________________
298
THE ALPHABET
Giuseppe Furlani, Religione dei Yezidi, 1930, and "RIVISTA DEGLI STUDI ORIENTALI", 1932, pp. 97-132, and so forth).
However, there is no doubt that the curious, probably cryptic, script of the Yezidis exists or existed. Fig. 139, 1 shows its alphabet, and Fig. 139, 2-3, reproduce two specimens. The script is partly based on the Persian-Arabic writing, and partly on the Latin alphabet, but the majority of signs do not resemble either Arabic or Roman letters. The date of origin of this script is uncertain.
According to Professor Furlani, not only was the Persian-Arabic alphabet the prototype of the new script, but the texts must have first been written in the Persian-Arabic alphabet, and then transcribed in the new cryptic script. It is quite possible; the phonetic values of all the letters of the new script are identical with those of the Persian-Arabic letters. On the other hand, the shapes of the signs of the new character are quite different, with the exception of a few letters, such as ', h, and w. The great majority of the letters seem to be arbitrary inventions, based mainly on geometric elements, such as straight strokes, little squares, triangles and circles, angles, and so forth, and some are similar to Latin letters having geometric forms (1, V, T, L, etc.), but they have quite different phonetic values.
II-Balti Alphabet
There are some manuscripts extant, which are couched in the Bhotia of Baltistan, or Balti, a Tibetan dialect spoken by about 150,000 people, in the province of Baltistan, formerly an independent state, and now part of the ex-State of Kashmir. These manuscripts are written in a script which according to Sir George A. Grierson (Linguistic Survey of India, III-i, pag. 33 sq.), was perhaps invented at the time of the conversion of the Baltis to Islam, about A.D. 1400.
Three kinds of signs can be noted (Fig. 139, 4):
(1) Some signs have the shape of Latin capital letters, but the phonetic values are not the same; in some cases there is a likeness (K representing g, P representing b, and R representing ), in other cases, there is no similarity at all (a reversed B, that is 9, represents an r, an E represents an #). I think that we may conclude that the inventor of the Balti scripts knew the Latin alphabet, and with purpose avoided giving to its signs the same phonetic value.
(2) Some signs (k, kh, ts, ng, th, etc.) have purely geometric forms, such as combinations of little squares and straight strokes.
(3) A few signs represent various sounds which are distinguished by the addition of diacritical marks; for instance, the sign b with a point below it, represents a p, with a point to its left, represents a t; an s with a point to its left, represents the sound sh, and so forth.
(4) The vocalization is rather peculiar; it is not unlike the Indian and
