Page:JSS 006 1b Bradley OldestKnownWritingInSiamese.pdf/14

 That is,—The Lao, having no voiced sounds corresponding to g and gh, utilizes these letters as tonal variants of the unvoiced pair k and kh taken without change from the Indian alphabet. The same thing is done with the corresponding pair in each of the five series. And the first pair—simple and aspirate—are uniformly "high" letters, while the second pair are uniformly "low." Thus without help of any accents beyond the original mai ek (◌่) and mai tho (◌้) of the inscription, a fuller scheme of tones than that of modern Siamese is provided for, without duplication of letters, without lack, and with fair balance of parts. Such symmetry and adaptation of means to ends surely is not the result of accident. If we make a similar comparison in the case of the Siamese, we find—

Two things, apparently, have happened to the Siamese version of the scheme:—1) No. 1 of each group has lost its power of indicating "high" tone, and has gone to swell the list of imperfectly equipped "middle" letters: and 2) No. 3 of each group has taken on aspiration, and thus merely duplicates No. 4.

An outcome such as this is perfectly intelligible and natural if the original were something like the Lao. But the Lao scheme could have hardly have been developed naturally from a scheme originally like that of present Siamese. And if our present twice, nay thrice, involved scheme of Siamese tonal notation—the dismay and confusion of all students,—together with the absurdly inflated consonantal alphabet which is part and parcel of it, were all really the work of Prince Ram Khămhæng it might well bear the palm among what Professor Whitney has called "devices of perverse ingenuity."

As might have been expected, the working of the new

scheme was not always sure. In the inscription a number of uncertain or variant spellings are