the eternal legacyAn Introduction to the Canonical Literature of Buddhism

s angha r a k sh i t a

windhorse publications

5

PREFACE preface tothis editionSince the publication of The Eternal Legacy in 1985 interest in Bud-dhism has grown steadily in the West. Centres and groups affiliated tothis or that branch of the Buddhist tradition have multiplied, andmore and more people are practising meditation in one form or other.Enterprising publishing houses on both sides of the Atlantic havemoreover continued to bring out books on Buddhism – both popularand scholarly – in ever-increasing numbers.

Much of this secondary literature is of a higher standard than wouldhave been the case forty or fifty years ago. Nonetheless, for the seriousstudent of the Dharma an acquaintance with the Buddhist scripturesremains indispensable. The difficulty is that the Buddhist scripturesare vast in extent and varied in content and exist in several canonicallanguages. Where shall the student begin? How is he to gain an over-view of the whole field of what is known as the Buddhavacana orword of the Buddha?

It was with such questions in mind that I wrote The Eternal Legacy. Inthe preface to the first edition of the work I expressed the hope that itwould soon be superseded by a more adequate treatment of its greatsubject by someone better qualified than myself. Though I have nowwaited twenty years, that hope has not been fulfilled. I have thereforedecided to bring out this new edition of The Eternal Legacy.

Since the original publication of the work, translations of many Bud-dhist texts have appeared. Among the more important of these areBhikkhu Bodhi’s Middle Discourses of the Buddha (1995) and ConnectedDiscourses of the Buddha (2000), Thomas Cleary’s The Flower Ornament

15

Scripture, and some of the volumes published in the Bukkyõ DendõKyõkai English translation series.

I would like to thank Windhorse Publications for agreeing to bringout a new edition of The Eternal Legacy, and Dharmachari Shantavirain particular, for his help in correcting errors and regularizingdiacritics. May this work help awake students of Buddhism to theunparalleled riches of that legacy!

SangharakshitaMadhyamalokaBirmingham14 October 2005

x v i T H E E T E R N A L L E G A C Y

16

B U DDH I S M A N D L A NGU AG E

1buddhism and languageBefore essaying a rapid survey of Buddhist canonical literature wemust briefly discuss our principal terms. By canonical literature ismeant the written records of the Buddhavacana or living word of theBuddha, or what purports to be such, whether original or translated,or what is traditionally regarded as such by the Buddhist communityor any section thereof. The whole of the vast derivative literature, inthe form of commentaries and expositions, is thus excluded from ourpurview.

Now the term ‘word of the Buddha’, and therefore the expression‘canonical literature’ also, can be understood either in a wider primarysense, or in a narrower secondary one, depending upon our definitionof the word ‘Buddha’. If we mean by Buddha simply the state ofSupreme Enlightenment by whomsoever experienced, then byBuddhavacana is to be understood any expression, or better reflec-tion, of this transcendental state in the medium of human speech. If,on the other hand, Buddha means the historic Buddha Gautama, theinitiator of the spiritual movement now known as Buddhism, thenBuddhavacana will be confined to the literary record of the sayings ofthis teacher. Buddhism as a whole tends to oscillate between the twoextremes. Even the Theravãdins, who are committed to a pedanticallynarrow and rigid doctrine of Buddhavacana, include in their Tipiìakadiscourses which, though delivered by disciples, are regarded asBuddhavacana inasmuch as the Master had approved them, thusmaking them, as it were, his own. Conversely, the Mahãyãna, whichin principle maintains that ‘Whatever is well said is a word of the

17

Buddha’,1 in practice certainly hesitates to accept as such any teachingthat conflicts with the scriptures.

In whichever way it may be interpreted, Buddhavacana consists ofan assemblage of words in a particular language or languages. Thisintroduces the extremely important question of the relation of Bud-dhism to language in general which, also, can be understood in vari-ous ways. Philosophically, the question is that of the relation betweenthe purely spiritual import of the teaching – ultimately coinciding withthe transcendental state of Enlightenment itself – and its conceptual-cum-verbal formulations: historically, that of the language spoken byGautama the Buddha. It will be convenient to deal with these twosenses in reverse order, proceeding from the narrower to the broaderone.

Modern Theravãdins are fond of making such statements as ‘theBuddha taught in Pãli’ or ‘Pãli is the language of the Buddha’. Theproblem of what linguistic medium the Buddha adopted in communi-cating his teaching to mankind does not, however, admit of sostraightforward a solution. To begin with, Pãli is not the name of a lan-guage at all. The word means, literally, ‘a line, a row (of letters)’ andthus, by extension of its meaning, ‘the (canonical) text’. Early Westernstudents of Theravãda literature, finding in the commentaries expres-sions such as pãlinayena, ‘according to the (canonical) text’, took theword for the name of the language of the texts and, through their writ-ings, gave currency to this misunderstanding. According to Thera-vãda tradition the Buddha spoke Mãgadhî which, since the Tipiìaka isregarded as a verbally faithful record of his teaching, for them alsodesignates the language of the canonical texts. In uncritical usage,therefore, Pãli and Mãgadhî have become synonymous, both of themnow being applied by the Theravãdins indiscriminately to theBuddha’s personal language and the language of the Tipiìaka.

But even to say that the Buddha spoke Mãgadhî does not really helpus. Mãgadhî is the language of Mãgadha just as Spanish is the lan-guage of Spain, and, in the absence of independent literary records inthat tongue, to tell us that the Buddha spoke Mãgadhî leaves us nowiser than we were before. Though born among the Šãkyas, who werefeudatory to the kingdom of Košala, the Buddha spent much timeafter his Enlightenment in the adjacent kingdom of Mãgadha. Thelanguage of Mãgadha, or ‘Mãgadhî’, was therefore undoubtedly hisnormal means of communication within that area. When in Košala he

2 T H E E T E R N A L L E G A C Y

18

must have spoken Kosalese. Being enlightened, he was exempt fromlinguistic prejudice, and his attitude, exemplified by a well-knownepisode, was tolerant and practical:

Two monks [it is related] of fine (cultivated) language and fine(eloquent) speech, came to the Buddha and said: Lord, heremonks of various (names, clan-names, races or castes, andfamilies) are corrupting the Buddha’s words by (repeatingthem in) their own dialects. Let us put them into Vedic(chandaso ãropema). The Lord rebuked them: Deluded men,how can you say this? This will not lead to the conversion ofthe unconverted.… And he delivered a sermon andcommanded (all) the monks: You are not to put the Buddha’swords into Vedic. Who does so would commit a sin. Iauthorize you, monks, to learn the Buddha’s words each in hisown dialect [sakkãya niruttiyã].2

As Edgerton points out, it is clear from this passage that in addition toSanskrit, the language of the upper classes, there existed a number ofpopular and more or less mutually intelligible Middle Indic dialects(among them Mãgadhî and Košalese), in one or more of which theBuddha himself was accustomed to preach, and that it was in thosedialects, therefore, that the monks were to learn, recite, and (accord-ing to the Chinese versions) to disseminate the Buddhavacana.3 In thisway the teaching, instead of being confined to a Sanskrit-educatedelite would, as befitted its universal character, be accessible to all.There was no question of compiling a single standardized version ofthe teaching in a learned tongue, such a procedure being expresslyprohibited. The freedom which the Buddha had allowed his followerspromoted, after his parinirvãœa, the growth of parallel versions of theteaching, first in different local vernaculars and afterwards in differ-ent languages. The oldest and most authentic portions of the ‘Pãli’Tipiìaka are based, ultimately, on one of these versions, being a liter-ary recension of a Middle Indic version originating not in Mãgadhabut somewhere in western-central India.

The insistence of the Theravãda that the Buddha spoke Pãli, the lan-guage of the Tipiìaka, stems less from ignorance of historical facts thanfrom a doctrinal misunderstanding. This misunderstanding, which isof the essence of the Theravãda, consists in the belief that the importof the teaching is inseparable from its original conceptual-cum-verbal

B U DDH I S M A N D L A NGU AG E 3

19

formulations, or what is believed to be such. Hence sweepingpronouncements such as ‘It is impossible to understand Buddhismproperly without studying Pãli,’ or ‘How can the Tibetans be real Bud-dhists? They don’t know Pãli.’

Reinforced by the belief that in the Tipiìaka they possess the onlycomplete and accurate record of the ipsissima verba of the Buddha, thisliteralistic attitude has given rise to that spirit of bigotry, exclusive-ness, and dogmatic authoritarianism for which some modernTheravãdins are notorious. Yet such an attitude is clearly incompat-ible with a number of passages in the Tipiìaka itself, including the onequoted. From the latter it is, indeed, obvious that according to theBuddha the spirit of his teaching, far from being dependent on anyparticular form of words, could be given equally valid expression inlanguages other than the one in which it had originally beenpropounded.

Buddhavacana was not to be identified exclusively with any one ofits linguistic versions. Hence for Buddhists there can be no scripture,no canon, in the sense of a single finally definitive, universally authori-tative text of the teaching such as the Bible constitutes for Christiansand the Koran for Muslims. The word of the Buddha, it must beemphasized, has from the beginning been extant in a multiplicity ofalternative versions – some more and some less complete – no one ofwhich is a priori more reliable than the rest, or can claim superiorityover them on any grounds other than that of greater depth and com-prehensiveness of content. This is not to deny that early versions ofthe teaching, especially when their language approximates to the lan-guage used by the Buddha (assuming this to be known), will alwayspossess a special historical significance. They will obviously be ofgreater help, moreover, in reconstructing the original form of histeaching than the later versions. What we deny, and deny emphatic-ally, is that by an extraordinary coincidence the language used by theBuddha (whatever it may have been) happens to be intrinsically morecapable of conveying his meaning than any other and that, therefore,a knowledge of the letter of the Dharma is indispensable to an under-standing of its spirit. Indeed it has been suggested, by a close and criti-cal student of the Tipiìaka, that the Buddha found the linguisticresources of his day inadequate, being in particular hard pressed forwant of a stronger word for ‘will’ than the feeble cetanã.4

4 T H E E T E R N A L L E G A C Y

20

Taking this as a starting point, one might even argue that classicalChinese, or modern English, being more highly developed languages,are intrinsically more capable of giving expression to the spirit of Bud-dhism than ancient Middle Indic or medieval Pãli. Some do, of course,maintain that it is impossible to translate Buddhist texts satisfactorilyinto modern European languages. This is to confuse fidelity to thespirit with capacity to reproduce the letter of the Buddha’s teaching.Moreover, were it in reality impossible to disengage the former fromthe latter and give it an independent expression it would mean, ineffect, that the Buddha’s spiritual experience, far from transcendingthought and speech, had on the contrary been conditioned by them.Thus his Enlightenment would be no enlightenment at all.

Contradictions of this sort can be precluded only by recognizing,once and for all – as the Mahãyãna has done – that the spirit of theteaching is capable of expressing itself in a variety of forms, no one ofwhich, however authentic or however excellent, is perfect or final, orcan possibly exhibit in full the infinite riches of its transcendental con-tent. The Buddha himself, as one might have expected, was as keenlyaware of the limitations of words in respect of spiritual reality as thepoet Marlowe was in respect of sensuous beauty:

If all the pens that ever poets heldHad fed the feeling of their masters’ thoughts,And every sweetness that inspired their hearts,Their minds, and muses on admired themes;If all the heavenly quintessence they stillFrom their immortal flowers of poesy,Wherein, as in a mirror, we perceiveThe highest reaches of a human wit;If these had made one poem’s period,And all combined in beauty’s worthiness,Yet should there hover in their restless headsOne thought, one grace, one wonder, at the least,Which into words no virtue can digest.5

Evidence of the Buddha’s awareness is provided by the list of ten‘inexpressibles’ (avyãkòtavastuni, Pãli avyãkatavatthuni), according towhich it is impossible to declare (1) whether the world is eternal or not,(2) whether the world is finite (in space) or infinite, (3) whether the

B U DDH I S M A N D L A NGU AG E 5

21

Tathãgata exists after (physical) death, or does not, or both, or neither,and (4) whether the soul is identical with the body or different from it.6

Moreover, in communicating his spiritual experience even a medi-ocre religious teacher has a definite advantage over the poet, howevergifted. By virtue of the very conditions of his art the poet is entirelydependent upon words. The religious teacher, on the other hand, cansupplement any deficiencies of language – whether intrinsic or due tohis own inadequate command over that medium – by the directimpact of his personality on the hearts and minds of his auditors,whether through looks and gestures, or in ways still more subtle andindefinable. In the case of the Buddha, the perfectly enlightenedTeacher of teachers, this impact is out of all proportion to either thenumber or the actual import of the words spoken. It may, indeed, beentirely independent of words. The Dhyãna (Ch. Ch’an, Jap. Zen)School, which claims to represent ‘a special transmission outside thescriptures’,7 is believed to have originated from an unverbalized com-munication of this kind. According to a late Chinese legend:

Šãkyamuni was once engaged at the Mount of the HolyVulture in preaching to a congregation of his disciples. He didnot resort to any lengthy verbal discourse to explain his point,but simply lifted a bouquet of flowers before the assemblage,which was presented to him by one of his lay-disciples. Not aword came out of his mouth. Nobody understood the meaningof this except the old venerable Mahãkãšyapa, who quietlysmiled at the master, as if he fully comprehended the purportof this silent but eloquent teaching on the part of theEnlightened One. The latter perceiving this opened his gold-tongued mouth and proclaimed solemnly, ‘I have the mostprecious treasure, spiritual and transcendental, which thismoment I hand over to you, O venerable Mahãkãšyapa!’8

The Tibetan branch of the Vajrayãna, no doubt following Indian trad-itions, reckons three different ‘lineages’ of the Dharma correspondingto the three different planes on which its transmission may take place.On the highest, the purely spiritual plane, that of the mind-lineage ofthe Jinas (or Buddhas), the transmission consists in a communicationof spiritual experience directly from the heart or mind of the enlight-ened master to the heart or mind of the disciple without recourse tolanguage or gesture. On the intermediate plane, that of the sign-

6 T H E E T E R N A L L E G A C Y

22

lineage of the Vidyãdharas or ‘Tantric initiates of high spiritual attain-ment’, it takes place by means of gestures only (according to some,through study of the written, as distinct from the spoken, word). Itwas in this way, apparently, that the Dharma was transmitted toMahãkãšyapa. Finally, on the third and lowest plane, that of theword-lineage of the ãcãryas or ‘teachers profoundly versed in thescriptures’, the Dharma is transmitted orally by means of language.The treasure handed over to Mahãkãšyapa and the two higherVajrayãna lineages represent, in different ways, the living spirit of theDharma which, unless it vivify the letter, the letter is dead.

In studying the canonical literature it is important to remember thatBuddhism is not to be understood by words alone, not even whenthose words are authentically the Buddha’s. If misunderstandings areto be avoided, it must be studied, not in isolation, but with reference tothe tradition of spiritual experience out of which it sprang, to which itreturns, and to which it all the time belongs. Moreover, besides thefact that the teaching expresses itself in a multiplicity of forms, itshould also be remembered that before its reduction to writing thecanonical literature existed in the form of oral tradition.

B U DDH I S M A N D L A NGU AG E 7

23