________________
SEPT., 1919]
THE LAKSHMANASENA ERA
173
executed in the fourteenth year of Govindapâla, is much earlier in date than inscription I of the year 51. Before proceeding to discuss the very possibility of this theory I must examine the evidence of the palæography of the inscriptions, as Mr. Chanda lays much stress on it, and declares it to be of a very highly convincing character.
The palæographic consideration of Mr. Chanda is chiefly based on the examination of the two test letters d and p occurring in the following six inscriptions: the Bodh-Gayâ inscriptions I and II of the time of Aśokachalla; the Gayâ stone inscription, dated 1232 V.E.-A.D. 1175; the Edilpur grant of Vievarûpasena; a Chittagong grant, dated A.D. 1243; and the Assam grant of Vallabhadeva, dated A.D. 1184-5. Now, for a comparative study of letters which may be of any practical use for determining dates, it is not desirable that we should mix up inscriptions incised on different materials, e.g., stone, copper, etc., or inscriptions though on the same material, yet connected with different localities far removed from one another by long distances. This procedure, I may say, is certainly, what may be called scientific' and that it is so, is clearly borne out by such an expert epigraphist as the late Dr. Fleet, who has made similar remarks in another connection (JRAS., 1913, pp. 975-6). In view of this general principle of paleography I am compelled to reject the last three inscriptions of the above list, for, they are, in the first instance, all copper-plates and therefore, not calculated to furnish any reliable data with regard to the palæography of stone inscriptions; and secondly, inscriptions discovered in Dacca, Chittagong or Assam cannot be brought in a line with inscriptions discovered in Bihar. The real comparison of letters that might be safely instituted therefore, is virtually confined to the first three records which are all on stone and belong to one and the same locality. Now, according to Mr. Chanda, the lettere p and d in the Gaya stone inscription represent the old Nâgarî type and those in epigraphs I and II almost resemble the modern Bengali specimens of the same letters. I quite agree with this observation, but cannot endorse the opinion, that the aforesaid appearances of letters only would justify us to fix in any way the age of the inscriptions, viz., that Nos. I and II are later in date than the Gaya stone inscription. One characteristic of the palæography of North-East India inscriptions from circa A.D. 1050 onwards is that they contain a mixture of Nagari and later Bengali forms. Curiously enough, we find the Nagari and the later Bengali forms of some letters used side by side not only at one and the same period but also at one and the same locality. Let us take, for instance, the case of letters v, land 8. The v of the Bodh-Gayâ inscription of the year 51 has practically no difference with a Bengali v of our own period. But strange to say, in the inscriptions of the years 74 and 83, the letter clearly represents its Nâgarî prototype. Exactly similar is the case of the letter I which is proto-Bengali in the first, and Nagarî in the second and third, inscriptions. Again in inscription I we have a Nâgari 8, in inseription II it is of proto-Bengali type, but in inscription III which is ex hypothesi later than II the old Nagarî type is again met with. The case of the two letters d and p is also not different. In inscription I, d represents an advanced type of the letter, and there is a close resemblance between this and modern Bengali d. In the Gayâ stone inscription referred to above, the d is doubtless of the Nâgart type; but then, this type we also notice in inscriptions II and III. With regard to the letter p, it must be admitted that the proto-Bengali type alone occurs in the three Bodh-Gayâ inscriptions of the years 51, 74 and 83. But, from this if we infer that the Nâgarî p was not in general use in the locality during this period, we shall commit a serious mistake, because, in a Gayâ inscription, which like inscriptions I
8
7 Banerji, Mem. ASB., Vol. V, No. 3, p. 109 and Plate XXVIII. 8 Ante, Vol. X, p. 342 and Plate.