means of valid cognition (pramaa.naa.h) -- which are themselves empty, hence delusory -- cannot establish emptiness as valid, or non-delusory. The literal statement (if not the underlying intent) of the puurvapak.sa is acceptable, because all conventional realities, such as proof and refutation, are delusory insofar as they are empty. Thus emptiness, though delusory, can negate inherent existence just as the death of a son in a dream prevents the conception of his existence from arising.(15) Thus goes the more conventional and commonsensical way of interpreting the sense of BCA 9:139.

Tsongkhapa's interpretation puts a rather different spin on "Saantideva's verse, and is typical of the Gelugpa style of Praasa"ngika Madhyamaka interpretation. On the face of it, "Saantideva is saying that without "contacting" (or imagining) an entity which is to be negated, a negation is not possible, as in the case of the death (negation) of a son (an entity) in a dream. This is not what is actually meant, according to Tsong kha pa and his disciple rGyal mtshab rje. For Gelugpas, to take the negandum of Maadhyamika negation to be any particular conventional existing entity -- whether it be, for example, a dead son qua dream-phenomenon or a dead son in "real life" -- is mistaken, because then emptiness would be simply the negation of that thing; that is, emptiness of "son" would be equivalent to "not-son," and so forth. Of course, that is not what emptiness means. Instead, emptiness is the absence of svabhaava, inherent existence, of any entity. Thus what "Saantideva means to say, according to Gelugpa Madhyamaka, is that without "contacting" (or conceptualizing) the inherent existence ((sva)bhaava) of an entity, its opposite (abhaava = ni.hsvabhaava) cannot be apprehended either. For Gelugpa Madhyamaka, that would be a shame -- for if one does not apprehend with critical awareness the nature of things (ni.hsvabhaava = "suunyataa), then one is bereft of wisdom (praj~naa).

So far this may seem uncontroversial. However, other commentators have detected a resonance between 9:139-140 and 9:35, which reads in Williams' translation:

When entity (bhaava) and negation (abhaava) |
Do not stand before the mind |
Then because there exists no other possibility |
Without intentional object it is calmed.(16)

This resonance is all the more palpable if one recalls that 9:138CD was translated into Tibetan as a refutation of the viability of meditation on emptiness, rather than as a refutation of the rationality of emptiness per se. On this reading, the following verse 9:139 naturally would be taken as illustrative of the actual method of Maadhyamika meditation. The reference in 9:35 to "without intentional object" (niraalambaa) clearly indicates that meditation on emptiness should be without object, and so 9:139 has been read by some commentators such as Mi pham as an exhortation to just such a meditation, where "one abides in the gnosis which is the calming of all verbal differentiations...the Great Madhyamaka free of all assertions."(17) This stands in rather marked contrast to Tsongkhapa's interpretation of 9:140, which emphasizes the verbal differentiation of negandum and negation, and implicitly, of the negandum (inherent existence) and its substratum (a conventionally existing phenomenon perceived to exist inherently).

Basically there are two models of "emptiness" at work here: emptiness as the absence of elaboration (ni.sprapa~nca = spros bral), and emptiness as the negation of a negandum. Though Gelugpa Maadhyamika commentaries are known for their emphasis upon the latter type of emptiness, the former is also recognized by dGe lugs scholars, if not always elaborated upon.(18) For example, rGyal mtshab rje glosses the statement at BCA 9:2, "reality is not an object of mind" (buddheragocarastattvam = don dam blo yi spyod yul min), as "reality is not an object of dualistic mind" (don dam gnyis 'dzin gyi blo yi spyod yul min)(19). Assuming that dualistic perception is inclusive of elaboration (prapa~nca), then ultimate reality is presumably to be known in the absence of elaboration. If emptiness is the ultimate reality, then in definitive knowledge emptiness should should always involve the absence of elaboration; on the other hand, emptiness as the negation of a negandum, while still a form of emptiness (referred to in the Lankaavataarasutra, notes Williams, as itaretara"suunyataa, the emptiness of one thing in another), would not be a definitive emptiness, because there would still be an elaboration of the absence or non-existence of svabhaava. Such is the position of self-described Praasangika commentators like Go ram pa, Mi bskyod rDo rje, and Mi pham.

Why then, if the ni.sprapa~nca interpretation of emptiness is commonly upheld by Tibetan Praasangikas, is Williams so quick to assimilate Mi pham's interpretation of 9:140 to "gZhan stong"? To return to the previous quote:

...(W)hen one abides in the gnosis which is the calming of all verbal differentiations (prapa~nca), there is the Great Madhyamaka free of all assertions. And Mi pham quotes to this effect from a suutra of the Praj~naaparamitaa type, thus assimilating the gZhan stong Great Madhyamaka into the message of the Praj~naaparamitaa. The quote concludes: "There is seen as objective realities neither agent nor all dharmas"...For Mi pham the point is to finish with existence and non-existence, abiding instead in non-conceptual gnosis which by definition can only occur when all concepts...have come to an end (pp. 98-99; my bold italics).

Perhaps it is the word "gnosis" (ye shes = j~naana) which Williams takes to be the gZhan stong smoking gun. If that is so, his suspicion is simply unwarranted. Clearly, as Williams himself says, Madhyamaka seeks the transcendence of dualities which are relative by nature. In the context of ultimate reality, gnosis and what it knows (emptiness) are non-dual; so why then should the presence of "gnosis" implicate a "gZhan stong absolutism" which -- at least in its opponents' characterizations of it -- inculcates a radical dichotomy of ultimate and relative truths? If emptiness is definitively (that is, non-dualistically) known by gnosis, then it is not necessarily unkosher to speak of gnosis in the context of Madhyamaka, nor is it necessarily incorrect to speak of gnosis as being ultimate, especially if one follows Bhavaviveka's gloss of paramaartha as "that which conduces to the realization of ultimate reality."(20)

Having said all that, in conclusion, Williams' article on the Maadhyamika negandum is an informative and thought-provoking exploration of a topic of central importance in Tibetan Maadhyamika interpretation and meditation practice, which should keep Tibetan Madhyamaka fanatics (like the present reviewer) busy for years to come.

Williams' final article in the volume, "The Absence of Self and the Removal of Pain: How "Saantideva Destroyed the Bodhisattva Path," on the other hand, ought to provide amusement for aspiring Mahaayaana polemicists. The very title seems absurd; how could "Saantideva, the most renowned advocate of the selfless bodhisattva way in Indo-Tibetan commentarial literature, have "destroyed the bodhisattva path"? If I have followed Williams' argument correctly, what the problem boils down to is the fact that, as Williams says at the beginning of the article,

"Saantideva takes as an assumption that the disinterested nature of morality is fulfilled by rational consistency, and a moral imperative can be drawn from what is, in the broadest sense, an ontological position.

On the face of it, the idea that "Saantideva's moral rationalism ought to undermine his project of elucidating and defending the bodhisattva path also seems preposterous. Has "Saantideva, one of the most compelling (and presumably coherent) of Indian Buddhist philosopher-saints, contradicted himself? Or, perhaps, has Williams come up with a better, presumably more rational approach to the bodhisattva path? The answer to the first question, in all fairness to Williams' extensive research and thoughtful analysis, should be "maybe"; the answer to the second, in all fairness to "Saantideva's good intentions, must be a resounding "no."

So far as I know, no other scholar in history, Buddhist or non-Buddhist, has ever attempted to depict "Saantideva as a shabby defender of the faith. In constructing a long, exhaustive (and in my opinion, fundamentally misguided) series of arguments in an attempt to show that he was, Williams has boldly blustered where Tirthikas and Buddhologists alike have feared to tread. And I think he has stuck his neck out. Before I let the polemical axe fall, I should quote the verses at issue here, and review what I take to be the most important points in Williams' argument.

Verses 8:101-3, in Williams' translation, read as follows:

A continuant and a collective -- such as a (caste) row (pa"nkti) or an army -- are fictions (m.r.saa) |
The one of whom there is a pain (du.hkha) does not exist. Therefore of whom will there be ownership of that? | | 101 | |

Pains without an owner are all indeed without distinction |
Because of its quality as pain indeed it is to be prevented. What limitation can be made there? | | 102 | |
If one asks why pain is to be prevented (Tib. "the pain of all is to be prevented"), it is (accepted) (Skt. "by all") without dispute |
If it is to be prevented, all is also thus. If not oneself also is like (other) beings. | | 103 | | (21)

To paraphrase "Saantideva's point: the sentient being is a composite entity; hence there is no self or owner of pain -- only, presumably, the illusion of ownership. Thus in truth pain really belongs to no one (or perhaps, as "Saantideva would have it, to everyone) -- but nonetheless, as pain, it should be prevented. Verse 8:103 elaborates on verse 8:102. If one wishes to prevent one's own pain, one should also prevent the pain of others; if one does not wish to prevent the pain of others, one should not prevent one's own pain -- because, as previously stated, there is really no owner of pain, so one's own and others' pains are not really different.

Williams finds fault with this argument on several grounds. His basic thesis is that "the fact" -- according to Buddhist philosophy -- "of no True Self does not in itself [entail] unselfishness...the 'ought' of unselfishness simply does not follow from the 'is' of anaatman." In other words, I may not accept that there is a metaphysical True Self (Aatman), but that does not necessarily mean that I would or should neglect, be incognizant of, or otherwise fail to show preference for "myself" -- in the everyday conventional sense of that word. Another point Williams raises, and illustrates at great length, is that an "ownerless" pain is really no pain at all. Without a subject of pain, there can be no pain. A pain is not a "free-floating" entity, like a pair of shoes which one person can slough off and another put on.

Although time does allow me to examine Williams' argument at every place I suspect he has taken a wrong turn, I would like to raise a few points which, even if they do not actually disprove his assertion that "the 'ought' of unselfishness simply does not follow from the 'is' of anaatman," might suffice to call it into doubt.

First of all, I think Williams is confused, or at least underinformed, about the negandum (prati.sedhya = dgag bya) of Maadhyamika analysis. That is a strong claim, of course; hasn't he just given us a beautiful analysis of this subject in a previous chapter? Nonetheless, I think that in this article his understanding (or practical assumption) what "self" (or Self) is negated by Maadhyamika reasoning is incomplete -- perhaps intentionally. The reader will notice Williams' capitalization of the word "self" as in "True Self," which I take to imply the Aatman of Upani.sadic and other traditions of Indian philosophy. The negation of the True Self, he says, will not suffice to eliminate our everyday conception of "myself," nor presumably our instinctive preference for that self over other selves. I also take it that Williams holds this to be the case because in Madhyamaka the ordinary conception of "John" as a person distinct from "Paul" and so forth is held to be a valid distinction; that is the way things are conventionally, and we wouldn't want to deny it lest chaos and confusion result. Thus, if negating a "True Self" doesn't negate the conventional self, which is the basis of everyday distinctions of "self" and "other" (and presumably, of selfish decisions and actions), then what good will it do for the cultivation of selfless compassion?

Here Williams seems to have overlooked some important distinctions of the Gelugpa Prasan-gika tradition which he seems to hold dear, namely the "rational negandum" (rigs pa'i dgag bya) versus the "path negandum" (lam gyi dgag bya), and innate ignorance (lhan skyes kyi ma rig pa) versus theoretical or conceptualized ignorance (kun brtags kyi ma rig pa = parikalpitaavidyaa). The rational negandum of Maadhyamika analysis would include the "True Self" or Aatman (which is a grub mtha'i sgro btags pa or philosophical conception) as well as the innate conception (lhan skyes kyis sgro btags pa) of self. In other words, any kind of self which is conceived to exist inherently is a rational negandum of Maadhyamika analysis -- whether such conception is conscious or not.(22) The root of sa.msaara, according to Tsongkhapa, is innate ignorance, which would consist in an innate misapprehension of a self, while the theoretical misconception of a "True Self" would be a conceptualized ignorance which only obtains for those who uphold the philosophical systems which teach it. Thus, the innate misconception of self is the more important negandum of Maadhyamika analysis, and because it is not particularly conscious, it is more difficult to deal with.

Williams is certainly correct in arguing that to negate the philosophical misconception of a True Self will not do much to undermine our tendency to selfishly prefer ourselves over others. However, notwithstanding "Saantideva's refutations of non-Buddhist philosophical theories, I think what "Saantideva is most concerned with is the innate misconception of self, and likewise, what Maadhyamika meditation is really concerned with is the practical application of reasoning to that innate misconception, that is, as a "path negandum." Thus in the dGe lugs pa tradition of analytical meditation there is much emphasis upon "identifying the object of negation" (dgag bya'i dngos gzung ba). A well-known example in the oral tradition of how this identification is made refers to how a sense of "I" or "me" arises when one is falsely accused or verbally abused. The sense of "I" about which one feels rigidly defensive under such circumstances is just a particularly glaring example of the innately misconceived self, which misconception is by definition a source of conflicting emotions. On the other hand, the valid conventional self is not necessarily associated with conflicting or selfish emotions; it is simply a fact, as in "I am typing this article," which, as statements go, is no more implicative of a belief in a truly existing, emotionally conflicted self, than the statement "roses are red" would be implicative of a belief that all roses are intrinsically red.

Maadhyamika analytical meditation should inculcate a profound certainty that the sense of self that is present in moments of emotional duress is a fabrication and does not exist in reality. It should also be kept in mind that part and parcel of such difficult moments are those "others," real or imagined, sentient or insentient, which are perceived to impinge upon that self in a negative way. Given these considerations, Williams' thesis that "the 'ought' of unselfishness simply does not follow from the 'is' of anaatman" can be brought seriously into question. If one's rigid sense of separation (via emotional clinging and rejection) from others can be dissolved through demolishing the misconception of a vulnerable emotional self, then naturally one's "selfishness," stemming as it does from a false belief in that self, will likewise be eliminated.

One of the faults in "Saantideva's argument, according to Williams, is that realizing selflessness would mean that one could not longer make meaningful distinctions between oneself and others. However, I think that to eliminate that selfishness would not have to mean, as Williams seems to believe, that one would no longer favor oneself in a natural quotidian fashion. For instance, to insist on brushing someone else's teeth before one's own would not be indicative of selfless compassion, but of confusion. To brush my teeth selflessly would not mean that I am unaware that my teeth -- not Paul's -- are being brushed. To the extent that the "self" which is negated is the source of conflicting emotions, however, selfless dental hygiene would be free of the anxiety of thinking, "Perhaps my teeth will not be white enough," or "maybe this toothpaste won't suffice to clear up my halitosis, so people will be repulsed by me" and so on. Again, the emphasis on the innate misconception of self, and taking it as a practical path negandum of Maadhyamika analytical meditation, must be understood to apply to the instances of self-conception which are invariably associated with negative emotions, and not to those which are emotionally neutral and conventionally valid. Though space does not allow for discussion of them here, the dGe lugs tradition is particularly rich in oral and written instructions for differentiating between these instances of "self." Clearly, to realize selflessness does not mean to become incognizant of conventional reality. As Tsongkhapa says in the Three Principles of the Path, "As long as the infallible appearance of dependent origination and the understanding of emptiness free of assertions appear separately, one has yet to realize the Muni's intention."(23)

It might still be objected that to eliminate the innate misconception of self will still not suffice to inculcate a profound identification with, and altruistic motive toward, other sentient beings, as it is granted that the conventional distinction of "self" and "other" will remain, even when the innate misconception of truly existing "self" and "other" has been eliminated. Here again, an important distinction can be invoked, that of "meditative absorption" (mnyam bzhag = samaapatti) and "meditative aftermath" (rjes thob = p.r.s.thalabdha). Once again to quote BCA 9:35:

When entity (bhaava) and negation (abhaava) |
Do not stand before the mind |
Then because there exists no other possibility |
Without intentional object it is calmed. | |

Thus, when one really meditates effectively on emptiness, no self or other is perceived; this is what is ultimately the case. In meditative aftermath, it is said, one should perceive all things in the manner of the twelve metaphors of illusion: like magical illusions, the reflection of the moon in water, a mirage, and so forth. In other words, though one perceives self and other in meditative aftermath, one knows they do not exist as they appear, and one acts accordingly. For example, if I know the entrance to the A train which appears one day out of the blue at Eighty-ninth Street is a magical illusion, I will not attempt to go down the stairs, nor will I be upset that I must walk three blocks further to Eighty-sixth Street to catch the train. Likewise, if I perceive self and other to be like illusions, I will not be upset or misguided by emotions; and seeing that others torment themselves needlessly through the misapprehension of a truly existent self as the subject of conflicting emotions, I will feel compassion and, whenever possible, will try to help them. Here Williams might quote the Praj~naaparamitaa to the effect that a bodhisattva sees no sentient beings. Again, I would take refuge in dGe lugs pa exegesis and say, "That only means a bodhisattva sees no inherently existing sentient beings -- not that (s)he sees no sentient beings in any way, shape or form!"

In conclusion, I would say that Williams' statement that ""Saantideva takes as an assumption that the disinterested nature of morality is fulfilled by rational consistency, and a moral imperative can be drawn from what is, in the broadest sense, an ontological position" is based on an erroneous assumption of his own. That is, Williams assumes that morality is essentially informed by a rational determination of the nature of reality, because that is how "Saantideva has presented his arguments in favor of a bodhisattva's universal compassion. I think Williams has left out an important piece of the puzzle, namely, personal practical experience. I would rephrase Williams' formula thus: ""Saantideva takes as an assumption that the disinterested nature of morality is fulfilled by rational consistency, but that rational consistency must be fulfilled by practical experience for a moral imperative to be drawn from what is, in the broadest sense, a personal experience of reality."

There is no evidence whatsoever that "Santideva ever intended that philosophical reflection alone should suffice to establish a moral imperative; on the contrary, Mahaayaana Buddhism teaches that perfect morality is impossible without perfect wisdom, and that perfect wisdom is impossible without meditation on perfect wisdom. Knowledge of emptiness through logical analysis alone (that is, the negation of a rational negandum) does not suffice to cure one of misknowledge and the moral foibles it entails, but meditation on emptiness (the gradual elimination of the path negandum, the innate misconception of self) does.

What it boils down to, in my opinion, is that to really understand the connection between emptiness and compassion, we must not only study about emptiness, but meditate on it as well. Perhaps this would make Mahaayana Buddhism "anti-rational" or "mystical." Pace Kant, I would simply suggest that the purpose of the BCA is, in large part, to clear up reason to make room for meditation.

The editors of the Journal of Buddhist Ethics invited Paul Williams to respond to John Pettit's review article, and his response can be read by clicking HERE.

End Notes

1. Altruism and Reality, Studies in the philosophy of the Bodhicaryaavataara: Curzon Press, 1998. Return to text

2. The reader will note that I have said nothing about the third essay in this volume, "An argument for Cittamaatra, reflections on Bodhicaryaavataara 9:28 (Tib. 27) CD." This article is a fine discussion of various commentarial traditions on that verse, and should be consulted by anyone interested in Maadhyamika critiques of Cittamaatra positions. However, as neither subject of the article nor Williams' conclusions are particularly controversial, I have declined examine it here. Return to text

3. There is some incongruence of verse enumeration in the ninth chapters of the Tibetan and Sanskrit editions. Unless otherwise noted all references are to verse numbers in Vaidya's edition of the Bodhicaryaavataarapa~njikaa. Return to text

4. The term proper is not found in the BCA, but the past passive participal of the root prak.r.tinirv.r occurs is in 103CD: yan na kaaye na caanyatra na mishram na p.rthak kacit | tan na ki.mcidata.h sattvaa.h prak.rtyaa parinirv.rtaa.h | | "That mind is not in the body or elsewhere, and is neither the same nor different than it | Since that (mind) is anything in particular, beings are naturally in nirvaa.na." Williams translates 103D as, "That (mind) is nothing at all. Therefore sentient beings are fundamentally (or 'inherently') ceased" (p. 12). Return to text

5. Mi pham Rin po che (1846-1912); cf. my Life and Works of Mipham Rinpoche (http://www.hvinet.com/jwp/Miphambio.html). Return to text

6. "That adventitious conventional mind, if it is examined, is not the slightest bit established. Therefore, because the ultimate dharmataa is invariable, sentient beings are established as having the essence (snying po can) of enlightenment, the clear-light nature of the mind." This line of reasoning is reminiscent of oft-quoted RGV 27: sa.mbuddhakaayasphara.naat tathataavyatibhedata.h | gotrata"sca sadaa sarve buddhagarbhaa.h "sariiri.na.h | | Tib. reads: | rdzogs sangs sku ni 'phro phyir dang | | de bzhin nyid dbyer med phyir dang | | rigs yod phyir na lus can kun | | rtag tu sangs rgyas snying po can | Return to text

7. In the "Sik.sasamuccaya "Saantideva quotes the La"nkaavataarasuutra, the "Sriimaaladeviisi.mhanaadasuutra, the Anuunatvapuurnatvasamudgataparivarta, the Ratnaku.ta, the Tathagatako"sa (= -garbha) suutra, the A"ngulimaalikasuutra, the Suvar.naprabhaasottamasuutra, and the Ratnaku.tasuutra. All of these are considered, at least in Tibetan exegetical traditions, to teach the tathaagatagarbha - concept. Return to text

8. yadaa na bhaavo naabhaavo mate.h sa.mti.s.thate pura.h | tadaanyagatyabhaavena niraalambaa pra"saamyati | | Return to text

9. Cf. Bodhipathapradiipa, folio 280A of the Derge bsTan 'gyur on the Asian Classics Input Project (ACIP) CD-ROM: slob dpon klu sgrub kyis ni bstan pa'i snying po bshad de | des shes rab kyi pha rol tu phyin pa'i don yod pa dang med pa las 'das pa'i dbu ma chen po'i don thugs su chud cing | mkhas pa gzhan gyi rgyud la yang de ltar gsungs so | | de ltar bla ma byang chub bzang po dang | rje btzun ku su lu pa yang de ltar dgongs so | | 'phags pa klu sgrub zhal gyi bdud rtzi des | | a'arya de ba zla grags bha bya dang | | zhi ba'i lha dang byang chub bzang po'i bar | | tshim par gyur ba bdag la'ang cung zhig 'thor | | de ltar gtan tsigs chen po bzhi dag gis | | chos rnams thams cad skye med bsgrub byas te | | sngon gyi slob dpon rnams kyi rjes 'brangs nas | | dbu ma chen po'i grub mthar gnas par bya | Return to text

10. In the colophon to his dGongs pa rab gsal commentary on the Madhyamakaavataara, Tsong kha pa refers to himself as "a yogi of the Great Madhyamaka": dbu ma chen po'i rnal 'byor pa mang du thos pa'i dge slong shar tzong kha pa blo bzang grags pa'i dpal gyis, 'brog ri bo che dge ldan rnam par rgyal pa'i gling du sbyar ba'o (Source: ACIP CD-ROM). Return to text

11. Williams, p. 30: taddu.hkhena na me baadhety ato yadi na rak.syate | naagaamikaayadu.hkhaan me baadhaa tat kena rak.syate | | Tib. reads: gal te de la sdug bsngal bas | bdag la mi gnod phyir mi bsrung | ma 'ongs pa yi sdug bsngal yang | gnod mi byed na de cis bsrung | | Return to text

12. Williams, p. 75. Skt: kalpitam. bhaavam asp.r.s.tvaa tadabhaavo na g.rhyate | tasmaad bhaavo m.r.saa yo hi tasyaabhaava.h sphu.tam. m.r.saa | | Tibetan reads: brtags pa'i dngos la ma reg par | de yi dngos med 'dzin ma yin | de phyir brdzun pa'i dgnos gang yin | de yi dngos med gsal bar brdzun | | Return to text

13. Michael J. Sweet makes brief mention of this verse in his article "Bodhicaaryaavatara 9:2 as a Focus for Tibetan Interpretations of the Two Truths In the Prasaa.ngika Madhyamaka." (Journal of the International Association of Buddhist Studies 2.2 (1979): 79-89). Return to text

14. BCA 138 | Tib. 139: pramaa.nam apramaa.nam. cen nanu tat pramitam m.r.saa | tattvata.h "suunyataa tasmaad bhaavaanaam. nopapadyate; gal te tshad ma tshad min na | des gzhal brdzun par mi 'gyur ram | de nyid du ni stong pa nyid | sgom pa de phyir mi 'thad 'gyur | | Return to text

15. Cf. BCA 141 | Tib. 140: tasmaat svapne sute na.s.te sa naastiiti vikalpanaa | tadbhaavakalpanotpaadam. vibadhnaati m.r.saa ca saa | | Return to text

16. 9:35: yadaa na bhaavo naabhaavo mate.h sa.mti.sthate pura.h | tadaanyagatayabhaavena niraalambaa pra"samyati | | Return to text

17. P. 98, paraphrasing Mi pham's Nor bu ke ta ka commentary which reads: spros pa thams cad zhi ba'i ye shes la gnas pa'i tshe na khas len thams cad dang bral ba'i dbu ma chen po yin no | Return to text

18. Yeshe Thabkay, professor of philosophy at the Institute of Higher Tibetan Studies in Sarnath, India, notes: "Most of the ancient Tibetan scholars were of the opinion that the view of non-elaborative nonconceptuality was subtler than the view of non-affirming negation (prasaajyaprati.sedha = med dgag) of true (existence). Je Rinpoche (rje Tsong kha pa blo bzang grags pa, 1357-1419) also asserted both these views. In his commentaries (Ocean of Reasoning and Illumination of the Thought, respectively) on the Root Wisdom (=MuulaMadhyamakakaarika of Naagaarjuna), he stated that there is no realization of the mode of existence without prior thought. In order to realize the mode of existence, it is imperative to know the mode of apprehension ('dzin stang) of the lack of true (existence). The view of selflessness as a non-affirming negation has been stated to be the ultimate view. . .However, in his Epistle to the Lord Rendawa. . .concerning the "view," he stated that ordinary beings cannot enter into the actual ultimate (rnam grangs min pa'i don dam) at first. In the beginning one should have clinging to the nominal ultimate (rnam grangs pa'i don dam) of non-inherent dependent arising and engage in analytical meditation whereby one will perceive the actual ultimate. In order to establish the actual ultimate, it is not feasible for it to be spoken, listened to or heard" ("The Four Assertions: Interpretations of Difficult Points in Prasangika Maadhyamika"; Tibet Journal 17.1 (1992), p. 4). Return to text

19. I was unable to locate a copy of rGyal mtshab's rGyal sras 'jug ngogs commentary on the BCA, so I have attempted to quote the gloss from memory. Return to text

20. Cf. Tarkajvala in the ACIP CD-ROM at \TEXTS \BYAUTHOR \BAVAVIVE\TOGEBAR\Td3856e.inc, @5B: dam pa'i don te rnam par mi rtog pa'i ye shes dam pa'i don yin pas dam pa'i don to | yang na don dam pa dang mthun pa ste don dam pa rtogs pa dang rjes su mthun pa'i shes rab la don dam pa de yod pas don dam pa dang mthun pa'o. The latter reference here is to shes rab (praj~naa), not ye shes (j~naana); but the analogy is, I think, appropriate. Return to text

21. sa.mtaana.h samudaaya"s ca pan-kti senaadivan m.r.saa | yasya du.hkham. sa naasty asmaat kasya tatsvam bhavi.syati | | asvaamikaani du.hkhaani sarvaany evaavi"se.sata.h | du.hkhatvaad eva vaaryaa.ni niyamas tatra kim. k.rta.h | | du.hkham. kasmaan nivaaryam. cet sarve.saam avivaadata.h | vaaryam. cet sarvam apy evam na ced aatmaapi sattvavat | | Return to text

22. Cf. Lam rim chen mo (425a in ACIP CD-ROM edition): sems can thams cad 'khor bar 'ching ba ni lhan skyes kyi ma rig pa yin pa'i phyir dang, kun brtags kyi ma rig pa ni grug ( read grub) mtha' smra ba de dag kho na la yod pas 'khor ba'i rtsa bar mi 'thad pa'i phyir ro | | 'di la nge spa (read nges pa) dmigs phyed par rnyed pa shin tu gal che'o Return to text

23. | snang ba rten 'brel bslu ba med pa dang | | stong pa khas len bral ba'i go ba gnyis | | ji srid so sor snang ba de srid du | | da dung thub pa'i dgongs pa rtogs pa med | Return to text


Updated: 3-5-2000

Return to "Book Reviews"

Top of Page