________________
No. 16.)
THE “PYU" INSCRIPTIONS.
131
The vowel au is used in the Myazedi inscription 88 & variant of ; but I must admit that it is odd that both si and sar (which is, moreover, a doubtful reading) should appear in such & short document as C. Also the word tru is not quite certain, that portion of the rubbing being by no means clear; it might conceivably be tra, though I prefer the reading tre. I assume that "five" has two forms, the shorter one being used as a multiplier. If these more or less hypothetical conclusions are correct, these inscriptions have by & fortunate concurrence of circumstances given us a series of " Pyu" numerals which is complete so far as it goes and seems to be in general agreement with the numerals of other Tibeto-Burman languages. But farther confirmation will of course be necessary before we can accept it as definitely established in every particular.
As stu8 means "year" it seems reasonable to suppose that de (or le, or whatever the true reading may be) and phuy stand for other divisions of time, probably "month" and "day" respectively for they also are followed by numerals, or words which we have found to form part of the numerical combinations used in connexion with sni8, or words used alternatively to such words. I take tio to be a postposition meaning "in", but I admit that there is some doubt as to this, and its use here seems rather capricious and irregular. The phrase-ta kio kha # sni8 common to all four texts is evidently a formula describing the type of yoar intended. As it is a constant it can only refer to some characteristic or quality common to all the years previously mentioned, and the most natural view of it would seem to be that it defines them by reference to some fixed point, in other words it denotes some era. Then follows another chronological phrase beginning with snis. This is wanting in Dl and I take it to refer to the ages of the deceased persons commemorated in these epitaphs. If urn D contained the mingled ashes of several members of the family, that might be a good reason why this phrase is not found in Dl. At any rate these numerals have no constant relation to the preceding sets of numerals, and they are too high to be probable lengths of reigns.
Let us now tabulate these chronological data. Assuming the first set of numbers to be referable to some fixed point or era, the inscriptions will fall into the order D1, C, A, B1. There is of course nothing in the texts (80 far as we can understand them at present) to determine what fixed point or ora is implied. But let us assume, for the sake of convenience, that it was the ordinary Burmese era of 638 A.D. We can then make out the following ohronological table :
(1) year 35 (673 A.D.); Sariyavikrama's relative or relatives died; (2) year 50 (688 A.D.), 5th month ; Sariya vikrama himself died, aged 64 years; (3) year 57 (695 A.D.), 2nd month, 24th day; Harivikrama died, aged 41 years, 7 months
and 9 days (4) year 80 (718 A.D.), 2nd month, 4th day; Sihavikrama died, aged 44 years, 9 months
and 20 days. From this it is obvious that the three personages named could very well have been grandfather, father and son occupying the throne of the Prome monarchy in lineal succession. Of course we are not entitled to assert that this really was the fact : but as a working hypothesis It seems to be consistent with the evidence at present available.
So too as to the era, the most one can say is that it is not an impossible one. But there is very little to guide us as to the age of these inscriptions. We know that the Myazedi record is only about 800 years old, yet its alphabet does not differ very materially from that of our arb-inscriptions. But then it shows signs of great archaism, the leading instance being the anchor-shaped subscript y, which was obsolete in India after the 4th oentury. This argues Buch a conservative attitude on the part of “Pyu" scribes that I cannot understand how anyone can profess to date their inscriptions by palæographical evidence alone. We know from history that the Btırmose of Pagan conquered the South somewhere in the 11th century and therefore one is naturally disposed to date the Vikrama dynasty of Prome before that
