________________
EPIGRAPHIA INDICA.
[VOL. XVIII.
A fourth attempt was made in the records of Ysamotika and his successors, viz., by means of the compound ys; and this same device became the rule in Ancient Khotani.
270
Erjhana accordingly represents an attempt at rendering the sounds of erzana or erzana, and there can be no doubt that in this form we have to see the same word which occurs in Ancient Khotani as alysānai, eysamnai, which is used to render Sanskrit kumara. The l before the voiced s in this word, as also in malysaki, is certainly a later development in Khotant, which, on the whole, agrees with Old Iranian in using r for 1 throughout.
M. Boyer, who read ejhshuna, saw in this word the name of a prince and thought that this name was an Iranian compound containing izad as its first part. But no similar name is known to exist, and in my opinion, my explanation of the word is absolutely certain.
If now erjhana belongs to the language which I have called Khotani, it would be natural to infer that the individual characterized as erjhana, i.e., kumara, belonged to the tribe or tribes who used that form of speech. His name must be contained in the fourth and fifth. aksharas of 1. 5, which M. Senart read as kapa, while M. Boyer, who admitted that the first akshara looks like ka, suggested to read [bhu]pa. There cannot be any question about the nature of the first letter. It is ka, and it is so clearly cut that we have no right to correct it. The second akshara is certainly pa, but underneath there are distinct traces of an akshara which I have formerly suggested to read as fa or sha. It now seems to me that it must be read as sha, there being distinct traces of a vertical below the semi-circle, i.e., of the letter sha. I still think that this kapsha cannot be anybody else than the Kushana prince whose name is written in many different ways, kadphises and kadaphes in Greek letters, and kasa, kaphsa, kadapha in Kharoshṭhi. With regard to the use of sh when the Greek rendering has we may compare Kharoshthi ayilisha, Greek azilises.
We know that the national tongue of the Kushanas was identical with, or closely related to Ancient Khotant, and Sir John Marshall's discoveries at Taxila have shown that Kujala Kadphises followed almost immediately after Gudufara as ruler of that place. Sir John writes*: "After the death of Gondophernes his empire was split up into smaller principalities, and it was then that Hermaeus and Kadphises I appear to have made their successful invasion of Gandhara and Taxila." I therefore feel no hesitation in identifying the prince mentioned in our inscription with Kadphises I, since he is characterized by a Khotani title and bears a name which is almost identical with the names used in his coin legends.
The remaining portion of the inscription is quite clear. After Kapsha follows the genitive termination sa, separated from Kapsha by a wide gap, and further puyae 'madu pidu puyae. The curious flourish to the left of the final è is, as stated above, probably accidental.
The inscription carries, as we have seen, a double date: in the 26th year, during the reign of Gudufara, and in the 103rd year. The latter date has usually been referred to the Vikrama era. Sir John Marshall has tried to show that that reckoning was instituted by Azes, and Professor Rapson endorses that view. I am unable to accept it, for several reasons. Sir John's theory is based on his interpretation of the word ayasa occurring after the figures denoting the year in the Taxila inscription of the year 136, which he takes to be the genitive
1 Cf. Lüders, S. B. 4. W., 1912, pp. 407 ff. The recent attempt made by Mr. N. B. Divatia, J. Bo. Br. R. 4. S., XXVI, pp. 159 ff., to vindicate the rendering of the compound as ghs is a failure. The learned anthor has failed to recognize that ys is the regular sign for voiced & in Ancient Khotaui.
28. B. A. W., 1916, p. 801 f.
Cf. Konow, Z. D. M. G. 68, pp. 85 ff.
• J. R. 4. S., 1914, pp. 979 f.
6 J. A. R. 8., 1914, pp. 973 ff.
The Cambridge History of India, Vol. 1, pp. 571 ff.
