________________
NOVEMBER, 1906.] THE ORIGIN OF THE DEVANAGARI ALPHABET. 313 =
• The alphabetic letters which bear the names of those and other gods or goddesses form the very bodily frames of those and other gods or goddesses. .
Thus, for example, the mantras, klim, called the bijákshara of Kâma, and hrim, the bijdkshara of Sakti, constitute, as in Plate IX., the forms of Kama and Sâkti. The same result would ensue in all cases of such mantras as are not later manipulations of ignorant mystics. Thus, it is only in the identity of the Tantric hieroglyphics with the letters of the Devanagari Alphabet that we can find satisfactory explanations for almost all kinds of Tantric technical terms and speculations.
Besides the ocular and documentary evidences proving the indigenous origin of the Devanagari Alphabet, as pointed out above, there is also the evidence furnished by the critical analysis which James Prinsep made of the alphabet of the edicts of Asoka. It is very well known that, but for his admirable skill in deciphering the then unknown alphabet of the edicts, the history of India, broken and unreliable as it is, would have missed its basis altogether. It is really astonishing and admirable that his critical analysis of the alphabet of the edicts of Asoka should have enabled him to arrive at almost the same primary letters that, as has already been seen, were first selected with no appreciable modification from among the hieroglyphics and next formed the basis for the complete manipulation of the rest of the letters by differentiation. As his observation (pp. 474–6, Vol. VI., J. A. S. B.) throws a flood of light on the process of derivation of the rest of the letters, and also on the peculiar compact forms of the letters in general, his remarks are quoted here in full :
“There is a primitive simplicity in the form of every letter, which stamps it at once as the original type whereon the more complicated structure of the Sanskrit has been founded. If carefully analysed, each member of the alphabet will be found to contain the element of the corresponding member, not only of the Devanagari, but of the Kavauj, the Påli, the Tibetan, the Hale Canars and of all the derivatives from the Sanskrit stock. This is not all: simplification may be carried much further by due attention to the structure of the alphabet, as it existed even at this early stage, and the genius of its construction, ab initio, may in some measure be recognised and appreciated. First the aspirated letters appear to have been formed in most cases by doubling the simple letters; thus, b, chh, is the double of d, cha; o, th, is the double of C,!; D. dh, is the half of this; and th, is the same character with a dot as a distinguishing mark. This may account for the constant interchange of n o and in the inscriptions. Again , dh, is only the letter produced from below; - if doubled, it would have been confounded with another letter (the b). The aspirated b, pha. is merely the b, pa, with a slight mark, sometimes put on the outside, either right or left; but I cannot yet affirin that this mark may not merely denote a duplication of the letter rather than an aspiration, -if, indeed, the terms were not originally equivalent ; for we have just seen the loubling of the letter made to denote its aspiration. The kh seems formed from the g rather than the k. The gh and jh are missing as in Tibetan, and appear to be supplied by y aud chh respectively. Bh is anomalous, or it has been formed from the d by adding a downward siroke.
"Again, there is a remarkable analogy of form in the semi-vowels,r, r. 1,, 1J J, L, which tends to prove their having been framed on a consistent principle. The first hardly ever oceurs in the Delhi inscription, but it is common in that form in that from Girnar. The Lih, is but the J reversed : the ti, so peculiar to the Sanskrit alphabet, is formed by adding the vowell to ther, thus P.
"As far as yet known, there are only one hand one s: the nasals and sibilants had not therefore been yet separated into classes ; for the written Páli of 200 years later possesses at last the various n's, though it has but one s. The four vowels, initials, have been discovered H ..,A L : the second seems to be the skeleton of the third, as if denoting the smallest possible vocal sound. Of the medial vowels, it is needless to speak, as their agreement in system with the old Nagari was long since pointed out. The two long vowels í anda are produced by doubling the short symbols.
The visarga is of doubtful occurrence; but the anusvdra is constantly employed; and when before mas D. 8, it is equivalent to the duplication employed in the more modern Pali writing.
