` Or should we say that the Tibetan orthography was constructed in places of learning in Turkestan, for instance at Endere, and that the Tibetan dialect of Endere was very archaic, even more archaic than the present dialects of Purig and Baltistan ? No. Think of the contrast of dialects that we should then have to accept for Central Tibet and Endere—in Central Tibet a dialect not very different from modern Central Tibetan, and at Endere a dialect more archaic than even the classical language.
` I think a way out of this difficulty is suggested by Mr. Barnett in his article (Y.R.A. S., 1903, p. 112). We are simply compelled to consider the story of Srori btsan sgam po as legendary, and to ascribe a much more ancient date to the introduction of Buddhism into Tibet, as well as to the invention of the Tibetan orthography and characters.
` There is also the possibility that Tibet was in possession of an archaic sacred language from time immemorial, that it was this language which was first reduced to writing, and that this already sacred language was accepted as the language of Buddhism.
` Palaeographic Note.—It is possible that the dbu can alphabet of Tibet was preceded by the dbu med alphabet (Indian Antiquary, vol. xxxii. 1903, pp. 361 ff.), and also that the two songs were copied from a dbu med MS., in which d and it look entirely alike. Thus in No. i. 13 odor was wrongly written where
odod was meant. Of special interest is the form of •9, which is differentiated from only by a little stroke.
The same form is found in the Ladakhi rock-carvings.
`Sandhi Laws.—Although the genitives and instrumentals are not confused in the two songs, the Sandhi laws are in many cases violated. This need not necessarily lead us to the conclusion that Sandhi laws did not exist in those times. That they exist in the vernaculars of to-day is beyond doubt. But there are many persons at the present day who, although they observe them very well in their speech, do not trouble about them when writing. Then I may repeat another observation of my own. Just as the Sandhi laws of Leh and Khalatse are not entirely the same, so also in ancient times the Sandhi laws of certain districts may have been different, and monasteries, where people from various districts were gathered, may have been just the places to destroy them.'