120 INDO-IRANIAN LANGUAGES 



there is still in existence the large encyclopedia called Tanjur, 

 comprising a collection of 200 translations of Sanskrit works, so 

 faithfully rendered into Tibetan, that where the corresponding 

 Sanskrit text has been preserved and it happens to contain a lacuna, 

 the missing Sanskrit words can be restored with certainty. Among 

 the Sanskrit originals discovered in the countries where these trans- 

 lations have been preserved is the grammar of the Buddhist Can- 

 dragomin, found in Nepal only a few years ago. The Tibetan trans- 

 lation enabled Professor Liebig, in his edition of the text, to emend 

 successfully some passages which were defective in the original. 

 Diligent search will, it is to be hoped, result in the recovery of many 

 Sanskrit works (at least as far as Buddhist literature is concerned) 

 of which at present only the Tibetan or Chinese translations are 

 known. Much might have been expected in this direction from the 

 British occupation of Lhassa, where the monasteries must contain 

 many manuscript treasures, but the absence of any Sanskrit scholar 

 in the expeditionary force will, I fear, preclude the discovery of valu- 

 able Sanskrit manuscripts such as would probably have resulted, 

 had an energetic scholar of the type of my friend Doctor Stein, 

 accompanied the British troops. 



Patient search may also lead to the recovery of some of the originals 

 of the numerous Sanskrit Buddhist works which were translated 

 into Chinese from the first century of our era onwards. Much may 

 be hoped in this direction from the labors of the Society of Oriental 

 Research recently founded in Japan, one of the objects of which is 

 to examine systematically the monasteries and temples of China 

 and Corea with a view to the discovery of Sanskrit manuscripts. 

 What is possible in this way will be apparent from the following 

 example. By the year 1879 all knowledge of Sanskrit had died out 

 in Japan. In that year two young Japanese Buddhists, named 

 Nanjio and Kasawara, were attracted by the influence of Max Miiller 

 to learn Sanskrit at Oxford, in order to study Buddhist texts in the 

 original Sanskrit as well as in Chinese translations. Through these 

 young scholars (whom I taught Sanskrit during the first year of 

 their studies), Max Miiller caused investigations to be made in 

 Japan, which soon led to the discovery, in an ancient monastery, 

 of a Sanskrit work dating from the sixth century A.D. and at that 

 time (1880) the oldest Sanskrit manuscript known. 



The works of the Chinese Buddhist pilgrims who visited India in 

 the fifth and seventh centuries A.D. mention a large number of 

 geographical and personal Indian names, the identification of which 

 is of great importance to Indian history. The transformations which 

 these words undergo, owing to the widely divergent character of 

 Chinese phonetics, often render their identification purely conjectural 

 in the present state of our knowledge. An important problem here 



