Mongolian Wa and the Seam Between Inherited and Added Letters

The thirty-five letters of the traditional Mongolian alphabet are not one thing. The first twenty-five are the inherited core, taken from Old Uyghur when the script was adopted and sufficient to write the Mongolian language. The last ten were supplied for sounds Mongolian does not have, met in words taken from Chinese, Tibetan and Sanskrit. Knowing where the seam falls changes how the alphabet reads. The core is a coherent system: harmony classes, paired consonants, restrictions on what may begin a word, and a set of ambiguities the language itself resolves. The additions are a toolkit, assembled as need arose, with no internal logic beyond the sounds they were required to write.

Counting them together as a single list of thirty-five is correct for reciting the alphabet and misleading as a description of the language. The distinction matters again just beyond the alphabet, where a further set of signs encoded in Unicode under the name Ali Gali serves to write Sanskrit and Tibetan in full, and again beyond that, where the same Unicode block holds the letters of Todo, Sibe and Manchu, three separate writing systems built on the same shapes and not part of this alphabet at all. Any count of the Mongolian letters is therefore a statement about which of those layers is being counted, and a count offered without that statement is not much use.