________________
26
THE INDIAN ANTIQUARY
[MARCH, 1919
Characters that do not tally with characters of any period yet known, that is to say, are unknown, cannot be necessarily considered archaic. An unknown thing cannot be recognised as archaic until its affinity to something that is known to be archaic is established. The principles of epigraphic evolution cannot be very different from the principles of organic evolution. In the organic world if points of similarity are noticed in the structures of two species of animals, the species with the less developed structure is either recognised as the ancestor of the species with more developed structure, or both the species are traced to a hypothetical common ancestor. So two known quantities are necessary for postulating an unknown third, either as an intermediate form or a common source. The process of evolution of an isolated species whether in the organic or in the epigraphic world cannot be traced backward with the assistance of imagination only.
The theory regarding the origin of Brahmi lipi that now holds the field is that of Bühler according to which it is derived from the oldest form of North Semitic alphabet which was introduced into India by traders about 800 B.C. But this theory is not universally accepted. Cunningham never subscribed to it. Another eminent authority, Fleet, suggests that either the oldest Semitic alphabet and the Brahmi lipi were derived from a joint original source," or Hindus "were the independent inventors of that which was emphatically their national alphabet."1 The relationship between certain Brâhmî letters and old Semetic letters is undeniable, and I prefer the first alternative proposed by Fleet to the second. But even if we accept the latter view and altogether ignore Semitic forms in our investigation of the origins of the Mauryan Brahmi alphabet it is impossible to recognise the letters of the Patna image inscriptions as fifth century (B.c.) predecessors of the third century B.C. forms without independent evidence. Not only has Mr. Jayaswal failed to offer any independent evidence to prove his case, but his statement that the characters used in these two short records do not "fully tally with characters of any period known to Indian epigraphy" appears to be absolutely wrong. I hope to show that the characters of the epigraphs under discussion nearly fully tally with the Brahmî characters of the Kushan period.
A
Cunningham-Yakhe Sanatananda.
Jayaswal-Sapa (Shapa ?)-Khate (Khete ?) Vasa (Vesa ?) Nandi. (p. 95). (1) Mr. Jayaswal's Sapa or shapa is a clear Ya of the Kushan period with equal verticals, and an angular right limb and a semi-circular left limb (CLI). Cunningham reads the letter correctly and any one can ea silyrecognise it from the good facsimile published with Mr. Jayaswal's article.
(2) Mr. Jayaswal's method is best illustrated by his remarks on the second (his third) letter. He agrees with Cunningham in reading it as kh. Like kh in the inscriptions of the time of the Kushan kings and in the Girnar inscription of Rudradaman it consists of a triangle with a hook turned to the left (2). Quite oblivious of this Mr. Jayaswal writes, "The third letter, kh, again, has an older feature. The body is formed of four lines, which becomes round or tends to disappear in Aśoka's time." (p. 94.) I do not see the medial e with Kh and so I read the two first letters as Yakha (Yakshu).
(3) The third letter which Cunningham reads as sa and Mr. Jayaswal as ta is a doubtful one. Its left leg is a little curved like the left leg of a sa, but its right leg looks more like the right leg of va. With Cunningham provisionally I propose to read this letter as sa.
1 Encyclopædia Britannica, 11th ed., Vol. XIV, p. 626.