> My point of departure is looking at a table which e.g. , lists
> p (plain) p' (glottalized) ph (aspirated) phonetically but writes
> them *p *b *bh, respectively when setting up correspondences. ... I
> guess some dispute the glottalized value
The system which we seem to have to reconstruct is imbalanced, and almost
unknown elsewhere. People have tried to make sense of it by suggesting
that what we reconstruct as *b was phonetically not /b/ but something else.
Various common 3-way systems have been tried. The only ones with any real
hope of success were the "glottalic" theories (there was/is more than one!).
They also seemed for a brief moment to offer explanations for other
phenomena, such as Lachmann's Law in Latin. More detailed discussion of
them has made some linguists a little more doubtful now - it may have been a
false dawn! (Although there are many who still prefer this explanation).
Alternatively, if there was an imbalanced system in PIE, we have a ready
explanation for the various changes that clearly took place: an imbalanced
system was adjusted to make it more stable. So the consonant shifts in
Germanic and Armenian, the collapsing of aspirated and unaspirated voiced
consonants in Hittite, Avestan, Slavic, & Baltic, and the development of
voiceless aspirates in Sanskrit all find explanations. The only problem
then is to find out how the system became imbalanced in the first place!
Peter