Old Burmese Sound Check

[LinguList]

https://github.com/digling/burmish/blob/master/sounds/sounds.md#old-burmese

[nh36]

In Old Burmese y is [j] c564 not [y] v166.

[nh36]

In Old Burmese ṅ is [ŋ] c795

[nh36]

In Old Burmese 17 o v042 5 should not occur. It should either be o1 or o2. I will have to look at the cases in question.

[nh36]

For our current purposes o₂ in Old Burmese can be interpreted as [u] v027. Although this is not an altogether un-problematic identification.

[nh36]

For our current purposes o₁ in Old Burmese can be interpreted as [o] v042, although this is not an altogether unproblematic identification.

[nh36]

Old Burmese nh nʰ c096, mh mʰ c243, and lh lʰ c541 are incorrect identifications. These should be identified as voiceless rather than aspirated. m̥ c235, l̥ c538

[nh36]

Same goes for rh rʰ. this identification is not correct it should be voiceless.

[nh36]

In Old Burmese c refers [ts] c015 and not to [c] c600.

[nh36]

We can treat ḥ as ɦ, I guess. It is a bit tricky, because in Burmese (lets say Rangoon Burmese) it is a character that is used to mark a tone and apparently the Burmese tones have regular correspondences with Lolo-Burmese tones, so they are tones all the way back, but Nishi suggests that in Old Burmese the tones may have been phonation types, and he reconstructs ɦ in (shall we call it) pre-Burmese.

[nh36]

The 'long vowels' are also ways of marking tone, but for the reasons mentioned immediately previously, we can interpret as follows ā > a: ī > i: ū > u:

[LinguList]

I guess I should think about introducing the possibility to add A/B-distinctions to include historical information in the lists. I mean, like the Old Burmese h with dot, this is a distinction we do not want to loose, but we may want to NOT give it any direct value. Without wanting to encourage people to be indistinctive about all their data, I could add either the possibility to tolerate a symbol by CLPA that is data-set-specific, or I could add dymmy-symbols meaning "unspecified", where one has but to choose to which type it belongs. In this case, we could just treat it as a tone for the moment, and say that h with dot is tone capital superscript H, or some other convention for EXPLICIT user-definied things with unspecified values.

In the end, this symbol would be a composite-symbol: one marker, that says, which basic type (consonant, vowel, tone) it is, and after-wards, a user-defined symbol that would be free for annotation. I'll reflect about this and see how feasible it is to get it integrated into the CLPA proof-check-system.