Latin Stress Development (was: Latin medial voicing of *p,*k & *t

--- In cybalist@yahoogroups.com, "Glen Gordon" <glengordon01@...>
wrote:

> As far as I know, the pre-Latin accent is only suspected to be

initial, not

> entirely proven though. I'd imagine that if the Latin accent

happened to

> be quasi-penultimate as it seemed to be, that the underlying system
> must be a stricter penultimate accent. So I would dare suggest for
> discussion the following alternative possibility:
>
> mobile > penultimate > quasi-penultimate (mixed accent)

I thought that that intermediate stage was only seen in
quadrisyllabic words like Plautus's _facilíus_, intermediate between
older *fácilius (arguably evidenced by the weakening of *facel- or
*facol-, which otherwise lead to Classical faculta:s) and Classical
_facílius_.

Classical Sanskit and Classical Arabic have automatic stress rules
very simlar to that of Latin. Do you think they also had a preceding
stage with penultimate stress?

Richard.