Re: voiced aspirates [was: Ligurian]

--- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy
<bhrihstlobhrouzghdhroy@...> wrote:

>
> > However, I disagree with the traditional reconstruction of series

III as "voiced aspirated", which only exist in Indic. To me they were
plain voiced *b,*d,*g, which is their outcome in most IE languages.
However, in Greek and Italic they shifted to voiceless aspirated
*pH,*tH,*kH which later became voiceless fricatives. As this shift also
happened in Etruscan, we can explain it as a result of language contact.

> >
> > Not to mention that "voiced aspirated" are specifically Indic and

hence

> > not reconstructable for PIE. IMHO, series III is plain voiced.
>
> Bhrihskwobhloukstroy:
>
> You surely know that they occur in their etymological position
> (and as sound-substitution for acrolectal WArm. /p/) in Sevas and
> Transylvanian Armenian; possibly in Myc. <pu2> as well.
>

No, I didn't know.

> Anyway, Your typological majority criterion is at variance with
> the classical procedure of selecting marked features and
> back-projecting them (according to the theoretical framework of an
> optimalist direction of diachronic sound change):
> b = + plosive, + labial, + voiced, - aspirate
> f (Latin) = - plosive, + 'labial', - voiced, - aspirate
> ph (Greek, Romani) = + plosive, + labial, - voiced, + aspirate
> p (Tocharian, partly Anatolian) = + plosive, + labial, - voiced, -

aspirate

> bh = + plosive, + labial, + 'voiced', + aspirate (properly murmured
> and aspirate or breathy and voiced)
>
> result: PIE = + plosive, + labial, + 'voiced', + aspirate = /bh/
>

What you call "classical procedure" was actually Neogrammarians' choice.
However, external comparison (including IE loanwords in Kartvelian)
don't support them.