1838.] Revision of the Bactrian Alphabet. 639 



I have long been pledged to my readers (and to the critics of the 

 Meerut magazine in particular) to give them a new alphabet for these 

 Bactrian legends, and I think the time has now arrived when I may- 

 venture to do so ; or at least to make known the modifications which 

 have been elicited by the abundance of fresh names and finely preserved 

 specimens which have passed under my eye since that epoch. It must 

 be remembered that the only incontestable authority for the determina- 

 tion of a vowel or consonant is, its constant employment as the equiva- 

 lent of the same Greek letter in the proper names of the Bactrian kings. 

 Beyond this we have only analogies and resemblances to other al- 

 phabets to help us, and the conjectural assumption of such values for 

 the letters that occur in the titles and epithets of royalty as may furnish 

 an admissible translate of the Greek in each and every case. 



It will be my object presently to shew that this can be done, as far as 

 the coins are concerned, by means of the Sanskrit or rather the Pali 

 language ; but in the first place it will be more convenient to bring 

 forward my revised scheme of the alphabet as far as it is yet matured. 

 Unfortunately the exceeding looseness of orthography and kalligra- 

 phy which could not but prevail when one foreign language, (for such 

 it was to the Greek die-cutters), was attempted to be rendered by the 

 ear in another character, equally foreign to the language and to the 

 scribes, that with abundance of examples before me it is impossible to 

 select the true model of some letters for the type-founder ! 



I begin with the initial vowels : 



9, a. This symbol continues to occupy the place of the vowel a in 

 all the new names, lately added to our list, beginning with the Greek 

 A, of which we have now no less than seven examples. The other 

 short initials appear to be formed by modifications of the alif as in the 

 Arabic : thus. 



f, 1 e, is constantly employed for the e of Greek names. 



a w, is found following it in the word Eucratides, as though put for 

 the Greek r, but other evidence is wanting. 



9-, i ? though seldom met with on the coins is common in the in- 

 scriptions, and by analogy may be set down as i. 



.9 and % a, an, is employed in words beginning with an. 



The medials seem to be formed in all cases by a peculiar system of 



names. In tbe modifications I now propose, however, I do not borrow one letter 

 from his list, because in fact he has followed quite another track. His reading of 

 'PiTTlU is, myrvd, a Syriac word I believe for prince or noble. It was this 

 which led to the expression of doubt of my own former alphabet, and to the just 

 satire thereon in the Meerut Magazine. 



4 L 



