--- In qalam@yahoogroups.com, "Peter T. Daniels" <grammatim@...> wrote:

> Richard Wordingham wrote:

> > I criticised the Unicode description of the effects of Thai consonants
> > on tone, and got told I should propose an alternative. This behaviour
> > is part of a family of effects seen in Thai, Lao and Khmer in the
> > Khmer group of encoded scripts, and in New Tai Lue but *not* Burmese
> > in the Mon group. It is also present in some unencoded scripts -
> > Lanna, Viet Thai, Mon (mostly encoded) and Cham - though I had to dig
> > hard to find evidence for the latter, e.g. on p119 of
> >

<http://www.people.cornell.edu/pages/mb236/Eastern%20Cham%20register.pdf>

> A first guess would be Gérard Diffloth, but Cornell does its email by
> initials, so this would seem to be by M.B., and I don't know who that
> might be. (Definitely not Martin Bernal.)

MB = Marc Brunelle. The article is his Ph.D. thesis.

> There have been a handful of studies of Southeast Asia as a Linguistic
> Area, which is where you'd look for unified treatments of the phenomena
> you mention.

The main phenomenon, the shift of the contrast in the voicing of the
onset to a contrast in the rhyme, also extends to most of Chinese (how
exceptional is Wu?), but the timing is uneven. For example, it
affects some dialects or languages known as Yi, but not the one on
which the Yi syllabary is based, and only some dialects of Hlai (the
main Tai-Kadai language of Hainan).

The splits can also be quite odd - I was surprised to see that Cham
resonants pattern as voiceless, whereas in Tai the plain resonants
pattern as voiced, as in Khmer. The tonal languages where voiced
stops give voiceless aspirates can have 3-way splits with some bizarre
effects, as in Southern Thai and some of the Lao dialects.

Richard.