The Study of Buddhism.
Problems and Perspectives[*]
By J. W. de Jong, Canberra

Buddhist Studies
pp. 15-28

Copyright 1979 by Asian Humanities Press
Berkeley, California




    The first chair for the study of Sanskrit in Europe was created in 1814 at the College de France in Paris for Antoine-Leonard de Chezy (1773-1832). He was succeeded by Eugene Burnouf (1801-1852) who occupied the chair for twenty years (1832-1852). It is not necessary to dwell upon the importance of Burnouf's contribution to Buddhist studies. Even today the student of Buddhism makes frequent use of his monumental works: Introduction a l'histoire du Buddhisme indien (Paris, 1844) and Le Lotus de la bonne loi (Paris, 1852). Since Burnouf, many distinguished scholars have studied Buddhist scriptures in Pali, Sanskrit, Chinese, Tibetan and Mongolian. Buddhist studies have developed into an important branch of Orientalism. However, until recently no special chairs for Buddhist studies existed in European universities. It was only in 1926 that a chair for "philologie bouddhique" was created in Paris for Jean Przyluski (1885-1944). A second chair was established in the University of Leiden in 1956.[1] The creation of these chairs has been prompted by the awareness of the fact that Buddhist studies cannot be undertaken as a subsidiary subject by a professor of Sanskrit, Chinese or Mongolian. This becomes obvious when one takes into account the various fields associated with the study of Buddhism. By analyzing some of the more important problems encountered in the past and at present, it is possible to find some indications as to the future perspectives of the study of Buddhism.

    As is well-known, Buddhism originated in the Eastern part of India, spread over the Indian continent and subsequently over a great part of South, Central and East Asia. However, the history of Buddhism Outside India will

* The original Dutch version of this paper was read as an inaugural lecture in the University of Leiden on the 28th September 1956. It was published in the same year by Mouton & Co., The Hague, under the title De Studie van het Boeddhisme. Problemen en Perspectieven. Apart from a few minor changes and corrections, the English version is identical with the Dutch version. The author is much obliged to Miss E.J.J. C. Kat and Professor O'Berkelbach van der Sprenkel for their help in preparing the English translation.

1. Since 1956 chairs of Buddhist studies have been established at the universities of Hamburg and Vienna,




not be taken into consideration, at any rate in so far as it does not assist the study of Buddhism in India. Indian Buddhism can be divided into three so-called vehicles, the Little Vehicle, the Great Vehicle and the Diamond Vehicle, terms which were used by the Buddhists themselves. The Little Vehicle comprises the schools, traditionally 18 in number, that developed in the centuries preceding our era. Around the beginning of our era the Great Vehicle emerges and finally after about five centuries the Diamond Vehicle appears. Each of these vehicles has produced a rich literature. Undoubtedly, this literature is the most important source of knowledge of Buddhism. Buddhist art, inscriptions and coins have supplied us with useful data, but generally they cannot be fully understood without the support given by the texts. Consequently, the study of Buddhism needs first of all to be concentrated on the texts which have been transmitted; and, indeed, it only made good progress after Buddhist philology had been established on a sound basis. Therefore, when discussing the problems of the study of Buddhism one has to take as a starting point the particular difficulties which have been encountered, and are still being encountered, in the field of Buddhist philology: difficulties arising from the way Buddhist literature developed and was transmitted over the centuries.

    The Buddha preached his message of salvation to all of those who came flocking to hear his words; he addressed himself not only to those belonging to the highest classes, as did the brahmans (who were loath to impart their knowledge to others) but to all who wished to hear. The Buddha preached in the local dialect; the brahmans used Sanskrit, which was unintelligible to the lower classes of the people. He desired to share with everyone the Awakening he had obtained under the bodhi-tree, whereas the brahmans instructed their pupils only in the traditionally transmitted texts. The Buddha's teaching did not consist of a number of dogmas, which he tried to inculcate in the minds of his followers. He endeavored to call forth in others the desire to try to obtain Awakening through their own efforts; the brahmans, on the contrary, taught their pupils to recite the sacred Vedic texts without the slightest variation of a single word or even a single accent. In these circumstances it is understandable that the Buddhavacanam, the word of the Buddha, spread over the whole of India and was preached everywhere in the local dialect. After the Buddha's death the schools of the Little Vehicle gradually developed, each having its own canon of sacred texts, each differing in content and language. For centuries these texts were transmitted orally. Were all these texts, in the course of time, committed to writing in the various languages in which they had been transmitted from one generation to the next? It is difficult to answer this question



since, as a result of the almost total disappearance of Buddhism from the continent of India, very few manuscripts of Buddhist texts have survived in India. However, one of the schools, that of "the Elders", spread to Ceylon before our era, according to the tradition, already during the time of Asoka in the third century B. C. The canon of this school, written in Pali, a middle-Indo-Aryan language, was put down in writing and its manuscripts have been preserved. Ceylonese tradition has it that Pali is identical with Maagadhii, i. e. the language of the country of Magadha, where Buddhism originated and that therefore the Paali-canon contains the word of the Buddha in the very language used by him. This tradition has been overthrown by modern scholarship. At present it is generally agreed that Paali is neither the language of Magadha, nor one of the languages spoken at the time anywhere in India, but a literary language containing elements derived from more than one local dialect. Attempts to discover the origin of Paali, or the main dialect on which it is based, have been fruitless. It has been established that, linguistically, the Paali-canon consists of various strata, and that at least parts of it are translated from texts in another language, designated by Sylvain Levi as 'langue precanonique du bouddhisme'[2] and by Luders as Maagadhii or old-Ardhamaagadhii[3]. The latter scholar is of the opinion that the Buddhist texts in Paali and Sanskrit are based on an original canon in Maagadhii. This assertion is not borne out conclusively, or indeed made plausible, by the results of his research. In the first place there is no evidence whatsoever of the existence of an original canon established either in oral form or in writing. It seems more reasonable to assume that only a small number of texts, above all verses and stereotyped dogmatic formulas, at an early stage attained a definite form, and that they were subsequently translated into Paali and other middle-Indo-Aryan languages. But it is difficult to believe that the extensive Paali-canon with its various linguistic strata-not to mention the Buddhist Sanskrit texts-was translated in its entirety from an original canon. Secondly, it has by no means been proved by Luders that Maagadhii was the language of this original canon. Contemporary texts in Maagadhii are totally lacking, and the earliest inscriptions at our disposal date from the time of Asoka, the middle of the third century B. C., or later. It is impossible to determine to what extent the language of these inscriptions reflects the spoken Maagadhii of that period. It is quite conceivable that Asoka's chancellery used a lingua franca in which Maagadhii probably dominated, as may

2. "Observations sur une langue precanonique du bouddhisme", FA, 1912, II, pp.495-514.

3. Beobachtungen uber die Sprache des buddhistischen Urkanons (Berlin, 1954), p. 8.



be deduced from the fact that inscriptions far beyond the borders of Magadha were also couched in the same language as the inscriptions in Magadha itself. Literary texts in Maagadhii appear only after the beginning of our era in a few scarce fragments of Buddhist plays, and not until much later more fully in classical dramas. The artificial character of the so-called theatre-Maagadhii is well-known. Consequently, neither the inscriptions nor the texts provide sufficient material to give a clear insight into the development of the language of Magadha. Even with the best will in the world it would still be impossible to
draw from this scanty material enough evidence to allow the conclusion that the original canon was written in Maagadhii.

    In view of these facts, the most reliable method would seem to be to scrutinize closely the Paali texts and parallel texts in other languages; thus to try to discover where these texts go back to a precanonic text, and then to try to determine the character of the language of this text. Luders has done much valuable work in this direction, and it is a task for the future to continue research in the way indicated by him, avoiding rash theories, which can only hinder progress. As to the languages in which the canons of the other schools of the Little Vehicle were written, only suppositions can be advanced, since such indications as the Buddhist tradition has to offer with regard to this point are completely useless, and no other texts in a middle-Indo-Aryan language have come down to us-with the exception of one text written in a language of the north-west of India: the Gaandhaarii Dharmapada. Unfortunately it is impossible to be certain whether this text formed part of an existing canon.[4] In Kashmir, Afghanistan and Central Asia fragments in Sanskrit have been found of texts belonging to two schools of the Little Vehicle, the Sarvaastivaada and Muulasarvaastivaada schools, and it is generally held that these schools made use of Sanskrit in writing down their canon. If we bear in mind the results obtained from the study of the Paali canon, we ought to be extremely wary in drawing any such conclusion. It has been ascertained that this canon was written in a literary language and not in a dialect. On the other hand the oldest layers of the Paali canon already show a Sanskrit influence, an influence which is increasingly apparent in the later strata. According to tradition,

4. John Brough believes that the Gaandhaarii Dharmapada [MS Dutreuil de Rhins] belongs to the canon of a school and declares that the Dharmaguptakas and Kaa`syapiiyas must be considered as eligible, but that still other possibilities cannot be ruled out; cf. The Gaandhaarii Dharmapada (London, 1962), pp. 43-45. Franz Bernhard decides in favour of the Dharmaguptakas, cf. "Gaandhaarii and the Buddhist Mission in Central Asia", A~njali. A Felicitation Volume Presented to O.H. de Alwis Wijesekera (Peradeniya, 1970), pp. 55-62.




which on this point has some evidential support, the writing down of the Paali canon took place in Ceylon several decades before the beginning of our era. Since then the manuscripts have for centuries been subjected to a process of normalization at the hands of commentators and grammarians who have replaced ancient readings by others more agreeable to Paali-grammar[5]. Traces of the influence of Sanskrit were in all probability greatly reduced in this way. If then the Paali-texts in India had already been strongly marked by Sanskrit influence well before the Christian era, would it then be too bold an assumption to make that they escaped a process of ever-increasing Sanskritization precisely because of the fact that they had so early been committed to writing in Ceylon? The canons of the schools in India must have experienced the influence of Sanskrit to an ever increasing degree. It is therefore not at all improbable, that the canons, of which only fragments in Sanskrit have survived, were also originally transmitted in a middle Indo-Aryan language. This hypothesis is supported by the linguistic aspect of the fragments, the language of which is not pure Sanskrit, but Sanskrit mixed with middle-Indo-Aryan forms Moreover, the oldest manuscripts do not date back further than to the 5th or 6th century. The process of Sanskritization is clearly visible in the scriptures of another school, the Mahaasaa.mghikas. To this school belongs an extensive text, written in Buddhist Hybrid Sanskrit, and transmitted in Nepalese manuscripts. A linguistic analysis of this text, the Mahaavastu, has proven the presence of different linguistic strata. The language of the oldest strata deviates much more from classical Sanskrit than does that of the later strata. Undoubtedly, the oldest parts go back to a text in middle Indo-Aryan. More uniform in character is the language of the recently published Vinaya texts of the same school.[6] These texts must originally have been composed in a middle-Indo-Aryan language but Prakrit forms have been replaced to a large degree by Sanskrit forms.

The above considerations justify the preliminary conclusion that the texts of the Little Vehicle have been subject to a process of Sanskritization to an ever-increasing degree. It is difficult to establish in detail the course of this process and the period of time involved, for the texts, handed down to us in

5. Cf. H. Smith, Saddaniiti, la grammaire Palie d'Aggavamsa, vol. I (Lund, 1928), pp v-vi; vol. IV (1949), p. 1139.

6. The of the Mahaasaa^nghikas. Edited by W. Pachow and Ramakantha Mishra, Journal of the Ganganatha Jha Research Institute, IX (1952), pp. 239-260; X (1952-1953), app. pp. 1-48; XI-XII (1953-1955), pp. 243-248; Abhisamaacaarika. Edited by B.Jinananda. Patna, 1969, Edited by Gustav Roth. Patna, 1970.




manuscripts, reveal only the final stage of the development. However, by taking into account the history of Indian languages, it is possible to obtain a better insight into this problem. Epigraphy is here of great importance, since inscriptions which can be reliably dated, are available to us. The oldest Indian inscriptions are those of Asoka, dating back to the middle of the third century B. C., and are in middle-Indian. The earliest inscription in Sanskrit probably dates from the end of the first century B.C. After the middle of the second century A. D. the use of Sanskrit increased, with the result that inscriptions in middle-Indian are hardly to be found at all in North India after the third century, or in South India after the fourth century. Sanskrit influence on inscriptions in middle-Indo-Aryan is noticeable in varying degrees, It is a task for the future to study the epigraphic materials and to try to trace how far the linguistic characteristics can be explained by historical, geographical, political, social and religious factors. It is to be hoped that in this way new light may be shed not only on the history of Indian languages, but also on the Sanskritization of Buddhist texts.

    In discussing the linguistic problems of the transmission of Buddhist texts we have limited ourselves so far to the literature of the Little Vehicle. However, it is necessary to study in the same way the literature of the Great Vehicle; and also the history of both Vehicles, because after the rise of the Great Vehicle in India, the schools of the Little Vehicle did not disappear but continued to transmit their texts. The surviving texts of the Great Vehicle in Indian languages can be divided into those in pure Sanskrit, and those in the so-called Buddhist Hybrid Sanskrit mentioned earlier. This last is a Sanskrit which, as shown by the extant texts, has been interspersed to a greater or lesser degree with middle-Indo-Aryan forms, words and constructions. In past years the question has been much discussed, whether these texts were originally composed in one or more middle-Indo-Aryan dialects, or in a lingua franca derived from several middle-Indo-Aryan languages. The second solution gives rise to the question whether this language was generally spoken; or was a religious language only, in which case one may also inquire whether this language was spoken amongst the Buddhist or used only for writing down texts and/or for recitation. Finally, the question was raised, whether those texts, whose language is almost completely free of middle-Indian influence, were written directly in this language or, alternatively, should be seen as the end product of a long process of Sanskritization. It is not my intention to discuss these problems in detail. From the discussions one often gains the impression that one is inclined too soon to have an answer and an explanation ready with-




out a careful study of all available material. First and foremost attention needs to be drawn to a very important source of material, which has so far hardly been made use of: namely the translations of Indian Buddhist texts into other languages. Only a small amount of the Buddhist literature that once existed in Indian languages has survived-though we must add that, in the past seventy years especially, important finds have been made in Central Asia, Kashmir and Tibet. Nevertheless, this amounts to only a very small quantity compared to the thousands of texts which have come down to us in Chinese and Tibetan translation. Buddhism came to China in the first century A.D. and to Tibet in the seventh or eighth century. The Chinese translations, collected in the Chinese Buddhist canon, were almost all made between the second and the twelfth centuries. The Tibetan translations collected in two canons, the Kanjur and the Tanjur, were completed between the eighth and seventeenth centuries. Being exceptionally literal, they are of the greatest importance for textual criticism; though only of secondary importance as far as the history of the texts is concerned, since for the most part they came into being in a period when Buddhism in India had passed its peak, and when most of the texts had already received their definitive form. The Chinese translations, on the other hand, though much less exact and (especially in the earliest period), often very poorly translated, are of incalculable value for the history of Buddhist texts. Chinese translations of many texts exist from various periods. They thus enable us to see how the texts in question gradually developed. It appears for instance that some texts of the Little Vehicle, by reason of changes and additions, were transformed into texts of the Great Vehicle. With the aid of the Chinese translations the various layers of a text can often be determined. It is obvious how important this is for the study of texts in Buddhist Hybrid Sanskrit. For the problem of the middle-Indian original text can only be properly posed when the original nucleus has been extracted from the text. Moreover, from a careful study of Chinese transcriptions of proper names it is possible to reconstruct their original form. However, before making use of Chinese translations one needs to determine when they were translated and to what extent the present text is identical with the original text of the translation.

    The Chinese, with their predilection for bibliography, already in early times began to compile catalogues of Buddhist texts. The data contained in these catalogues need to be critically analyzed by comparing one catalogue with another, and confronting the results with data supplied by the biographies of the translators. Finally, the texts themselves must be analyzed critically. The translations were often improved and changed, or wrongly attributed. Such




critical analysis has recently been carried out quite excellently, mainly by Japanese scholars, and has already yielded important results. Up till now the work of Japanese scholars in this field, as also in other fields of Buddhist studies, has been insufficiently studied by non-Japanese scholars. This has been either because of inadequate knowledge of the Japanese language, or because of the absence of Japanese publications in University libraries. Today one can see a noticeable change: for, in addition to Sanskrit, Paali, Tibetan and Chinese, more and more attention is being paid by non-Japanese scholars to work in Japanese. And, as well, there is a growing tendency in Japan to give more general currency to the work of Japanese scholars by translations into English or French, or by summaries in these languages.

    From the foregoing it will be clear that the study of Chinese translations is essential for a better understanding of Indian Buddhist literature. A critical scrutiny of these translations cannot be made without studying the history of Chinese Buddhism. In the same way it can be shown that the study of Buddhism in the other countries to which it spread is also necessary for the understanding of Indian Buddhism. In particular, it is of great importance to study Buddhism in those countries where it is still flourishing today, and where an uninterrupted tradition extends back to Indian Buddhism. For the Great Vehicle, Tibet and Japan are of particular interest; and for the Little Vehicle, Ceylon, Burma, Siam and Cambodia. This is not only because texts of Indian origin have been transmitted, translated, explained and analyzed in those countries ever since the introduction of Buddhism, but even more so because the study of present-day Buddhism, and the direct contact with the Buddhist mentality which these countries afford, can contribute to a deepening of our understanding. The study of Buddhism in Europe has been mainly concerned with its philological aspects. Since, in the field of Asian studies, university chairs were established only for Asian languages, it was often a Sanskritist or a Sinologist who specialized in this study.

    Here we may find the answer to our question, why the need of special chairs for Buddhism has hardly made itself felt up till now. It was only the realization of the scope of the task of Buddhist philology which led to the creation in Paris of a chair for "philologie bouddhique". It is hardly surprising therefore, that excellent work was done in the field of philology. Furthermore, we must not overlook the important progress that the study of Buddhism has made in India and Japan since the 'twenties', through the adoption of the methods of Western philology there, .especially as developed by the French school under the guidance of Sylvain Levi.




    Activities in Europe were by no means restricted to editing, translating and studying the texts. The number of books and articles attempting to explain the teaching and history of Buddhism, are as numerous as the grains of sand on the banks of the Ganges. However, if one considers how effective these publications have been in providing a better insight into the essence of Buddhism, the results appear to be disappointing. It is necessary to examine in what ways, and for what reasons, this has been so, if the mistakes of the past are to be avoided in the future.

    In order to give a clear example of the difficulties encountered in understanding Buddhism as a religion, I would like to draw attention to the various concepts which are to be found in the works of Western scholars concerning the figure of the Buddha On the analogy of the term Christology, it would be possible to designate this branch of Buddhist studies by the term Buddhology. However, it may be better to avoid this term in order to prevent any assumption of an analogy between the problems in those two fields of study.

    Burnouf. the brilliant founder of the study of Buddhism, occupied himself only incidentally with the figure of the Buddha. In his Introduction a l'histoire du Buddhisme indien, published in 1844, he gave a systematic description of the Buddhist Sanskrit manuscripts which Hodgson had sent to Paris from Nepal. With remarkable intuition Burnouf divided these texts into three groups,[7] corresponding to the three Vehicles mentioned above. Basing himself on the texts of the first group, Burnouf regarded the teachings of the Buddha as dominated by ethical doctrines.[8] He was of the opinion that the Buddha, during his life-time, had always been a human being even for his most ardent followers.[9] We hear quite a different sound in the Essai sur la legende du Buddha by Senart, which appeared in 1875. The story of the Buddha's life contained in the texts, as he interprets them, is the story of a sun-hero. As a god of light the Buddha descends in the cloud-womb ("le sein nuageux") of his mother, who perishes after his birth in the blaze of his rays.[10] In this way Senart explains the legend of the Buddha, from beginning to end, as a naturalistic myth. He regards the non-mythic elements in the legend as of secondary importance, as later additions to the myth, which has gradually been transformed into a divine legend and subsequently into a hero-legend, having in the process assumed more realistic elements.[11] Seven years later, in his Geschiedenis

7. Op, cit., p. 581.

8. Op. cit., p. 126, 152, 335.

9. Op. cit., p. 340.

10. Op. cit., 2nd ed. (1882), p. 433.

11. Op. cit., p. 435, 445.




van het Buddhisme in Indie [History of Buddhism in India], Kern carried this method of interpretation even further. Even Buddha's preaching becomes a manifestation of a sun-god. As an example I quote what Kern says about the first sermon of the Buddha: "The first sermon falls on the day of midsummer, and therefore, according to all rules of mythology, the Buddha was not allowed to preach about anything else but the text which nature offered him: he recommended the middle way. Seldom has the golden middle way been commended in a more trivial way, surely, but then this recommendation of a generally acknowledged, praiseworthy principle was not the main consideration."[12]

    The mythological explanations of Senart and Kern found little approval. Naturalistic mythology had already passed its peak; although, for example, in the field of the Veda, its influence would continue to be felt for a long time. Moreover, more attention was soon given to the oldest Buddhism and to the Paali texts, of which only a few had so far been published and translated. Schayer has pointed out that the study of Buddhism has been greatly influenced by the view that in the history of a religion only the original is real, all that comes later being more or less a degeneration. Just as protestant theologians tried to reveal true evangelical Christianity, so one wanted to get to know "the wonderful figure of the Buddha" and "the original world of ideas" on the basis of the Paali-texts.[13] To this observation by Schayer can be added the fact that the influence of 19th century ideas about Christianity is also noticeable in the attempt now made to regard the Buddha as a teacher of humanity, preaching a sublime morality. While on the one hand one tried to point out similarities between Buddhist and Christian morality, on the other, there was no lack of emphasis on the fact that the oldest Buddhism was a religion without a god. Although undoubtedly such views were more prominent in the more popular literature about Buddhism, one cannot ignore their influence on the work of scholars. All these factors contributed to the wide response given to Oldenberg's work: Buddha, sein Leben, seine Lehre, seine Gemeinde, published in 1881. However, it was certainly the masterly way in which Oldenberg presented his views which turned the scale. Regularly reprinted for more than 80 years, this book, as no other, has left its mark on the wav Buddhism has been understood in the West. Oldenberg reproved Senart for having made use of texts dating from a time in which the memory of the Buddha had become overgrown by legendary and mythological elements. According to him, one should first and foremost turn to the Paali-texts, most of which were already compiled

12. Op, cit., Vol. I (1882), p. 240.

13. Ausgewahlte Kapitel aus der Prasannapadaa (Krakau, 1931), p. ix.




during the century following the Nirvana of the Buddha.[14] Legendary traditions about Buddha's life in these texts should be ignored. Once these non-historical elements have been eliminated, there remains a series of positive facts which are, historically, completely reliable.[15] With a sound philological knowledge, and always with references to his sources, Oldenberg carried out this programme. Up to the present his method has found many followers who, however, frequently did not observe the same caution and carefulness shown by Oldenberg himself, and therefore often went much further than he in their search for historical facts in the texts.

    In the long run, however, a reaction against Oldenberg's viewpoint was bound to come. This was caused, in the first place, by a change in the evaluation of the Paali texts. It appeared more and more impossible to date the texts as early as Oldenberg had done. At the same time, more emphasis was placed on the fact that the Paali canon embodied the tradition of one school only of the Little Vehicle and consequently could not be regarded as fully representative of original Buddhism. It was gradually realized that the legendary elements in the Paali texts were not to be so easily put aside. They both appeared in the oldest strata and were closely interwoven with the other elements. Moreover, from the study of the inscriptions and of archaeological remains it became apparent that already at an early stage the legendary element played a great role. All this resulted in a more careful approach toward determining the historical element in the texts. Some scholars even gave it as their opinion that no adequate criterion existed for distinguishing between historical and non-historical elements. This view has been advanced very forcefully in an extremely interesting article[16] by Lamotte in which he surveys the study of the legend of the Buddha and analyzes its sources on the basis of texts in Paali, Sanskrit and Chinese. In this article he shows that the legendary element already formed an integral part of the oldest documents, and has played an ever increasing role as time went on. It appears, moreover, that it took almost ten centuries to write a complete biography of the Buddha, as the oldest texts only relate some episodes from his life, while other episodes were added later on. According to Lamotte every attempt to discover an historical kernel in the legend of the Buddha is doomed to failure; one can only endeavour to trace the growth of the legend, and if possible to explain this process by means of a critical study

14. Op. cit., p. 77.

15. Op, cit.,p. 92.

16. "La legende du Buddha", Revue dc l'histoire des religions, t. CXXXIV (1947-48), pp. 37-71.




of the texts and the archaeological monuments.

    Following Lamotte, one can distinguish three different tendencies in the views of Western scholars. Firstly, the mythological explanation, as given by Senart and Kern, then the rationalistic explanation with Oldenberg as its most important representative; and finally the pragmatic attitude, which has been formulated most clearly by Lamotte himself. Adherents of the first view tried completely to reduce the Buddha to a god by eliminating all historical or pseudo-historical elements; whereas, contrariwise, those adopting the second view wanted to emphasize Buddha as a human being by eliminating all that pertains to the legend. In both cases too great a reliance was placed on the method used. Each school, by assuming either the mythological or the historical element to be the "original" one, thought themselves able to prove their thesis by critically sifting the sources. The mythological explanation failed to pay attention to the Buddha as founder of Buddhism. His very existence was doubted by Kern, who claimed that "the Buddha of the legend is a mythical figure, who no longer bears the marks of the historical founder of the school, even if he existed."[17]

    Senart was much more careful on this point, and stated emphatically that it was not his intention to prove that the Buddha never existed.[18] He even considered it possible that memories of true events were interwoven with the legend; but he believed that the historicity of these elements could not be proven. With the rationalistic explanation, on the other hand, there was no doubt about the possibility of eventually revealing an historic kernel in the legend of the Buddha. There was, however, a difference of opinion about the question whether to his immediate followers the Buddha had been a human or a supernatural being. It was not fully realized that this is the fundamental question-the question which needs to be put at the very beginning of the enquiry. This is because, to form an opinion as to the historic value of the sources regarding the Buddha, it is necessary first to trace the ideas with regard to the Buddha which are found in the sources themselves. Reading the oldest Paali texts, it becomes clear that the alternative man/god does not apply to the Buddha. For the Buddhists, gods were beings who as a result of their merits in a previous existence, enjoyed a state of bliss in heaven. With the passage of time, however, their existence as gods ends. Although in many respects the life of these gods is more pleasant than that of human beings, only human beings are able to hear Buddha's sermons, to obtain salvation

17. Op. cit., vol. I, p. 233.

18. Op, cit., p. 452.




from the cycle of rebirth and to become free from all impurities-impurities from which the gods cannot be completely free. How possibly could Buddha's followers ever have regarded him as a god? Only an outsider could ask himself this question. A famous text[19] narrates how a brahman asks the Buddha whether he is a god, a demi-god, a demon or a human. The Buddha replies that the impurities which could have made him one of these beings had been rooted out:

"Just as a lotus, born in the water, rises above it without being soiled by it, so have I, born in the world, overcome the world and remain in it without being soiled by it. Realize, brahman, that I am a Buddha."

    How evident it is from this passage, to which numerous others could be added, that the Buddha must be regarded as a being sui generis-as a unique being. No matter how, in later texts, the concept of the Buddha changes and the miraculous and supernatural become more prominent, he nevertheless always remains the Buddha. This must be the starting-point in the study of Buddhist literature. Its authors are not historians in the sense which a modern European scholar would ascribe to this word. Their history of the Buddha, his birth, his Awakening and his death, is a history of salvation. In the same way, what has been written by the schools of the Little and the Great Vehicle is religious history. From the texts one can learn the Buddhists' point of view about the Buddha, and learn which traditional beliefs existed in the various schools. The actual historical truth can only be found if one resorts to other, non-Buddhist sources. If they confirm the Buddhist tradition on a particular point, then one may assume, according to the rules of historical criticism, that this point is an historical fact. Too often another method is followed by scholars who seek to find the historical truth by establishing an agreement in the traditions of the various Buddhist schools. Some years ago, in connection with a work in which this method was used, Demieville once again clearly pointed out that in this way one can only show that in so far as the various traditions of different schools agree, a common tradition underlies them; but this affords no ground for assuming this common tradition to be a reflection of historical reality.[20] Undoubtedly, if this method is rejected, only very few historical facts can be established: for the number of cases in which different, independent sources relate the same facts is extremely limited in the history of Indian Buddhism. The establishment of historical facts, however, requires a rigorously critical

19. Ang Nik., vol. II, pp. 38-39.

20. "A propos du concile de Vai`saalii", T'oung Pao, XL (1951), pp. 269-270.




approach, and one cannot rest content with traditions simply for lack of historically reliable sources. No matter how carefully one tries to free 'traditions' from all miraculous elements' and reduces them to their common source, they stubbornly resist transformation into historical documents. Using such a method, one falls inevitably into a way of history-writing which is in direct conflict with the character of the sources.

    In the past European Orientalists have applied themselves especially to the history of Buddhism, as has recently been underlined by Eliade.[21] Educated in the historical tradition of the nineteenth century, scholars believed they could learn all about Buddhism by studying its history. In the first place they tried to obtain a knowledge of the facts and data in order to form a picture of the development of the Buddhist ideas. This method is doomed to failure because in the spiritual life of India the historical dimension is of much less importance than it is in Western civilization. The most important task for the student of Buddhism is the study of the Buddhist mentality. That is why contact with present-day Buddhism is so important, for this will guard us against seeing the texts purely as philological material and forgetting that for the Buddhist they are sacred texts which proclaim a message of salvation.

21. "Le bouddhisme et l'occident", Le Figaro, 6 August 1952.