________________
NOVEMBER, 1924 )
KOTTAYAM PLATE OF VIRA RAGHAVA CHAKRAVARTI
247
exactly as it is in Tamil Grantha. We do not find gh in any inscriptions of Malabar, except in the Vira Raghava plate. The symbols for gh in Malayalam and Tamil Grantha are almost the same.
In the Vira Raghava plate we have m twelve times and mm once. The symbol m is twice written exactly as it is in Malayalam. It is written ten times, and mm once quito differently, just as we find in all other inscriptions.
As for , it is not Arya-eluttu. This character is not necessary for writing Sanskrit, of which Arya-oluttu is the alphabet. It was, no doubt, taken from Vatteluttu and adopted into the Malayalam alphabet. The symbol of our copper plate is not like the Malayalam, but like the Vatteluttu ? of Rajasekhara and Taņu Iravi Tiruvalla. The symbols & and u are added to it, just as in Malayalam. This character, therefore, was adopted into the Malayalam alphabet. No one can, however, say when it was adopted, or whether it was newly adopted, by the writer of our copper plate.
We have ! in the Tiruparappu copper plate of the ninth century (column 13). We find ? in the Jațila varman plates (c. eighth century) and others, almost like the Malayalam ạ.
As to the remaining characters a and a, we do not find any other inscription in Keralam in which these characters occur. But we find a symbol for d almost like its modern Malayalam equivalent in the inscription of Marañchataiyan (column 9). We find a and a in the eighth century inscriptions of Jaţila varman and Nandivarman in a slightly different form, i.e., almost like their Tamil Grantha equivalents.
Now let us take the symbol a again. In the plate showing the early Pallava Grantha alphabet of the seventh century (Trav. Arch. Series, Vol. I, p 222), the symbol a is almost like the modern Malayalam a, though all the other characters are far from being similar to their modern equivalents. In the Grantha alphabet of the eighth century (ibid., p. 223) the symbol a is far from being similar to the modern a, but some other characters approach their modern equivalents. Lastly in the table showing the Grantha alphabet of the last quarter of the eighth century (Ibid., p. 224), the character a by no means approaches the modern a; but many of the other symbols are just like their modern equivalente. From all this we find that the development of the Grantha alphabet was not uniform throughout Southern India.
So the argument that the Grantha characters of our copper plate are too developed to be ascribed to any date earlier than the fourteenth century falls to the ground. For we find them in the eighth century inscriptions of the Tamil country.
Evidence in support of an early date from the Grantha characters.--Let us see whether we can arrive at any positive idea regarding the date, from the Grantha characters used in this plate.
The poouliarity of the symbol ai should be noted. It is written twice with two symbols, called pulli, unlike its modern equivalent. Since we do not find that character in any other inscription, we cannot say when this gave .place to its modern form.
Let us again compare the characters of Vira Raghava's copper plate with those of other plates. The few characters we find in the Rajasekhara plate are, no doubt, more archaic than those of Vira Raghava's plate.
(1) Now compare the Vira Raghava plate with the Taņu Iravi plates. In Tâņu Iravi there are only very few Grantha characters. The characters for comparison are t, 1,8, and m. The symbols of Tånu Iravi is just like its modern equivalent. We have no pure : but stu in the Vira Raghava plate. The symbols of that stu is far from being similar to the modern which we
