Treasure Trove of Ancient Literatures

by J. K. Nariman

A report of the archaeological findings coming out of Central Asia in the early 20th century.




I. Scientific Expeditions This paper is mostly a translation of Lüders, Uber die litterarishen Funde von Ostturkestan. It was printed as Appendix V of J. K. Nariman’s Literary History of Sanskrit Buddhism. I have shortened the headings in this edition.

[224] The country of East Turkestan has been one of eternal unrest since the beginning of the second century before Christ. Historical notices, especially by the Chinese, supplemented by our finds, show that it had as guests, one after another, Indian clans, Tokharians, Huns, Scythians, East Iranians, Tibetans, Turks, the people of Kirghiz and Mongols. The picture of the country, as it was in the seventh century, that is, at a time when the majority of the MSS. now discovered were written, is drawn for us by Hiuen­tsiang. He went on a pilgrimage to India in 629. His object was to see the cities between which the Founder of his faith travelled, and to acquire some of the holy books; He chose the northern route and passed through Chotjo, the capital of modern Turfan. On his return he traversed Kashgar, Yarkand, and Khotan. On the eastern confines of Khotan begins the desert, where the sand is kept shifting by the perpetual movement of the wind. The only landmarks visible are the whitened bones of pack-animals. Hereabout lay the ancient kingdom of Tokhara – already in ruins – and beyond was the silence of death. Flourishing life was, however, visible towards Khotan. All along, Buddhism was the dominant religion. Many thousands of monks lived in the monasteries of the countries, the northern side belonging to the school of the Sarvāstivādis, Yarkand and Khotan being Mahāyānists. The Chinese traveller has noted for us the various characteristics of the people who had nothing [225] in common, except their religion. They were various as regards dress, customs, manners, languages and modes of writing. The last was borrowed no doubt from India in each case. A new period of culture began for the country with the appearance of the Turkish clan of the Uigurs. They absorbed the inhabitants and united them into a people known to this day by their name. East-Turkestan in the matter of religion was only a province of India. Then side by side with Buddhism appeared Nestorian Christianity and Manichæism.

The ruler of Turfan was the first to embrace it. Soon after came upon the scene a new arrival which showed itself to be stronger than Buddhism, Christianity, or the doctrine of Manes. The first conversions to Islam took place in Kashgar and the first Islamic dynasties took their rise there. The older faiths continued their existence, but there was no stemming the tide of Islam. From the fourteenth century onwards Turkestan became definitely Muhammadan. China acquired the country in 1758 without altering its religion.

The words of the Buddha, of the Christ, and of Manes ceased to be heard; yet the works which embody them survived. Ruins of monasteries, which are proved to be Christian from wall-paintings, inscriptions, and the find of MSS., have come to light in the capital of Turfan. In the centre of the city there was a large Manichæn colony. In this part was discovered a wall-painting, which is the most valuable find of an original fresco in the Berlin collection. It is a picture of a Manichæn priest surrounded by believers, men and women, in their characteristic dress. The building was ransacked by the peasants in search of buried treasures when the German scientific expedition arrived. It appeared just at the moment when the real treasure would have been destroyed. The place abounds in traces of Buddhistic monuments. Without the help of illustrations [226] it is difficult to gain an idea of the architecture of the times – the temples, the stūpas, the monasteries. The art of Gandhāra was transferred from its home in India to Central Asia. Over all a strong Iranian influence is noticeable. The further we come down the stream of time, the more mixed and complex becomes the style and the problems of civilisation studied by Stein, Grünwedel and Le Coq. It will require several decades to study the entire collection of finds. Philologists and archaeologists will not be the least interested investigators.

The first find of MSS. by a European, which gave the impetus to further archæological search in Central Asia, was a bark MS. which was found by two Turks in 1890 in a ruined stūpa. They sold it to Lieut. Bower, who was then the British Resident at Kucha. Bower presented the find to the Asiatic Society of Calcutta. The next year, Dr. Hoernle, the Secretary of the Society, published a report on the MS. which evoked considerable interest. The antiquity of the MS. was noteworthy. Indian MSS., according to the western standard, are relatively young. The destructive effect of climate and the pest of insects require their continual renovation. The oldest MSS., preserved in Nepal on palm leaves, date back to the beginning of the eleventh century. Only two palm leaves were hitherto known which had crossed the Indian border in 609 and reached Japan through China. They were preserved there in the celebrated monastery of Horiuzi, as venerable relics. The Bower MS. however was a considerable and complete one. It was written in the Gupta character, and hence had come undoubtedly from North­West India, and dated at the latest from the fifth century. Later investigations have proved, that it must date from the second half of the fourth century. The possibility of such a discovery incited to further research. The Russian Archæological Society asked the Russian Consul-General in Kashgar, [227] and the British Government commissioned the political agents in Kashmir, Ladak, and Kashgar, to look out for similar MSS. Thus have been acquired the MSS. which are known as the Petrovsky, the Macartney and the Weber. They are housed either at Petrograd or Calcutta. They belong to a large find made soon after the discovery of the Bower MS. by Turkish peasants in Kucha. For a long while the collection had remained in the house of the local Kazi, as a plaything which amused his children!

Meanwhile there was another discovery in 1892. The French traveller Dutreuil de Rhins found three MSS. in Khotan which he despatched to Paris. In 1897 Sénart made known their contents and value. By now we are quite used to surprises from Central Asia. At that time, however, Sénart’s communication created a sensation in the Aryan section of the Oriental Congress held in Paris. The find represented a Kharoshti MS. The Kharoshti character till then had been known only from inscriptions in the outermost boundary of North-West India. Epigraphical comparison proved the date of the MS. to be the second century. As to its contents, it was a recension of the Pāḷi Dhammapada in a Prakrit dialect, which was till then unknown in literary compositions. The manuscript was only a fragment. Another portion of the same MS. was brought to Petrograd.

The impetus given by an accident transformed itself into systematic research. The Russians were first on the scene. In 1898 Klementz set to work on this spot, and the next year Radloff started the initiative, which formed an International Association for Investigation in Central and Eastern Asia. What surprise awaited the seeker, was shown by the results of the labours of Sir Aurel Stein supported by the British Government in the country round Khotan in 1901. Stein’s personal travels led to a secondary discovery. He found out [228] and exposed the manufacture and sale by Turks of fabricated MSS.

Stein’s success led to the German expedition under Grünwedel and Ruth to Turfan in 1902. Meanwhile with the exertions of Pischel there was formed a German Committee of Research which, with State help, in 1904 and 1907 sent out two expeditions under the leadership of Le Coq and Grünwedel. And Kucha and Turfan were thoroughly searched. The result was brilliant. In 1906-1908 Stein set out on his second journey. His most beautiful discoveries he made in the territory of Tun-huang. He came across a portion, altogether forgotten till then, of the great wall built by the Chinese as a protection against the incursions of the Huns. Here a windfall awaited him in the shape of a literary treasure. A few years before Stein’s arrival, a Taoist priest in the hall of the Thousand Buddhas, or Tun-huang, as it is called, discovered among the caves a cellar which had been walled up. It contained a huge library of thousands of MSS. To judge by the date of the MSS., the cellar must have been closed up in the beginning of the eleventh century. Stein secured a considerable portion of the treasure. A portion fell to the lot of the French scholar Pelliot, who journeyed to Turkestan in 1906-07. Even Japan was not behindhand. In 1902 it sent a Buddhist priest who made excavations with some success. To preserve the remains of the Tun-huang library from destruction, he despatched them to the National Library of Peking. Thus, in addition to archæological discoveries, there has been collected a huge mass of MSS. and block-prints in the libraries and museums of Petrograd, London, Oxford, Calcutta, Berlin, Paris, Tokio and Peking. Almost every material used for writing purposes is represented – palm-leaf, birchbark, wood, bamboo, leather, paper and silk. The number of alphabets represented is very large. The languages in which these MSS. are [229] written are counted by the dozen, including several, of which, till the other day, we had no knowledge.

Among the first finds which reached Calcutta and Petrograd, there were fragments of MSS. written in a variety of the Indian Brahmi character. The language, however, was not Sanskrit. The writing was tolerably clear and Hoernle succeeded in deciphering Indian names and expressions of Buddhistic terminology and Indian medical terms. Next Leumann proved that we had here to do with two different tongues. The merit of discovering the exact nature of the first of these belongs to Sieg and Siegling, who in 1907 proved its Aryan character from the names of domestic animals, parts of the body, terms of relationship, and figures. The name of this language was the Tokharian. It was mentioned in the colophon of a MS. deciphered by F. W. K. Müller. The manuscript represented the Turkish version of a Tokharian translation from a Sanskrit original. One dialect of it seems to have been widely common. Caravan passes written in it have been discovered, and dated and deciphered by Pelliot and Sylvain Lévi. Further results may be expected from the studies of Mironov and Meillet. There is a vast number of MSS. which represent translation and redaction of Sanskrit works relating to Buddhism and medicine. There are also some Buddhistic dramas; they can be traced to Indian modes as is shown by the mention of the vidūṣaka.

The second new language is represented by two groups of texts, and is studied especially by Stael von Holstein and Konow. The first represents business papers, mostly dated, though the current era is not known. The second group embodies Buddhist texts, partly dated. While the Tokharian fragments are of works belonging to the Sarvāstivādi school, the texts of the second language belong to the later Mahāyānist literature – for example the Vajracchedikā, the [230] Aparimitāyu-sūtra, the Suvarṇa prabhāsa Sūtra, Saṁghāta Sūtra, and the Adhyardhaśatika Prajñāpāramitā.

II. New-old Tongues

In 1904, F. W. K. Müller succeeded in deciphering a couple of fragments of paper, letter, and silk, originating from Turfan. He declared the alphabet to be a variety of the Estrangelo, the language as Middle Persian or Pahlavi, and the contents as pieces from Manichæn literature believed to have been lost. This was the commencement of a long series of brilliant discoveries, the results of which have been registered in contributions to learned journals. A heap of dogmatic and liturgical works has been recovered of the religion of Manes, which spread from further Asia to China, and in spite of sanguinary persecutions of centuries asserted itself on the coast of the Mediterranean as a rival to Christianity. It is, though but debris, a priceless possession, because for the first time we perceive here from its own books the doctrine, for a representation of which, up to now, we had to rely on the hostile writings of Augustine, the Acta Archelai, the formula of abjuration of the Greek Church and the celebrated Fihrist, a kind of detailed catalogue of contemporary Arabic literature by an-Nadhim. So far, as can be ascertained, the principles of the doctrine have been correctly characterised: here the ethical and physical elements have been indissolubly united in a fantastic fashion. Kessler was inclined to see in it a preponderating influence from Babylonian sources, and now it can be asserted as certain that at least the immediate basis of Manichæism was the religion of Zoroaster. Apart from the pronounced dualism, which is common to both the religions, the names bear witness to this. Here we find the whole mythology of the [231] Avesta reproduced. A fragment from Shapurakan, composed by Manes himself; makes mention of Mihir, and the demons Az, Ahriman, the Pairikas and the Azhidahaka. In a fragment which, according to the superscription, belongs to a hymn of Manes himself, he is named as a son of God Zarvan, who represents Time in Zoroastrianism and who in later times is exalted as the highest Principle. In a hymn, Fredon is invoked together with Mihir. Fredon is the Thrætaona of the Avesta and the Faridun of the Shahnameh. Many of the Zoroastrian angels, like Srosh and Vohumano, occur side by side with Jesus. For Manes claimed to be the perfector of Christianity. In the fragment discovered by Müller, Manes calls himself the apostle of Jesus, as has already been told us by Augustine. To judge, however, from the fragments, the syncretism of the Christian elements has not been perfectly achieved. There has been no complete amalgamation. The different layers of belief lie one over another. Thus the description of the end of the world in the Shapurakan presupposes the Day of Judgment and has a close connection with the words of the Gospel of Matthew. Further Christian influences are evidenced by reference to the history of the crucifixion and resurrection of Christ.

Manes acknowledged the Buddha as also a predecessor of his. Clear evidences of Buddhistic influence, however, only appear in the fragments belonging to later times, like the confession of sins. It is quite possible, therefore, that what we meet with here is a later development of Central Asian Manichæism. Probably here, in the ancient soil of Buddhism, it took the Buddhist colour, just as in the West it assumed a Christian tinge.

In their exterior get-up Manichæan MSS. are distinguished by the great care bestowed on them. Many are adorned [232] with pictures, which must be regarded as magnificent specimens of miniature-painting. This taste for artistic book ornament was a legacy from old Iran. Augustine, as we know, turned with flaming wrath against the bibliophiles. Manes’ name has been connected from ancient times with painting, and legend ascribes to him the knowledge of secret signs. In Persian he is known as Manes, the painter.

From the philological standpoint the Iranian writings fall into three groups. The first group is composed in a dialect which comes very near to the Pahlavi, the official language of the Sasanian empire. We know this language from a few inscriptions and texts of the Zoroastrian religion, and especially from a translation in it of the Avesta. Accordingly, the texts from Turkestan published by Müller and Salemann indicate an infinite advance of our knowledge. The writings on the monuments known up to now are wholly uncommon. They do not give back the pronunciation of the time, and they employ Aramæic cryptograms for ordinary words, so that, for example, people wrote Malka while they read Shah, or King. In the script of the fragments recently discovered this method is avoided, so that here for the first time we find an actual presentment of the proper Middle Persian language.

The second group is composed in the dialect of North­Western Persia, which no doubt was the language of the Arsacids who proceeded from these regions and who preceded in sovereignty the Sasanians. Andreas surmises that the so-called Chaldeo-Pahlavi, which appears in the inscriptions of the Sasanian kings, is identical with this tongue. He has now in hand a rich amount of inscription material for the investigation of the question, and we may hope in the near future to hear from himself the confirmation of this theory. [233]

The third group occupies the premier position in importance, if not in number. It is written partly in the Manichæan and partly in a younger alphabet, called the Uigurian. Andreas sees in this the Soghdian dialect. It was only an accident which has preserved for us in al-Biruni the names of the months current in this language. The discovery of the Soghdian has led to another important discovery. F. W. K. Müller has ingeniously succeeded in showing, that in the celebrated polyglot inscription of Kara-Balgassum, which informs us of the introduction of Manichæism into the land of Uigurs the difficult text in a character, which was up to now regarded as Uigurian, is in reality composed in Soghdian. He also demonstrates, that the Iranian terms in Chinese astronomical writings of the eighth century do not belong to modern Persian, but to the Soghdian idiom.

Another find furnishes a proof to the fact that Soghdian was used not only by the Manichæans, but was the common language of intercourse of all the Iranian inhabitants of Turkestan, while to Pahlavi was assigned the role of a written language.

Among the MSS. which are acquired in the northern parts are found pages in Syriac writing and language, which have been published by Sachau. They are connected with the hymns of Nestorian Christianity. The activity of the Nestorian missions, which, starting from Assyria and Babylonia, spread into the interior of China, is attested further by 12 leaves from a charming little book, the Pahlavi translation of the Psalms with the canon of Mar-Abba which to this day is in use in the Nestorian church. The MS., to judge from the characters must date from the middle of the sixth century. But the translation lies some 150 years before the oldest MS. of the Peshita Psalter and promises to prove of the greatest importance for the history of the text criticism of the Syriac [234] originals. Then, in Syriac writing, but in a language which, owing to certain peculiarities, can be designated as a younger phase of Manichæan Soghdian, considerable fragments relating to Christian confessions of faith, legends, and acts of the martyrs are found. The major portion has been edited by Müller. They show that the Christians employed the Pahlavi and the Soghdian languages for the spread of their doctrine quite as much as their Manichæan rivals.

Also the third religion, Buddhism, made use of the Soghdian for its propaganda. The Berlin collection possesses fragments of the Vajracchedikā, the Suvarṇaprabhāsa etc. The cave of Tun-huang is, however, a peculiar treasury of Buddhistic Soghdian texts which are written in a particular alphabet of Armaæic origin. Among the texts published by Gauthiot, the most interesting is that of the Vessantara Jātaka, the gem of didactic story-literature (forgotten in India but known to every child in Burma and Ceylon), which we find here in a new version. Gauthiot has deciphered also the oldest form of this writing, as well as language, which was found by Stein in the desert between Tung-huang and Lobnor, along with Chinese documents of the beginning of the first century. Above all, there can be no doubt as to the character of the Soghdian. It was the language of the Iranian population of Samarkand and Ferghana, and was spoken as a kind of lingua franca from the first to the ninth centuries in Turkestan and farther in Mongolia and China. From a Buddhist MS. of Stein’s, it appears that it was written in Singangu. An echo of the Soghdian is still found in certain modern dialects in the higher valleys of the Pamir. Especially the Yaghnobi lay claim to the designation of modern Soghdian.

When it is further mentioned that the Stein collection also contains a document in Hebrew letters, and written [235] according to Margoliouth, in the year 100 of the Hijra, the most ancient Judo-Persian piece of writing, which at the same time is also the most ancient piece of writing in modern Persian, it must suffice to measure the importance of the Turkestan finds for the Iranist; and yet Turkish philology is in greater debt to the country. Upto now there was almost an entire dearth of its ancient literature. The earliest Turkish book known to us was the Kutatku-bilik, written at Kashgar in 1069. Now we have acquired an ample collection of MSS and block-prints in the land of Uigurs, which is 200 years older in language and in character than that book. A splendid number of old Turkish texts which, however, represent only a small portion of what we possess, have been edited by Radloff, Thomsen, Müller, Le Coq and Stonner.

III. Buddhist Sanskrit literature

The varieties of scripts employed in these manuscripts are as curious as their contents. We meet with a Manichæan Estrangelo, the Uigurian alphabet, the Brahmi, the Runes of a particular kind, (which the genius of Thomsen was able to read twenty years ago for the first time on the stones at Orkhon andYenissei). From the standpoint of their contents the texts fall into three divisions. The Christian literature has up to now been very sparsely encountered, the largest document dealing with the adoration of the Magi, who are here described after the manner of the Apocrypha. Among Buddhist texts, those of a comparatively later date occupy a large place – the Saddharmapuṇḍarīka, the Suvarṇaprabhāsa Sūtra, (of which both Berlin and Petrograd boast of complete texts), passages from the diaries of travellers, from the peculiar species of literature, not always of a cheerful nature, the Dhāraṇīs, and the penitential formulas [236] with their lively portraiture of all manner of imaginable sins. They bear a strong resemblance to the Zoroastrian Patets. Then there are again fragments of works with interlinear versions, which are not without value for the originals, since though they are somewhat younger in age they reflect the oldest accessible texts. From the standpoint of history and literature the most interesting of our acquisitions are the miscellania of Indian legends. Who could have ever conceived an expectation of coming across in Turfan the old legends of the Mahābhārata related by Bimbasena or more correctly Bhimasena and his fight with the demon Hidimba, or of the svayaṁvara of Indian princesses? We have confessional formulas of the Manichæans, which are without doubt framed after the Buddhist exemplars, like the Khuastuanift which is valuable even in its dogmatic contents, and another which witnesses to a considerable tolerance of Buddhism. In this text, in the same breath, are enumerated the sins committed by one against one’s own brother in religion as well as the sins shared in Vihāras dedicated to Śākyamuni! Further, our inventory of the treasure trove has to notice fragments of hymns, sermons, divine judgments, and dogmatic transaction; next, a small complete book of prognostications or a dream book in the Rune script. It bears resemblance to similar products of China, but is of Manichæan origin. A special value is to be ascribed to two leaves from Berlin which from their exterior can be marked as Manichæan and not Buddhistic. The first relates to the setting out of the Bodhisattva or as he is here called, the Bodisav, on the path of renunciation, and those who meet him. The other contains the revolting story of the youth, who in his intoxication embraces the dead body of a woman. It is of Buddhistic origin and S. Oldenburg has shown, that it occurs as the first parable in the Persian version of the legend of Barlaam and Joasaph. This discovery as good as confirms the conjecture [237] of Müller and Le Coq, to which the peculiar name Bodisav had led them, that here we have to do with the vestiges of the Manichæan version of the celebrated Buddhist romance. But it is not at all impossible, that the original was a Manichæan work possibly in the Soghdian language. It would constitute a remarkable instance of involuntary syncretism, if the Manichæans had contributed to the turning of the founder of Buddhism into a Christian saint.

There is hardly a single nation among those of the East Asiatic continent possessing any civilisation of its own, which has not left literary traces in Turkestan. Müller has in certain fragments recognised the script employed by the Hephthalites or White Huns on their coins. We have Mongolan letters and xylographs in the enigmatical Tangutian written language. Tibetan manuscripts are numerous of which only a few, the fragment of a sūtra and a couple of religious songs, have been brought out by Barnett and Franke. The number of Chinese writings is enormous. The oldest of these excavated from the sand by Stein are now before the public in a magnificent work by Chavannes. Of the paper manuscripts a few go back to the second Christian century. They are at any rate the oldest paper documents in the world. A large majority of the documents are on wooden tablets. Some are on bamboo chips; they mark the condition of the oldest Chinese books. The wooden pieces, the oldest of which date from 98 B.C. come from the archives of the garrisons stationed here in the outermost west of the empire on the Great Wall. Here are gathered the detailed particulars regarding the daily life of the military colonies in the first centuries of Christ. They deal with the duties, the wages, the equipments of the soldiers, an optical telegraphic service, a postal department; and, a complement to the picture of the realities of the day, a poem of later days describing the miseries and dangers of the frontier legions [238] guarding against the barbarians of the West. The mass of later Chinese manuscripts seems to belong to works of the Buddhist canon and to business documents. A stranger has sometimes strayed into the collection as is shown by the “Lost Books in the Stone Chamber of Tun-huang,” published five years ago in Peking. It is a pleasant sign that China is willing not merely to guard the ancient literary treasure entrusted to her, but also to make it useful.

For us, in India, the manuscripts in Indian languages are of supreme importance. Historic interest is claimed before all by documents on leather and wood discovered by Stein on the Niya river. They contain, as is evidenced by the publications of Rapson and Boyer, dispositions and reports of local authorities, instructions, regulations, official and private correspondence – all inscribed in the Kharoshti script and drawn up in a Prakrit dialect. The date of the Prakrit documents is fixed by the Chinese wooden tablets which have been mixed with the latter, and one of which is dated A.D. 269. In the third century, therefore, there were Indians in Khotan of Gandhāra origin, who were living mixed with a Chinese population. It is, therefore, not improbable, that an historic fact lies at the basis of the legend, according to which Khotan in the days of Aśoka was colonised by Chinese emigrants under the banished son of the Emperor as well as by the inhabitants of Takṣaśilā, whom the Indian king, wounded over the blinding of his son Kunāla, which they had not prevented, had ordered to be banished to the deserts to the north of the Himālayas. In the circle of these Indian colonies lies also the Kharoshti manuscript of the Dhammapada which is known after Dutreuil de Rhins, Professor Lüders thinks that it is by no means a private anthology, but the remnant of a particular tradition of the [239] word of the Buddha, which up to now has undoubtedly remained the only one of its kind.

Since the time of Pischel, who deciphered the first pages of the xylograph of the Saṁyuktāgama, the remnants of the Buddhist canonical literature in Sanskrit have been infinitely multiplied. What up to now has been placed before the public out of the Vinaya and Dharma of the Buddhist Sanskrit canon by Sylvain Lévi, Finot and de la Valée Poussin is only a small portion of the salvage. Of the Udānavarga, which seems to have been unquestionably the most favourite Sanskrit Buddhist work, 500 leaves are preserved in the Berlin collection alone, out of fragments and leaves belonging to some 100 manuscripts, so that the text is almost completely restored. Pischel recognised that these vestiges belong to the canon of the school of Sarvāstivādis lost in the original Sanskrit. He already noticed that the Sanskrit texts were not translations from the Pāḷi canon, which is the only canon preserved intact to us. A penetrating research has revealed, that both the Sanskrit and Pāḷi canon are traceable to a common source, which, as is proved by mistakes in the translations, was drawn up in the Eastern dialect which was spoken as the common idiom in the territory of the Buddha’s activity. This is an event which is of decisive consequence in the history of Buddhism. We are now in a position to restore the Sanskrit canon from the debris of tradition. It existed in the pre-Christian centuries in Magadha. That, however, is not equivalent to saying that we have come upon the original word of the Buddha. What the Buddha himself exactly taught will always remain a subject of speculation, although Professor Lüders believes we are not yet justified in resigning ourselves to the position of ignorabimus. That, however, which the Church thought He taught at a time to which no direct documents go back, is now in our hands, thanks to the Turkestan discoveries. [240] Another region in literature has now been made accessible from this quarter – the pre-classical Sanskrit poetry. Thirty years ago the Kāvya appeared to begin with Kālidāsa, who was placed in the sixth century. Before that seemed to lie centuries of complete sterility and Max Müller coined the phrase about “Sanskrit renaissance.” To-day we are positive that Kālidāsa lived in the beginning of the fifth century, that his name signifies the zenith of courtly poetry, and that it was preceded by a spring. Inscriptions and a couple of lucky discoveries in India have given us an idea of the beginnings of the Kāvya. Turkestan intimates to us the existence of an unsuspected wealth of hymns, epics, romances and anthologies which in the majority belong probably to this period. The material is always religious, but the form is that of the secular Kāvya. This differentiates the poetry from the old Buddhistic, though the old Church did not by any means stand hostile to poetry.

[The present writer may be allowed to dwell for a moment – a moment only – on the brilliant confirmation of the discovery of the Buddhist canon in Sanskrit. A short eight years ago his refusal to look upon Pāḷi as the prime word of the Buddha, and Sanskrit Buddhist books as later fabrications, drew on him a storm of indignation from Burmese monasteries. Unfortunately for the time being the excavator’s spade is left for the shrapnel; else it were easy to make a present to the Shwe-da-gon shrine of an anthology of Sanskrit Buddhism, as voluminous as any in Pāḷi, issued from Leipzig or New York.]

IV. The hiatus in classical Sanskrit literature supplied

People appropriated the popular specie of poetry called the Gāthās by putting over it a Buddhistic veneer. The [241] first age of profound religious passion gave rise to a number of poets who, however, had not the ambition to hand down their names to posterity. Many of the strophes which were placed in the mouth of the Buddha himself or his disciples are among the finest produced by the literature of any age. But only when Sanskrit was given the position of a church language, instead of the popular dialect, doubtless with a view to a wider spreading of the doctrine, it was, that poetry began to be composed according to the rules of the Sanskrit court singers. Our manuscripts prove, how much, under the influence of this artificial poetry, gradually the ear of the monk himself in the Turkestan monasteries was refined. Scholars were constantly at work improving upon the old translations of canonical works which were in many ways crude and unpolished. They laboured to reduce the text in language and metre to the stricter requirements of later ages.

Two names belonging to this early period are mentioned in the Middle Ages with enthusiastic admiration, Mātceta and Aśvaghoṣa. Both belong, as it seems, to the beginning of the second century. Mātceta’s fame is based on his two hymns to the Buddha, which, according to I-tsing, in the seventh century every monk in India learnt by heart, whether he was attached to the Hīnayāna or the Mahāyāna, and gave rise to the legend that the author in his previous birth had rejoiced the Buddha with his songs as a nightingale. They were up to now known only from Tibetan and Chinese translations. From the fragments in the Berlin collection about two-thirds of their text has been restored. The work of Mātceta has great value in the history of the Sanskrit literature as the earliest example of Buddhistic lyrics; although the enthusiasm, with which the Chinese Buddhist-scholar and translator I-tsing speaks thereof, is not altogether intelligible to us. Dogmatic punctiliousness can scarcely [242] compensate us for the monotony with which synonym after synonym has been heaped. Also the alaṅkāras which constitute the regular decoration of a Kāvya are only sparingly employed. Incomparably higher as a poet at any rate stands Aśvaghoṣa. Fragments of his epic, the Buddhacarita and the Saundarananda in the original Sanskrit are found in Turkestan. Here we have also palm leaves eaten up and ruined on which was inscribed the Sūtrālaṅkāra which is at present known only from its Chinese translation. A French version of the Chinese rendering was done by Hüber. The ruined remains, however, give us an idea of the style of Aśvaghoṣa. We likewise possess a wholly unexpected fund of remnants of dramas of which at least one in the colophon is expressly designated as Aśvaghoṣa’s work. One of the two palm leaf writings, in which it is preserved to us, is a Palimpsest prepared in Central Asia. The other was probably written in northern India during the lifetime of the poet. It represents the oldest Brahmi manuscript we know. One leaf has come out of a dramatic allegory, in which Wisdom, Endurance, and Fame entertained epilogue or an interlude. A fragment represents a comic piece, in which the principal part seems to have been played by a courtesan. The drama, which undoubtedly is a production of Aśvaghoṣa, treats of the story of the two chief disciples of the master, Śāriputra and Maudgalyāyana, up to the time of their conversion to Buddhism. The fragments do not suffice to enable us to judge of the individuality of Aśvaghoṣa, although they furnish valuable suggestions for a general history of the Indian theatre. We here come across, apart from divergences of little consequence, forms as in the classical period. The speeches are in prose intermixed with verse. The women and the inferior dramatis personae speak a Prakrit dialect, which undoubtedly stands here on a more [243] ancient phonetie level. The comic person of the piece, the vidūṣaka, is also here a Brahman perpetually suffering from hunger in the company of the hero, and the manner of his jokes is the same as in Śākuntalā. All this demonstrates that the Indian drama at the close of the first Christian century was fully developed in all its characteristics and this has been completely established by the discovery in Southern India of the dramas of Bhāsa, by Gaṇapati Śāstri. Bhāsa is one of the poets mentioned by Kālidāsa as his predecessor.

It is a variegated picture this, presented to us by research in Turkestan. It is all still almost in confusion, the flickering light of accident. It will require years of labour before we are able to judge of the whole huge collection. The question with some is, whether the results will be commensurate to the labour. There are many in the West who have hardly any appreciation for the work of scholars engaged on the investigation of peoples and speeches of Southern and Eastern Asia. But the sinologues’ views at least must count. Chinese is a “colonial language.” The Sanskritist, however, is something more than a tranquil man who worships dead deities worlds apart. These gods are not dead. The knowledge which Gautama Buddha acquired in the holy night under the Bodhi tree is still the credo of millions of mankind, and thousands and thousands of lips still repeat the prayer at sunrise composed by a ṣi thousands of years ago. Nor are those countries far from us. Only 18 days’ journey divides the heart of Europe from Colombo, in whose harbour steamers from their journey to the ends of the earth take shelter. The world has become narrower, the peoples of Asia have been brought close to us and will be brought still closer. Whether this will be peaceful or will lead to strife this nobody knows. It is nevertheless our duty to endeavour to study the ancient systems of culture, to endeavour to appreciate [244] them in the only possible way – that of historical research. In the history of this research the discovery of the Ancient and Middle Ages of Turkestan constitutes only a single chapter but that happens to be one of the most important.