Re: a discussion on OIT

--- In cybalist@yahoogroups.com, "Andrew Jarrette" <anjarrette@...> wrote:

> Is it correct that the glottalic theory posits the series
> *p-*t-*k/*p'-*t'-*k'/*b-*d-*g (ignoring for the sake of argument the
> palatals and the labiovelars), with *b-*d-*g representing the
> *bh-*dh-*gh of traditional reconstruction?

> If this is so, then the
> glottalic theory says that voiceless aspirates in Greek and
> proto-Italic come from *b-*d-*g while *p-*t-*k remain unaspirated.
> This seems incredibly implausible to me, that voiced stops should
> become voiceless aspirates while voiceless stops remain unchanged.

It happened in a contiguous group of Tai dialects (most notably
Siamese and Lao) and is also known from other parts of the world. The
shifts there from Proto-Tai to the modern language are exemplified by:

t > t
th > th
?d > d
d > th

/?d/ is a preglottalised stop. In terms of its effect on tone splits,
it most typically has the effect of a voiceless consonant, though
sometimes the glottalised stop and preglottalised consonants have an
intermediate effect.

Loans from Indic appear from the spelling to have undergone the d > th
shift, e.g. Sanskrit _dvija_ 'twice-born' > Thai thawit 'bird, brahmin'.

Richard.