On Sat, 3 Sep 2011 09:39:34 +0600 Christopher Fynn <chris.f...@gmail.com> wrote:
> You can find quite a few "non-standard" stacks (those used in Tibetan > abbreviations) in the book བསྡུ་ཡིག་གསེར་གྱི་ཨ་ལོང། which is freely > available in PDF format from > <http://www.dzongkha.gov.bt/publications/PDF-publications/Duyig.pdf> The visible form of everything I've seen up to p55 (paper document numbering) fits the *visible* form of consonant stacks with marks above and below. (There are some PUA stacks which might have failed to render properly.) Most of the stacks in the abbreviations that could be interpreted as CVCV have U+0F39 TIBETAN MARK TSA -PHRU as the second consonant. However, it is the only Tibetan character of its combining class, so its position is immaterial in the simple examples. It functions as an abbreviated form of TSA to DZA and ZA. I should revisit the examples with the Tibetan anusvara in the contractions, as it has combining class 0. It can abbreviate syllable initial MA as well as final nasals. I did find one 'logical' CVCV form, on p55, an abbreviation of རྩེ་དྲུག་ _rtse drug_ (3 stacks) that looks like རྩྲེུག་ _rtsreug_ (2 stacks). I'm not sure that abbreviations have logical orders - I note that both the full form to abbreviated form tables and the abbreviated to full form tables are sorted alphabetically according to the full form. There are a lot of 'logical' CVC stacks. For example, the very first itme is the abbreviation of ཀི་ཀང་ _ki kang_ (3 stacks) as what looks like ཀྐིང་ _kking_ (2 stacks). However, there are a great many dislocations in the abbreviations, such as a vowel being transferred to a following stack, defeating any attempt to store the abbreviation in logical order. I've been eyeballing the text, so I may well have overlooked relevant examples. Richard.