Page:Proposals for a missionary alphabet; submitted to the Alphabetical Conferences held at the residence of Chevalier Bunsen in January 1854 (IA cu31924100210388).pdf/17

 "bon," "Lundi," or the English "to sing." It is true that in most languages the final guttural nasal becomes really a double consonant, i.e. n+g, as in "sing," or n+k, as in "sink;" still, as the pronunciation on this point varies even in different parts of England, it will be necessary to provide a distinct category, and afterwards a distinct sign, for the guttural nasal.

In some languages we meet even with an initial guttural nasal, as in Tibetan "nga-rang," I myself. Whether here the initial sound is really so evanescent as to require a different sign from that which we have as the final letter in "rang," is a question which a native alone could answer. Certain it is that in the Tibetan alphabet itself both are written by the same sign, while Csoma de Körös writes the initial guttural n by ň, the final by ng; as "ňa-rang."

We have now, on physiological grounds, established the following system of consonants:

According to Sanskrit grammarians, if we begin to pronounce the tenuis, but, in place of stopping it abruptly, allow it to come out with what they call the corresponding "wind" (flatus, wrongly called sibilans), we produce the aspirata, as a modified tenuis, not as a double consonant. This is admissible for the tenuis aspirata, but not for the media aspirata. Other grammarians, therefore, maintain that all mediæ aspiratæ are formed by pronouncing the mediæ with a final h, the flatus lenis being considered identical with the spiritus; and they insist on this principally because the aspirated sonants could not be said to merge into, or terminate by, a surd sibilant. Accepting this view of the formation of these aspirates, to which we have no corresponding sounds in English, we may now represent the