Aaa

works of the earlier Pali Canon was made from some other ; dialect, even if the other well-known arguments against such a >• notion did not exist

To be able to pass on textual complexes as large as these by word of mouth while still maintaining an acceptable level of accuracy requires a special system, and it is precisely this that is attested to by the tradition that there existed specialists in the skill of recitation (bhanaka), which represented a parallel with the methods of transmission used by the Vedic schools. .To a certain extent the Buddhist practice of oral transmission continues to exist side by side witht the written even today, especially in Burma.

Thus, there cannot be a shadow of doubt - and at this point I believe I can pass from asking a question to making a flat • assertion - that what we are dealing with in the early period is an oral tradition. Indeed, literary historians have long since determined with great exactitude the effect of a long oral tradition on the form of literary texts (see G. von Simson, ‘Zur Phrase yena . . . tenopajagama/upetya und ihren Varianten im buddhistischen Kanon’, Beitr&ge zur lndienforschung, pp.479-88).

Now that we have come to this conclusion we can answer more accurately the question as to the nature of the ‘transmission’ of the texts. If we look for remnants of earlier linguistic forms in the available texts, we must do so bearing tn mind the characteristic features of oral tradition; to interpret the differences between the versions of the Buddhist text we must bring to bear an entirely different methodological approach from that which we would use, say, in comparing the versions of the Asokan inscriptions, even- though these inscriptions

belong to the same linguistic and chronological domain.

Thus, in seeking out traces of earlier linguistic forms, we must heed the principle already formulated by S. Levi for* our own question and later applied successfully by Hermann Berger (in Zwei Probleme der mittelindishcen Lautlehre , Munich 1955) to the solution of a large number of individual problems; namely, we must always look for the specific conditions which have led to the. preservation of forms from an alien dialect in these linguistic monuments. This precept applies whenever we see in the language in question not simply a ‘hybrid dialect’ but a specific linguistic forpi into which the given textual material has been ‘transformed’ or ‘transmitted’. We have accepted as a premise that this applies to Pali. Thus H. Berger has designated as ‘Magadhisms’ (op. cit ., p.15 ff.) such linguistic doublets as occur only or chiefly in stereotyped series of synonyms (e.g. kiqha along with kanha), or which are found in verses whose metrical structure would be distorted if the normal Pali form (e.g. kiccha for the ‘Magadhism’ kasira) were used. Both premises are in keeping with the special demands of oral transmission and oral conversion.

I should like to cite as an additional example the use of bhikkhave and bhikkhavo in the earlier prose sections of the Pali Canon. We find the ‘Magadhism’ bhikkhave in the actual sermon of the Buddha, while the vocative bhikkhavo occurs in the introductory formula. The text of the Majjhima Nikaya begins as follows:

tatra kho Bhagavd bhikkhu amantesi: bhikkhavo ti. bhadante ti te bhikkhu Bhagavato paccassosum. Bhagavd etad avoca: sabbhadhammamulapariyayam vo bhikkhave desessami _

Buddhist Studies Review 8. 1-2 (1991) - Bechert

The form bhikkhave is thus established as a specific usage | in the Pali text which can be explained as a way of recalling the j actual speech of the Buddha. Once such a standard procedure has been devised, it could be applied to newly created texts without further ado, and thus the occurrence of this ‘Magadhism’ would tell us nothing about the original language of the text in j question. On the other hand, it would explain why we find only bhikkhavo throughout the verses of the Suttanipata, which otherwise is so full of ‘Magadhisms’.

The forms in -e (for Sanskrit -as), which of course were determined very early to be Magadhisms in the Pali Canon (Kuhn, Beitrage, p.9; V. Trenckner, Pali Miscellany, Copenhagen 1879, p.75 etc.), also provide exemplifications of this • methodological principle, which are plausible in other ways. If we refer to the list of such cases compiled and expanded by H. Luders ( Beobachtungen, §§ 1-24), we' find that - except for set expressions to which e.g. seyyatha and yebhuyyena owe their adoption into Pali - the causes for the preservation of such forms are generally speaking misunderstandings in transmission. This applies also to those passages in the Patikasutta (Luders, o p. cit., § 5) that can obviously no longer be correctly understood. As with seyyatha and bhikkhave, the easily remembered formulation - and thus the existence of a stereotyped mode of expression - may have contributed significantly to the preservation of the -e in the passage of the Sakkapanhasutta (Geiger, op. cit-, § 80; Luders, op. cit., § 6) and the Sunakkhattasutta (Trenckner, op. cit n p.75; Luders, op. cit., §7).

On the other hand, this very form, provides an example of how we can go astray if we rely exclusively on the grammatical

Buddhist Studies Review 8, 1-2 (1991) - Bechert

form and do not pay attention to the context. Luders, for instance, explains ( Beobachtungen, § 8) the nominative in -e in the language of the heretics in the Samannaphalasutta as ‘Magadhisms’, although it is difficult to perceive why an historical peculiarity of the language of the Buddha should be preserved in the language of the heretics only, while it is not found in the speech of the Buddha himself. I have attemp ted to explain these forms and related passages in the JStaka as ’Sinhalisms’, i.e. as forms first adopted in Ceylon from the local vernacular to characterise the uncultivated patois of the heretics (‘Uber Singhalesisches im Palikanon’, WZKSO 1, 1957, pp.71-5X This implied that these forms were inserted in the text in early Ceylon during the period of oral tradition. K.R. Norman disagreed (‘Pali and the Language of the Heretics’, Acta Orientalia 37,1976, pp.113-22), but I am not at all convinced by his arguments which I shall discuss elsewhere. In any W e may not consider these forms as ‘Magadhisms 4 in the usual sense of the term. They do not seem to be residua from the fang na y of the oldest tradition, but are forms which came into the text later, even though they look like ‘Magadhisms’ purely from the standpoint of form. If, on the other hand, the ending -ase in the nominative plural, which occurs in the verses, was not transformed into -aso in the Pali texts (with one or two possible exceptions under peculiar conditions only), it was for the reason that the form in -aso was not usual in ’genuine’ Pali and thus there was no point in substituting it

I am still in agreement with a thesis advanced by H. Berger (op. ciu P-15) that, jn general, forms like pure which appear in the traditional Pali‘texts should not be regarded as ‘Magadhisms’, although -e appears for -ah instead of *puro which the laws of Pali phonetics would lead us to expect; hence Berger’s comment

Buddhist Studies Review 8 , 1-2 (1991) - Bechert

Buddhist Studies Review 8, 1-2 (1991) - Bechert

(ibid.), ‘It is hard to understand why the Pali translator would have neglected to put this particular word, common as it is, into the corresponding western form while they never made the same slip with other adverbs (tato, bahusoe tc.). This must be a case of formation by analogy (and indeed with a significance corresponding to that of agge and similar forms, cf. Karl Hoffmann in Berger, op. cit., p.15, n.6). The same holds true for Pali sve or suve (Skt. svah). Here again we must not allow ourselves to be misled by a merely apparent congruence with the Eastern dialect.

Thus we can clearly see the general applicability of the principle enunciated above to the example of the occurrence of -e for •as in Pali, and, as we proceed to exclude, on the basis of convincing arguments, forms like these, which are not ‘Magadhisms’, we can then turn to working out the complex of true ‘Magadhisms* which remains. The example has also shown us how important it is to take-note of the further destinies of the transmitted texts. Aspects of the history, of the transmission of the Pili Canon have been examined recently by O. von Hinuber, K.R. Norman and other scholars. Various orthographic and grammatical peculiarities result from the influence of the vernaculars of the countries in which the texts were handed down, or from the influence of Sanskrit.

These basic considerations also hold true for that form of the language known to us from the ‘Gandhari-Dharmapada’ (J. Brough, The Gandhari Dharmapada, London 1962>, this was tentatively identified by F. Bernhard (‘Gandhari and the Buddhist Mission, in Central Asia’, Ahjali. O.H. de A. Wijesekera Felicitation Volume, Peradeniya 1970, pp.55-62) and even earlier by H.W. Bailey (‘Gandhari*, BSOAS 11,1946, pp.764-97) as the

language of the Canon of the Dharmaguptaka school before its Sanskritisation. (Cf. also J.W. de Jong, A Brief History of Buddhist Studies in Europe and America , Varanasi 1976, pp.62f.).

The situation is more complicated in the case of the texts in Buddhist Hybrid Sanskrit*. There was an indigenous term for this language, viz. ar$a. It is used in Kaumaralata’s grammar, as has been pointed out by H. Luders (Philologica Indica, .Gottingen 1940, pp.686 f., 693 f., 713 ff.) and more recently recalled by D. Seyfcrt Ruegg (‘Allusiveness and Obliqueness in Buddhist Texts', Dialectes dans les literatures indo-aryennes ed. C Caillat, Paris *1989, p.285 f.) 2 . Most of these texts were written in various forms of Middle Indie before Sanskritisation. We can proceed on the basis of the traditions of the themselves, that - depending on which sect was involved - they are based on different languages. The familiar tradition that , four different languages were used by the four main sects (Lin Li-kouang, L'Aide-memoire de la vrai Id, Paris 1949, ppJ75-81) is not, of course, an actual description of the historical facts, yet we can perceive that it represents a recollection of the linguistic differences of the various versions of the canonical texts. Akira Yuyama has presented a detailed critical discussion of this

uiai mis

been omitted from the Sanskrit-Worterbuch der buddhistischen Texte aus den Turfan-Funden even though the term, as noted by Luders, is attested in the 'Turfan" collection*. However, this use is found in grammatical literature only but not in the corpus of texts to-be evaluated in this dictionary. The guidelines governing the choice of material to be included in this dictionary were explicitly approved by Seyfort Ruegg in his review in JAOS 106 (1986) p.597, so that his criticism concerning the entry for arsa is not justified.

Buddhist Studies Review 8. 1-2 (1991) - Bechert

tradition (*Bu-ston on the Languages Used by Indian Buddhists at the Schismatic Period’, Die Sprache der altesten buddhistischen Oberlieferung, pp.175-81). Accordingly, the thesis once expressed by F. Edgerton concerning an ‘essential dialectic unity’ of the Prakrit underlying the hybrid Buddhist Sanskrit (see, e.g. Buddhist Hybrid Sanskrit Grammar, § 1.80) no longer requites any specific refutation.

Our task now lies in differentiating between the various strata of dialectic change. There is good reason to believe that Sanskritisation began when the texts were committed to writing, and, we can be helped along by the fact, well-known from the lessons of textual criticism, that textual changes occurring in the course of written transmission come about in a different manner from those developed in an oral tradition. Sanskritisation itself is known to have been a multi-stage process, and we are much better informed about it than we are about the previous stages of textual development, especially since we actually have available to us earlier versions of many texts which are closer to the Middle Indie variants as well as later, more strongly Sanskritised versions. Naturally we are speaking here only of the Buddhist works in Sanskrit which are actually .based on a Middle Indie original. Various other Sanskrit Buddhist works were written from the beginning in the so-called ‘hybrid dialects’; for a discussion of this question, see C. Regamey, ‘Randbemerkungen zur Sprache und Textuberlieferung des Karandavyuha’ ( Asiatica . Festschrift Friedrich Weller, Leipzig 1954, pp.514-27).

As has already been demonstrated by the foregoing discussions, the question of the relationship of the individual versions to the earliest tradition must be viewed in connection

Buddhist Studies Review 8, 1-2 (1991) - Bechert

with the problems of the history of the early Buddhist sects, and we must also enquire into their localisation. The home of Pali, for example, cannot be determined exclusively on the basis of linguistic arguments, but only with due regard to the carly history of the Theravada. Consideration of that history made it possible to classify Pali as the language of Vidisa (cf. H. Frauwal!ner, The Earliest Vinaya, Rome 1956, p.18 ff.), a determination which would not have been possible on the basis of current arguments from the standpoint of historical linguistics, but which nevertheless was in close agreement with the results of philological research. Local factors also help to explain the noteworthy similarities between Pali and the language of the texts of the Lokottaravadins, which the history of the formation of the sects leaves quite obscure.

Yet we must still keep in mind the linguistic aspects of the problem. The comparison of the language of the early Buddhist texts with the language of the Asokan and other early Prakrit inscriptions has been carried out in the minutest detail. Indeed, much of the research has, if anything, been undertaken too systematically. For example, we can only view with the greatest scepticism any attempts to come to conclusions about pronunciation on the basis of orthography, since we must never lose sight of the broad spectrum of possible divergences between orthography and pronunciation that we are familiar with from our knowledge of the development of other languages and from examination of later stages in the evolution of the Indie languages themselves.

Similarly, the questions of the conditions necessary for the emergence of a written language must be approached by methods which are predominantly linguistic. Fortunately wc

Buddhist Studies Review 8, 1-2 (1991) - Becherl ,

possess a number of examples from other areas - such as the origin of the written form of the Romance languages - for which we have developed an extremely useful research apparatus. The question of the language of the earliest Buddhist tradition and its progressive development into the corpus of material as it stands today must undoubtedly be viewed as part of the formation of standardised (and therefore also in certain ways ‘hybrid’) languages during the developmental stages of Middle Indie, which ultimately came to be written languages. Moreover, the use of Middle Indie languages in the earliest Indian inscriptions, which of course constitute the oldest written evidence of the Indo-Aryan languages, suggests the hypothesis that we have here the earliest written Indie language, to which, however, the established tradition of a language of priests and scholars that was transmitted orally at first and nevertheless became standardised down to the last detail - i.e. Sanskrit stands in the same relationship as Latin does to the written Romance languages. We can infer from the passage in the Vinaya that we have mentioned, and also from the actual development of language, that originally, and indeed in deliberate contrast to the Brahmanic tradition, the Buddha had definitely not been striving to bring about a linguistic standardisation to be used in the propagation of his teachings.

Does it not seem reasonable, then, to assume that the earliest tradition actually consisted of a linguistic multiplicity, and that a specific ‘language of the earliest Buddhist tradition’ does not exist at all? In view of all this there would hardly seem much point in continuing to look for this language; instead we should redirect the thrust of our enquiry towards the process of ’standardisation’ of the linguistic form of the tradition as such. In this connection it would be quite helpful if we could answer

Buddhisi Studies Review 8, 1-2 (1991) - Bechert

the question as to how the traditional canonical texts of the Jains developed up to the point when they took definitive form, and how the Ardhamagadhi of the Svetambara texts actually originated. The significant differences between the language of the canonical prose of the Pali Canon and the language of the early verses give- rise to the further question as to whether or not a poetic language existed in Middle Indie, which was possibly supra-regional in use but in certain places may have been subjected to a process of assimilation with local languages, as Helmer Smith conjectured. Whatever answers we finally come up with to all these questions, it would seem imperative, in any case, always to keep in mind the wide variety of points of view and be wary of supporting just one principle argument

Considered in- isolation and viewed only with reference to individual linguistic phenomena, this question might well appear to be one of those abstruse problems of detail in a highly specialised science the solution to which touches on the progress of that science as a whole only with reference to a narrowly limited issue. If, however, we view our question in its broader ramifications, its answer will prove to be an important element in the task of elaborating an accurate understanding of the entire linguistic, literary and religious development in India during the fifth to the first century B.C.E

DANDAPANI

As a general principle, the Buddha always spoke to the point and only taught Dhamma to those capable ( bhabbo ) of understanding. He did not waste words but spoke only what was appropriate on any particular occasion according to the capacity of his audience. Then, it may be asked, what about the concise teaching to Dandapani (‘Stick-in-Hand’) the Sakyan (Madhupindika Sutta, M 18) which was quite beyond his comprehension? The whole episode was subsequently related to the bhikkhus and was beyond them too until explained by Mahakaccana. “However, there are a number of indications in this story that make one suspect Dandapani was not a ‘real’ person at all in the usual sense. Perhaps we should regard him as a ‘type’; a hypothetical case, employed by the Buddha as a teaching device. In fact, looking at this episode closely, Dandapani was actually a Mara-like figure. Mara the ‘Evil One’ can also be viewed symbolically, as a psychological entity - in a sense the personification of the ego and sensual attachments, and an obstacle to be overcome before enlightenment is attained. For the arahant Mara poses no problem; he is always recognised immediately and is, accordingly, sent packing. As in a great number of Mara episodes, Dandapani appeared when the person, in this case the Buddha, was in solitude and in an open place, ‘under a tree’. Like Mara he was always roaming about seeking a ‘victim’ to debate with. Again, as so often with Mara, he assumed an arrogant stance, leaning on his stick, when putting his question. Finally, he departs, like MSra once did when defeated, with a wrinkled brow and leaning on his stick (see Mara Samyutta, S I, p.118). Mara defeated and recognised departs dejected, downcast and uncomprehending.-

Buddhist Studies Review 9, 2 (1992) - Manne

THE DiGHA NIKAYA DEBATES’.

DEBATING PRACTICES AT THE TIME OF THE BUDDHA 1

Joy Manne

Eighteen out of thirty-four suttas in the DIgha Nikaya (D 1-13, 23-25, 28, 31) are debate suttas, that is to say that each of these has all or most of the following features', a central character, most usually the Buddha, and a statement of his credentials; an adversary, and a statement of his credentials a description of a location that functions to set the scene and the atmosphere; an audience; a greeting ceremony; a challenge; a refutation of the adversary’s position; the establishment of the Buddhist position; a hypothetical case history 2 3 ; a surrender, in the form of an acceptance formula, by the adversary; a reward*. Witzel has already drawn attention to similarities between the debates in the Vedic texts and those in the P&li texts, notably on the

1 These investigations were supported by the Foundation for Research in the field of Theology and the Science of Religions in the Netherlands, which is subsidised by the Netherlands Organisation for the Advancement of Pure Research (Z.W.O.), and constitute Chapter IV of my doctoral dissertation. ‘Debaies and Case Histories in the Pali Canon’ (Utrecht 1991).

2; ' Most usually a repetition of S 40-98 of the Samannaphala Sutta, D 2.

3 See J. Manne, ’.Categories of Suita in the Pali Nikiyas and their

implications for our appreciation of the Buddhist' Teaching and Literature

JPTS XV. 1990. pp.29-87 (abbrev. Manne. 1S90X cf.' pp.44-48.

issue of the severed head 4 , on the relationship between the sahadhammika type of questioning ‘which takes place in a kind of open challenge or tournament, (which is) similar to the Vedic brahmodya ’ 5 , and on the similarity of both the anati- prasnya and the sahadhammika questions and the general rules of discussion found in the Vedic and Pali texts 6 . He particularly observes, ‘As often, it is the early Buddhist texts which provide more detailed and useful information. The Pali texts frequently describe in lively and graphic detail what is only alluded to in the Vedic texts which were, after all, composed by Brahmins for Brahmins: one did not have to explain ritual matters of everyday occurrence or of common knowledge to one’s fellow Brahmins or to bralimacarin students . . ,’ 7 . Witzel comments further, ‘Interestingly, the challengers seem to be the best among the various groups of Brahmins (and both Yajnavalkya’s and their personalities require further study)’*.

The Buddhist debates of the DIgha contain information regarding contemporary debating practices, including customs or conventions related to the debate situation, information regarding the types of utterance that were usual in religious

4 M. Witzel, 'The case of the shattered head'. Sludicn zur Indologie l ■;. Iranislik 13/14, 1987, pp.363-415 (abbrev. Witzel, 1987), but see S. Insler. '1 he shattered head split and the Epic tale of Sakuntula', Bulletin d'etudes indiennes 7-8, Paris 1989-90, who lakes a different view of the history of the theme of the shattered head.

5 ' Witzel, 1987, p.408.

6 ‘Both the saccaldriyd and the analtpraina I sahadhammika statements deai with truth, but both do so in a formalised context: cither a discussion ssith a challenger and one or more opponents.' Ibid., p.110.

. 7 Ibid* p381.

8 Ibid , p365.

Buddnlst Studies Review 9, 2 (1992) • Manne *

debate, and criteria for judging success in debate, beyond those that Witzel discusses in his article (by no means all of which have been referred to above). It is the very large number of features in common between Vedic and Buddhist debates that Witzel has drawn attention to in his article* and others that I- have pointed out 9 that permits me to say this. The purpose of this article is to present this material. It is beyond its scope to make extensive comparisons with the Vedic tradition. This article then analyses the Buddha’s debating style and techniques in terms of these conventions and compares them with those of one of his disciples, Kumara Kassapa.

In three of the debate suttas, the Brahmajala (D 1), the Kassapa-Sihanada (D 8) and the Udumbarika-Sihanada (D 25), contemporary debating practices, including customs or conventions related to the debate situation, are specifically mentioned. In the Brahmajala there is information regarding the types of utterance that were usual in religious debate (and the Buddha’s attitude towards them). In the Kassapa-Sihanada a r e the criteria for judging success in debate, and in the Udumbarika-Sihanada the value placed upon discussion between religious practitioners of different persuasions is demonstrated. In these suttas the Buddha is the debater on behalf of the Buddhists. This is the normal state of affairs in the Pali texts, which lends support to Witzel’s observation cited above that ‘interestingly, the challengers seem to "be the best among the

Buddhist Studies Review 9, 2 (1992) - Manne

various grdups of Brahmins,.. .’ 10 . In a fourth sutta, the Payasi (D 23), the wordy Kumara Kassapa takes this role. It is because he is so explicit about his tactics in the discussion that this sutta also provides useful information on debating techniques.

In the Brahmajala Sutta the Buddha criticises the dis¬ putatious habits of brahmans and samanas, narticularly the use of expressions like:

(1) ‘You don’t understand this doctrine and discipline, I do.’ ‘How should you know about this doctrine and discipline?’

‘You have fallen into wrong views. It is I who am right.’

‘I am speaking to the point, you are not.’.

‘Ytou^re putting last what ought to come first, and first what ought to come last’

‘What you have excogitated so long, that’s all quite upset.’

‘Your challenge has been taken up.’

‘You are proved to be wrong.’

‘Set to work to clear your views.’

‘Disentangle yourself if you can’".

Because of the many features in common between the Vedic

10 Witzel, 1987. p365.

11 'Na tvam imam dhamma-vinayam ajanasi, aham imam dhamma-vinayam a j an ami, kim tvam imam dhamma-vinayam ajdnissasi? - Micchd-patipanno tvam asi , aham asmi sammd-patipanno - Sahitam me, asahitan le • Pure vacaniyam paccha avaca, paccha vacaniyam pure avaca - Avicinnan te viparavattam - Arapito te vddo, niggahito *si - Cara vddappamokkhdya , nibbethehi vd sace pahositi t D 8, § 18. Tr. T.W. Rhys Davids. Dialogues of the Buddha I, p!4f. See also his extensive notes.

Buddhist Studies Review 9, 2 (1992) * Manne

and sthe Buddhist debates, the reference to these types of utterance may be taken to indicate that they were in general use in contemporary debating practice.

The expression of criteria for success in debate in the Kassapa-Sihanada Sutta takes the form of a categorical denial, uttered by the Buddha, of a set of criticisms that h& suggests might be made against him by religious wanderers of other sects?. The structure of the sutta. show-, that these criticisms are important: it is the Buddha himself who, unprovoked, first in¬ troduces them and then denies that they can be applied to him. Once again, because of the many other features in common between the Vedic and the Buddhist debates, this suggests that these were genuine contemporary criticisms which accurately re¬ flected contemporary conventions of the debate situation. In this case, however, because Kassapa was a naked ascetic ( acelo ), they may not apply strictly to the Vedic debates 13 . The points that the Buddha disputes provide us, nevertheless, with the criteria of the time for judging and evaluating thf competence of the debater.

The following are the potential criticisms that the Buddha suggests might be made against him: that although he issues his challenge 14 .

12 • thannm kho pan etam Kassapa vijjati yam ahhatitthiya paribbajaka evam vadeyyum D 1 175. 5 22.

13 *JUB (Jaiminiya Upanisad Brahmana] 312 sqi*. expressively stales that such discussions were held only among the Brahmins and Ksatriyas (and Vaisyas?) but not among the $udra& Wittel. 1987, pi410.

14 siha-nbdam nodali - 'utters his lions roar*, 'makes his assertion*, 'issues his challenge*.

Buddhist Studies Review 9, 2 (1992) - Manne

(2) 1. he does this in empty places, and not in assemblies' 5 ,

2. he issues his challenge in assemblies, but he does it without confidence' 4 ,

3. he challenges with confidence,... but people do not ask him questions' 7 ,

4. people ask him questions,, but he does not answer 18 ,

5. he answers their question, ... but he does not win over their minds with his exposition 19 ,

6. he wins over their minds with his expositions . . . but they do not find him worth hearing 70 ,

7. they find him worth hearing but after they have heard him they are not convinced 71 ,

8. having heard him, they are convinced, ... but the faithful make no sign of their belief 77 ,

9. the faithful give the sign of their belief, ... but

15 ten ca kho suhhagare nadati no parisdsuti. D 11 175. parisa - ‘group’, 'assembly'.

16 parisasu ca nadati , na ca kho visarado nadati. Ibid.

17 visarado ca nadati . . na ca kho nam pahham pucchanti . Ibid.

18 pahham ca nam pucchanti . . na ca kho pan dam [NaUnda ed. nesam] pahham putt ho vyakaroti. Ibid,

19 pahhah ca nesam putt ho vyakaroti . . na ca kho pahhassa vcyydkaranena cittam aradhetL Ibid

20 pahhassa ca veyyakarancna cittam drddheti . . na ca kho sotabbarn assa mahhanli. Ibid.

21 sotabbam c f assa mahhartli . . na ca kho sutva pasidanli. Ibid. ' pasldati - 'a mental* attitude which unites deep feeling, intellectual appre¬ ciation and satisfied clarification of thought and attraction towards the teacher*. K~N. Jayatilleke, Early Buddhist Theory oj Knowledge . London 1963, § 655.

22 sutva c’assa pasidanli . . na ca kho pasanna pasanndkdram karonti. Ibid. Presumably this means that they utter no acceptance formula, provide no meals for the bhikkhus, etc.

Buddhist Studies Review 9, 2 (1992) - Mannt

they do not follow the path to the Truth (Nibbana) 23 ,

10. they follow the path ... but they do not succeed 24 .

The Udumbarika-Sihanada Sutta contains a list of criticisms which provide further evidence that a religious leader was required to discuss his views and indeed to put himself before his critics in the public debating arena rather than to remain in solitude. These criticisms are made by Nigrodha, a wanderer iparibbajaka) and not a brahman, against the Buddha. Nigrodha challenges Sandhana, a householder (gahapati ) and lay disciple, on the subject of the Buddha’s habits:

(3) ‘With whom does he talk?

With whom does he engage in conversation?

With whom does he attain wisdom and distinction?

His wisdom is damaged by solitude.

The samana Gotama is outside the assembly.

He does not converse enough.

He busies himself with peripheral matters’ 25 .

He ends his criticisms with the boast: ‘If the Samana Gotama were to come to this assembly, with a single question only could - we settle him; yea, methinks we could roll him over like an empty pot’ 26 .

23 pasanna pasannakaran ca karonti . . na ca nho taihaitaya patipajjanti.

Ibid.

24 tahatlaya ca patipajjanti . . na ca kho patipanna aradhenti . Ibid.

25 . . kcna Samano Cotamo saddhim sallapati? Una sakaccham samapaj- jati? Una pahha-veyyattiyam dpajjati? Suhhagara-hata Samanassa Got amass a panha, aparisavacaro Samano Gotamo. nalam sallapaya . so antamanta eva sevati. D III 38. § 5.

26 Ihgha gahapati, Samano Gotamo imam parisam agaccheyya, eka-pahheri eva nam samsadeyyama , tuccha-kumbhi va nam marine orodheyyamati. D III

Buddhist Studies Review 9, 2 (1992) - Manne

As-thts criticism comes from Nigrodha, whose followers have been criticised for their talkativeness by Sandhana (§ 4), and who will be criticised for the same fault by the Buddha later in the sutta (§ 21), its content is evidently defensive in character. For this reason it might be expected that the Buddha, as he is represented by the composers of the texts, would not take it entirely seriously. As in the Kassapa-Slhanada Sutta, however, these criticisms are given importance in the sutta: the hears Nigrodha’s accusations by means of his clair- audience, and takes them seriously enough to come out of his solitude on the Vulture Peak into the area where the discussion was taking place in o r der to refute them.

Finally, Kumara Kassapa, who is so explicit about what he is doing in the debate situation, by suggesting an earnest desire to conform to standards, provides samples that support the rules in the previously cited suttas. He provides further examples of the techniques a debater was expected to use, and indeed was admired for using These*are supported by examples of similar strategies in debates where the Buddha is the protagonist.

Kumara Kassapa attempts the Buddha’s technique of gradually leading the adversary on ‘by the usual Socratic method adopted in so many of the Dialogues, to accept one self-evident truth after another* 27 , explaining to his adversary:

(4) ‘Therefore, Prince, I will question you in this matter

38. § 5. Tr. T.W. and CAP. Rhys Davids, Dialogues oj the Buddha II, p35.

27 T.W, Rhys Davids’ introduction to the Sonadanda Sulla (D 4). Dialogues I, p.138.

Buddhist Studies Review 9, 2 (1992) - Manne

Buddhist Studies Review 9, 2 (1992) - Manne

and you answer if you please’ 28 .

In the same explicit way he offers a simile:

(5) ‘Well then. Prince, I will make you a simile, for by a simile some intelligent persons will recognise the meaning of what is said* 29 .

The text tells us that Kumara Kassapa was considered a skilled debater. At the end of the debate his opponent says to him, ‘I was delighted, satisfied, by Master Kassapa’s first simile, but I wanted to hear the variety of (his) answers to the question’ 30 .

Tlie suttas above provide information concerning the con¬ ventions, rules and customs connected with the debates that took place between religious leaders of one sect, or their senior followers, and those of another. They refer explicitly to a num¬ ber of debating techniques or strategies. How far. does the Buddha’s performance in the debate suttas conform to these conventions?

In the Brahmajala Sutta the Buddha’s choice'not to express himself in certain ways (see (1) above) is reported, and indeed the Buddha adheres to his standards throughout the Digha debates.

28 Tena hi Rajahna lam yev'etlha palipucchissami. yatha te khameyya tatha

nam vyakareyyasi. D 11 319, 5 5. 0

29 Tena hi Rajahna upaman te karissami. upamaya pi ida‘ ekacce vihnh parish bhasitassa allham ajananli. Ibid, S 9. Tr. Rhys Davids, Dialogues II pOtt.

30 . Purimen evaham opammena bhoto Kassapassa atiamano abhiraddho, api caham imani vicilrani panha-pal ibhanani sotu-kamo ... Dll 332.

The criticisms in the Udumbarika-Sihanada Sutia emphasise certain features of the customs that formed part of the debate situation, notably the expectations placed upon a religious leader, that he should be willing to enter into public debate and discussion. The large number of debate suttas in D alone attest to the Buddha’s conformity to these expectations.

The criteria of the Kassapa-Sihanada Sutta (see (2) above) relate to the conventions of the debate situation. The debater was expected confidently to issue a challenge or make an asser¬ tion to an assembly (see (2), points 1 and 2). The challenge or assertion should be so important (or interesting?) that people wish for further information or elucidation, i.e. they ask ques¬ tions (see (2), points 3 and 4). Questions should be so com¬ petently answered that the attention of the questioner is captured, he appreciates the value of the message, and he be¬ comes so convinced that he makes his convictions publicly maniest (see (2), points 5 - 9). Furthermore, he should under¬ take to follow the path being taught and he should succeed in his efforts, thus proving that the assertions were well-founded (see (2), point 10).

The defeat of and surrender by the adversary is a signi¬ ficant feature of the Buddhist debate suttas as well as of the Vedic debate tradition 31 . It regularly attests to the Buddha's success as a debater. There is, however, only one occasion where the eventual attainment (see point 10 in (2) above) of the

31 ‘In the course of the discussion, participants who do not know the whole truth have to stale this clearly, they must cease questioning. ... and thus declare defeat, or they must even become the pupil of the winner.' Witzel. 1987, p372.

Buddhist Studies Review 9, 2 (1992) * Manne

erstwhile adversary is attested (Kassapa-Slhanada Sutta). The Payasi Sutta adds to the above requirements a point of style: the technique, richly adhered to by the Buddha in the debate . suttas, of furthering one’s argument through the use of similes and analogy.

The seemingly simple conventions of the debate situation are used in a variety of powerful ways.