Buchbesprechungen Comptes Rendus

Author(s): Johannes Bronkhorst
Publisher: Johannes Bronkhorst
Catalog link: https://jainqq.org/explore/269484/1

Page #1
BUCHBESPRECHUNGEN / COMPTES RENDUS 501 BUCHBESPRECHUNGEN / COMPTES RENDUS ON THE METHOD OF INTERPRETING PHILOSOPHICAL SANSKRIT TEXTS: CLAUS OETKE: Zur Methode der Analyse Philosophischer Sutratexte. Die pramana Passagen der Nyayasutren. Reinbek: Inge Wezler, 1991. (Studien zur Indologie und Iranistik, Monographie 11.) 86 pp. ANDREW P. TUCK: Comparative Philosophy and the Philosophy of Scholarship, On the Western interpretation of Nagarjuna. Oxford University Press, 1990. x, 127 pp. Much is being written about hermeneutics these days. Yet few books come out that deal with the specific questions of interpretation that confront the Indologist who studies philosophical Sanskrit texts. The two books under review, however, do deal with these questions, even though it is from altogether different points of view and with completely different results. It will be interesting to study them side by side A.P. Tuck's Comparative Philosophy and the Philosophy of Scholarship presents a survey of modern Nagarjuna studies. The titles of three of the four chapters of the book leave no doubt as to the point its author wants to make. They are: "Nineteenth-century German idealism and its effect on second-century Indian Buddhism"; "Analytic India"; "Buddhism after Wittgenstein". If these titles - as well as some of the contents of these chapters - sound somewhat ironic, they do draw attention to the prejudices and presuppositions that have always influenced scholars in their work and are likely to continue doing so in the future. Tuck speaks in this connection of isogesis, which he defines as "a 'reading into the text that often reveals as much about the interpreter as it does about the text being interpreted" (p. 9-10). Isogesis, Tuck further explains, is an unconscious phenomenon that is to be distinguished from exegesis, which is conscious intent. All this is very interesting, not only for the philosopher but also for the philologist who studies Indian texts. The latter in particular 1 For some earlier reflections on the methodology of interpreting technical Sanskrit texts, I refer to the Introduction of my Tradition and Argument in Classical Indian Linguistics (Dordrecht: Reidel, 1986).

Page #2
502 BUCHBESPRECHUNGEN / COMPTES RENDUS will wish to know how he can avoid repeating the mistakes of his predecessors. Tuck describes this aspiration in the following passage (p. 13): ... scholars for the past two centuries have defied the isogetical nature of their work by attempting to put aside their own prejudices and presuppositions. By attending as carefully as possible to lexical questions, historical detail, and the accumulation of more and yet more texts to translate and interpret, they have created for themselves as well as for their readers, an illusion of a progressive increase in knowledge about Indian philosophy and of a steady accumulation of better readings of Indian philosophical texts. The standard belief has been that there are more or less correct interpretations of texts and that the meaning of a text is recoverable if all of the necessary philological and historical research is carried out. Concomitant with this belief is the view that disputes between interpreters can be adjudicated, and that there are ways of finding 'correct' readings that are not dependent on the assumptions of the interpreter. Deficiencies in textual interpretation are understood to be a result of 'an imperfect acquaintance with primary source materials' and it is assumed that greater familiarity with original texts and the restriction of the scholar's modern Western biases will give us 'accuracy' and greater understanding of Indian thought. As is clear from this passage, Tuck finds this position problematic. As he points out on p. 15, "for contemporary Indologists to naively accept nineteenth-century objectivist principles betrays an ignorance of the methodological debates that have been taking place throughout the twentieth century in the closely related fields of literary criticism and post-positivistic European/American philosophy". Tuck knows, then, that many contemporary Indologists - who form at least part of his intended readership - are ignorant of these recent debates which could yet seriously affect their way of working, or even convince them of the utter futility of their efforts. One expects therefore some arguments that support these claims, and that might induce the uninformed philologist to mend his ways. But no such arguments are given. It is true that Tuck presents some observations that are no doubt correct and valuable, but they in no way support his conclusions. We have seen, for example, that scholars "have defied the isogetical nature of their work", which seems indeed true for many of them. A particularly important observation is that "[t]here are no interpretations that are not the result of some creative effort on the part of the interpreter" (p. 15); it is this creative aspect of interpretation that Tuck refers to as isogetical. But from this observation to "the fact that knowledge can be understood only in

Page #3
BUCHBESPRECHUNGEN / COMPTES RENDUS 503 specific, culturally embedded forms" (p. 13) is more than an interence; it is an unsupported claim, and an incorrect one at that. It is clear that Tuck underestimates our possibilities of understanding. We can, and we actually do, refine our understanding of a text by confronting it again and again with the principal evidence we have, viz., its exact wording. In this way we can discard false interpretations, which are not simply outdated with reference to the latest philosophical theory in vogue in the West, but really false because in contradiction with the exact wording of the text. By eliminating one false interpretation after the other, we can be sure to get ever closer to the correct interpretation of the text, even if we are to believe that that correct interpretation can never be fully reached. Another point that has not been sufficiently appreciated by Tuck is the following: Scholarship is a collective enterprise, in which mutual criticism plays a vital role. If one scholar is unable to break away from the patterns of thought provided by his culture, someone else may point out the shortcomings of his interpretation. It is no coincidence that Tuck has chosen, in order to illustrate his point of view, the Western interpretation of Nagarjuna. Nagarjuna's works do not state in general terms what they are up to, thus leaving the interpreter the freedom to think more or less what he likes without running too great a risk of colliding with them. This does not necessarily imply that confrontation with the texts will never allow us to make a choice between these various "interpretations". In fact, Stcherbatsky and Schayer's idea that the Madhyamaka absolute exists, and is constituted by the whole of all there is, is an example to the contrary: some Madhyamaka texts say quite clearly that the absolute does not exist." Yet it should be stated that the problems connected with the "interpretation" of Nagarjuna do not so much illustrate the difficulty of crossing a cultural boundary, but the difficulty of finding the opinion of an author on a subject about which he does not express himself. 2 Compare Richard Hayes' recent observation about Nagarjuna (e-mail Buddhism Discussion Group, 16th July 1992): "Not many Indian thinkers have been capable of so many radically different styles of interpretation. He then wonders "what features of Nagarjuna's presentation make it so difficult to interpret definitively and so easy to interpret in whatever way one wants. He's a bit like an oracle in whose words one can hear any message that one wants to hear." 3 See Bronkhorst 1992:71 f.

Page #4
504 BUCHBESPRECHUNGEN / COMPTES RENDUS * It would have been much fairer on the part of our author to study, say. the Western interpretation of Nyaya-Vaisesika, whose texts offer, by and large, fewer fundamental difficulties of interpretation than Nagarjuna. Tuck does mention the Nyaya-Vaisesika school in his first chapter, where he points out that this school drew little attention during the days of European idealism, whereas it did evoke interest during the analytic period. But interest, or lack of it, is not the same as misinterpretation, and Tuck's description of the fluctuating preferences for different schools of thought in India under the influence of changing philosophical fashions in Europe do nothing to support his claims. Systems like Nyaya-Vaisesika have offered relatively little resistance to interpretation, not because there are no cultural barriers here, but simply because the texts express themselves rather clearly about most of the points that interest the Western interpreter. At this point I will briefly discuss an example of a text whose interpretation has been improved by moving away from the model suggested by modern Western thought. This text is the Nirukta. It deals with etymologies, according to its standard interpretation dating from Max Muller in the middle of the nineteenth century. No need to add that these etymologies were looked upon, by Max Muller as well as by his successors, as historical etymologies, of the same type as those provided by historical linguistics, and meant to throw light on the historical development of the word concerned. The etymologies of the Nirukta were thus criticized for being "incorrect", or occasionally praised for being "correct". In reality these etymologies were never meant to elucidate the historical development of words. This becomes clear from a precise study of the wording of the text. It is confirmed by the circumstance that Indian culture tended to look upon its holy languages as stable, not subject to change in the course of time. Here, then, it is possible to arrive at a more correct (without quotation marks) interpretation of the Nirukta, which yet does not correspond to anything in modern Western thought. The idea of non-historical etymologies, though not unknown to an earlier phase of Western culture (cp. Plato's Cratylus), is completely foreign to modern linguistics and to any other modern school of thought. This means that this more correct interpretation of the Nirukta is arrived at by confronting an 4 Bronkhorst 1981; Kahrs 1983; 1984. 5 Bronkhorst, forthcoming.

Page #5
BUCHBESPRECHUNGEN / COMPTES RENDUS 505 initial interpretation that was provided by Western culture, with the letter of the text. This procedure requires, not that the scholar is aware of all his presuppositions or the like, but that he is willing to put question marks behind all his interpretations, especially there where a passage of the text does not appear to agree well with them. In the case of the Nirukta we have not, of course, reached the finally correct interpretation, or the original intentions - all of them - of its author. But only a philosophical nitpicker could deny that we have come a great deal closer to them. If the value of this example is accepted, it is clear (i) that at least in some cases it is possible to get closer to the "real" meaning of a text, and (ii) that such an improved interpretation does not have to be inspired by ideas current in Western thought. But once these possibilities are admitted in principle, it becomes imperative to look for really better - i.e., objectively better - interpretations elsewhere, too. In the Afterword to his book (p. 96 f.) Tuck warns against the extreme of "relativism" - the view that we are irrevocably confined to linguistic and cultural communities, and that real understanding cannot exist among cultures, historical periods, or even individuals. But "G]ust as extreme as the relativists ... are the ordinary scholars who believe in unconditioned facts and objective readings of texts: the unwillingness to question presuppositions is as much a failure of moderation as ... the paralysis ... that can come from too much self-consciousness" (p. 97). I must admit that I am perplexed by this passage. It obviously means that Tuck does not consider himself a relativist, and that he considers willingness to question presuppositions a requirement of good scholarship. Does this imply that Tuck, after all he has said, now joins the scholars of the past two centuries in "defy[ing] the isogetical nature of his work by attempting to put aside [his) own prejudices and presuppositions"? Or do we witness here a feeble attempt to "rescue" scholarship whose very reason of existence had been rejected in the preceding pages? It is a fact that, when it comes to giving practical advice to scholars in the field, Tuck recommends them to continue as before. They should not however believe that they will ever find "knowledge in the sense of a correct "representation of reality", that they are pursuing objective truth. Yet Tuck's book "is not intended to suggest that every previous attempt at cross-cultural philosophical study has failed" (p. 99). Unfortunately it does not tell us what it means for a cross-cultural study to be successful.

Page #6
506 BUCHBESPRECHUNGEN / COMPTES RENDUS All in all, the impression created by this book is that its author is. carried away by some fashionable ideas of which he does not dare, when it comes to it, to draw the consequences. What we must retain from Comparative Philosophy ... is the observation that a creative effort is involved in reading a text. This important insight seems to be lacking in C. Oetke's Zur Methode der Analyse philosophischer Satratexte. Oetke is clearly not interested in the hermeneutical questions that occupy Tuck; there is not a single reference in his book to the methodological debates that form the basis of Tuck's study. His problem is that of the practical philologist who is confronted with the obscurity of philosophical Sutra texts. There is no doubt that what he wants from these texts is their "real meaning" (even though he is aware of the fact that this may mean different things in different circumstances). Tuck's qualms about the possibility of there being a real meaning are not entertained. The interpretation of Sutra texts, never easy, is particularly difficult in cases where sutras may have been added, or removed, in the course of time. This last hypothesis seems the most plausible way to account for the form in which some of these Sutra texts have reached us. Note that this type of internal evidence is as a rule the only justification we have to conclude that a certain Sutra text is not the unitary creation of a single author. Where there are no special reasons to doubt the unitary creation of a Sutra text - as in the case of Panini's Astadhyayf as understood by most scholars - single authorship is taken for granted. This is hardly surprising: it is logically possible to doubt the single authorship of each and every piece of writing. Logically it is hard to exclude the possibility that every sentence of Oetke's book has a different author. In practice most readers will accept single authorship of a text that is presented to us as a unit, until and unless this view presents us with difficulties which an hypothesis of multiple authorship can more easily explain. Oetke is clearly of a different opinion. The sutras which engage his attention in this book - Nyaya Satra (NS) 2.1.8-2.1.19 - allow of a consistent interpretation, as he argues esp. on p. 34-35. Yet later in the book a long discussion is dedicated to the possible original meanings of these sutras, and to their relative chronology with regard to each other. In 6 The following observations on Oetke's book are presented with a certain reserve: Oetke's style is so difficult for me to read that, even after several rereadings, I am still not sure that I have correctly understood him.

Page #7
507 BUCHBESPRECHUNGEN / COMPTES RENDUS a more general way - and here we come to the method announced in the title Oetke presents the view (p. 63) that, at least in principle, first all possible meanings of individual sutras should be traced and examined. Subsequently one should search for plausible combinations: Das Ziel ware die Auffindung von Kombinationen von Interpretationsalternativen einzelner Sutras, die nach verschiedenen "Bewertungsparametern" gemessen insgesamt gunstige Wahrscheinlichkeitswerte ergeben. It must be admitted that Oetke's proposal constitutes a complete novelty, and one can only hope that it will not be followed by other workers in the field, at least not in this extreme form. Not only is Oetke's "atomistic" approach to the sutras of dubious value. The very idea of enumerating all possible meanings of individual sutras, which must then be combined, overlooks the creative element in interpreting texts: one cannot reduce the interpretation of a text to a mechanical enumeration of possibilities. It is typical for Oetke's approach that he says a great deal about logical possibilities, and little about what we actually know about the history of Sutra texts. As noted above, we only know about modifications in Sutra texts in cases where these modifications have left their traces. This, together with the fact that the earliest commentators already choose rather to present a forced interpretation than to change the wording of a sutra, suggests that sutras were not easily changed, i.e., adjusted to a different situation. When, in these circumstances, Oetke enumerates on p. 47 ways in which older sutras may have been incorporated satisfactorily into later works, we are in a realm of pure speculation, which does not become any the less speculative by the fact that the speculations represent logical possibilities. Oetke's approach is further characterized by the extent to which he holds that texts should not be interpreted in the light of other texts (dass man Texte nicht im Lichte anderer Texte interpretieren darf/soll; p. 46). This position gives rise to a long discussion about the meaning of pradipaprakasavat 'like the light of a lamp' in sutra 2.1.19. This comparison occurs a number of times in Indian philosophical literature of the period. Oetke refers, besides to NS 2.1.19, to NS 5.1.10 with Bhasya, Vigrahavyavartant under v. 33, and Vaidalyaprakarana sutra 5. One could add Bhartrhari's Mahabhasya Dipika I p. 3 1. 20, Prasastapada's Padarthadharmasangraha, alias Prasastapadabhasya, vol. II p. 284 (ed. Gaurinath Sastri), and Nagarjuna's Mulamadhyamakakarika 7.8 and 12. Let us look at these passages one by one.

Page #8
508 BUCHBESPRECHUNGEN / COMPTES RENDUS Bhart[hari introduces the comparison of the lamp in the following words: dvisaktih sabda atmaprakasane 'rthaprakasane ca samarthah / yatha pradipah atmanarh prakasayan nidhyarthan prakasayati/ "The word has two powers: it is capable of illuminating itself and its meaning: like a lamp which, while illuminating itself, illuminates the wealth in a treasury." Prasastapada states: yatha ghatadisu pradipat (pratyayo bhavati), na tu pradipe pradtpantarat "E.g., the lamp brings about the cognition of the jar etc.; but no other lamp brings about the cognition of the lamp." Mulamadhyamakakarika 7.8 reads: prad[pah svaparatmanau samprakasayate yatha "Just as a lamp which illuminates itself and something else ..." The Vigrahavyavartant has: dyotayari svatmanamh yatha hutasas tatha paratmanam / svaparatmandy evan prasadhayanti pramananiti // "Just as a fire brightens itself and something else, so the means of knowledge make known themselves and other things." And the Vaidalyaprakarana: tshad ma mnams la ni tshad ma med do / hdir mar me bsin tshad ma ni rart dan gsan sgrub par byed pa yin no / ji ltar mar me ni ran dan gsan gsal bar byed pa mthon ba de bsin du tshad ma mnams kyan ran dan gsan sgrub par byed pa yin no // "(sutra:) Means of knowledge have no means of knowledge (by which they are known). In this respect a means of knowledge is like a lamp: it establishes itself as well as other things. (Comm.:) Just as a lamp is seen to light up itself as well as other things, so do also means of knowledge establish themselves as well as other things." The Myaya Bhasya on sutra 5.1.10 presents the same image: atha pradipar didrksamanah pradipantararh kasman nopadadate / antarenapi pradipantaram drsyate pradipah / tatra pradipadarsanartham pradipopadanamh nirarthakam / "But why don't those who wish to see a lamp fetch another lamp? (Because] the lamp is seen even without another lamp. Here it is useless to fetch a lamp in order to see another lamp." The Nyaya Bhasya on sutra 2.1.19, too, knows an interpretation of that sutra that uses the same image: yatha pradipaprakasah pradipantaraprakasam antarena grhyate tatha pramanani pramanantaram antarena grhyant[e]. "Just as the light of a lamp is grasped without the light of another lamp, so the means of knowledge are grasped without another means of knowledge." 7 Note that the author of the Nyaya Bhasya prefers another interpretation of sutra 2.1.19. Oetke is however right in pointing out (p. 33) that this interpretation has little to recommend itself.

Page #9
BUCHBESPRECHUNGEN I COMPTES RENDUS 509 NS 2.1.19, then, allows of an interpretation that uses an image known from a variety of texts. But here Oetke's principle that texts should not be interpreted in the light of other texts comes in. Oetke warns against a "gleichmacherische Tendenz" (p. 33) and observes that "mit Verweisen auf Parallelen in anderen Texten ... sehr bedachtsam umgegangen werden sollte, weil damit meist nur Unterschiede verwischt werden und dies der ... Tendenz zur Gleichmacherei dienlich ist" (p. 40). In five pages (36-40) he shows that other interpretations of sutra 2.1.19 are imaginable. Unfortunately no independent evidence is provided to support these other interpretations (not even from parallel texts). As so often, all these pages of heavy prose show no more than that other interpretations are logically possible. But who ever doubted this? Oetke's principle would obviously have been served better with one single example where he could show that non-observance of his principle leads (or has lead) to an incorrect interpretation. But clinching examples (or counter-examples) are obviously not his strong side. Mention was made above of the principle that a text be accepted as a single whole unless there are compelling reasons to doubt this. It was also pointed out that Oetke does not accept this principle, at least not where Sutra texts are concerned. This leads to amazing pronouncements, such as the following. On p. 47 Oetke refers to the suspicion of Ruben and others, according to which books 2 to 4 of the Nyaya Sutra have been inserted later into the text. He observes that the fact that all of the Nyaya Satra, including chapters 2-4, constitutes a systematic whole, is no valid argument (his words are: vollig unbrauchbar) against this suspicion. One wonders, of course, what kind of argument Oetke would consider useful to support the unitary nature and origin of a text. Do we have to conclude that for Oetke the non-unitary nature of, at any rate, Indian Sutra texts is axiomatic? Oetke's methodological reflections constitute something like an appendix to a detailed study of Nyaya sutras 2.1.8 - 2.1.19. He criticizes at length another interpretation of these sutras, and then presents one of his own (p. 34-35). Here Oetke makes a remark with which one cannot but agree, and which one wishes he had heeded himself. This remark shows that he is, to at least some extent, aware that finding the meaning of a text is not a purely mechanical affair, the sole requirement for which is, supposedly, applying the correct method. For here he admits that in comparing interpretations of a text, it is the relative superiority of one over the other that counts (p. 35: Die relative Uberlegenheit ist es aber, auf die

Page #10
510 BUCHBESPRECHUNGEN I COMPTES RENDUS es in diesem Zusammenhang allein ankommt). In other words, it is always possible that someone else will, in spite of one's best efforts, find an even better interpretation. Presenting an interpretation that can be corrected by others is not in itself a sign of methodological shortcomings; bad methodology is not responsible for Newton's failure to discover the theory of relativity. Comparing interpretations is of the essence of textual scholarship. It is useless to criticize an interpretation if one has nothing better (or at least equivalent) to offer. Oetke does not seem to realize this: at the very least he would have eased the task of his readers considerably if he had made clear at every step that he criticizes other interpretations because he thinks he can offer a better one. Simply criticizing other interpretations not only makes for tedious reading, it is even methodologically indefensible. It is not possible to discuss Oetke's ideas here in further detail. One general observation must however be made. Whereas most philologists will see it as their task to interpret texts in their historical and cultural context, Oetke has the tendency to abstract the statements he seeks to interpret from any context whatsoever. He is primarily interested in logically possible interpretations, much less in interpretations that fit best the cultural and historical context. Such a procedure may perhaps occasionally rectify interpretations that have been too heavily influenced by contextual, at the expense of textual considerations. Unfortunately Oetke presents no example where this can be shown to be the case. This takes us to the contrast that exists between the two books here reviewed. In an important sense they represent two opposite extremes. For Tuck, there is no way to break away from one's own cultural universe and enter into that of the Indian authors whose texts we study; not even detailed textual scholarship can help us cross the barrier. Oetke, on the other hand, comes close to denying the very existence of such a barrier. Accordingly, he seeks to provide a mechanical method to get at the meaning of the text. In reality there is no justification, neither for Tuck's hopelessness nor for Oetke's methodological optimism. We can in many cases get close to the intended meaning of a text, yet success is not guaranteed by simply applying a supposedly right method. There is a creative element in reading any text, even, or especially, a sutra. But our creativity is not limited, it is no prison. It can get us closer to the meaning of a text if we confront it, as strictly as we can, with the letter of the text concerned. Johannes Bronkhorst