Re: [tied] Daughter (was: Short and long vowels)

From: Patrick Ryan
Message: 39411
Date: 2005-07-24

----- Original Message -----
From: "Richard Wordingham" <richard.wordingham@...>
To: <cybalist@yahoogroups.com>
Sent: Sunday, July 24, 2005 3:51 PM
Subject: Re: [tied] Daughter (was: Short and long vowels)


--- In cybalist@yahoogroups.com, "Patrick Ryan" <proto-language@...>
wrote:

> I am not sure I understand why Greek _thugáte:r_ would argue for PIE
> *dhúgh2ter-.
>
> I would think it would argue for PIE *dhug(h)-Há(H)tR-; I believe
the Greek
> form correctly reflects the original PIE position of the stress-accent.

a) The second vowel is reduced / zero-grade (as readily demonstrated
by the Greek-Sanskrit correspondence a~i), and therefore unlikely to
have borne the accent, though that is not beyond the bounds of
possibility. (Or is it impossible for a laryngeal?)

b) In Greek, all morae between the final and the accented mora must be
in the same syllable. If other rules would put it in such a banned
position, it moves towards the end of the word until its position is
complies with the rule. Thus PIE *dHúgh2te:r would yield Greek
_thugáte:r_. As another application of the rule, this stress is the
nearest the nominative singular can come to having the stress of the
vocative, _thúgater_.

Richard.

***
Patrick:

Regarding a): At a stage later than *dhug(h)-Há(H)tR-, the pre-Greek PIE
would have been *dhug(h)á:tR-.

The pre-I-I misanalysis to which I referred would have tranformed
*dhug(h)-Há(H)tR- into *dhug(h)H(H)-tér, with understandable subsequent
developments.

Regarding b): PIE *dhúg(h)Ha(H)te:r is impossible because:

1) Until the final stage of PIE, when the stress-accent was moved to the
root-syllable, the very form of the syllable would have excluded the
stress-accent: *dhu- is a zero-grade of *dheu-, suggesting that the
stress-accent was farther to the right;

2) -te:r- is impossible as a final syllable for *dhúg(h)Hte:r; the *e
indicates the stress-accent was on it. IF the first syllable had the
stress-accent, the form would have had to have been *dhúg(h)HtR-, or
*dhúg(h)HtR:- from *dhúg(h)HtRs.