Re: [tied] Latin medial voicing of *p,*k & *t before penultimate?

On Mon, 17 Nov 2003 22:09:57 +0000, Glen Gordon <glengordon01@...>
wrote:

>Piotr said a few posts ago:
>>*wik^- > *w(i)g^- > *wdz- > *dzw- > *dz^- > z-
>>
>>This is strangely reminiscent of Latin vi:ginti:, with /g/ from *k^.
>
>I started thinking about that and when I thought about all I
>knew about Latin... not much :)... it seems almost like medial stops
>were voiced before a strict penultimate accent in a pre-Latin stage.
>Am I wrong?

Yes.

>The word /quadra:ginta:/ "forty" has voiced both former *t and
>*k to /d/ and /g/. Both stops occur before a penultimate accent,

Same in quadra/quadrus, quadri- etc. This is a special development of *tr.
I've seen it attributed to the schwa secundum before it, but also to the
/w/. Accent plays a role (*kWtwrós > quadrus, *kWtwó:r > quattuor), but
I'm afraid I've forgotten the details.

>Maybe it would also explain the particle /ab/ "away", /ob/
>"toward" and /sub/ "under", accentless themselves with voicing
>due to original *p being before the accent of the following word.

Final stops do get voiced in Latin.

>Is this an actual rule in pre-Latin? If not, why not?

It's unclear to me whether you're talking about the Latin accent
(determined by syllable weight), the pre-Latin accent (initial) or the PIE
accent. Whatever the case, just open a Latin dictionary and let the
counterexamples leap up at you.


=======================
Miguel Carrasquer Vidal
mcv@...