Re: [tied] Re: Italic root *ored(h)- and ready; short and long vowe

From: Patrick Ryan
Message: 39220
Date: 2005-07-14

 
----- Original Message -----
Sent: Wednesday, July 13, 2005 12:01 PM
Subject: [tied] Re: Italic root *ored(h)- and ready

<snip>

It now so happens that a synchronic root *reH1dh- may simply
have "preaspiration", i.e. aspiration of a stop induced by a preceding
H1 or H2. This is a rule discovered by my wife Birgit Olsen and
presented a number of times already, apparently with little echo among
those in a position to shape general opinions. There is a short
presentation in the Zürich Fachtagung (1992, publ. 1994), and there is
also the first presentation in her study on The PIE Instrument-Noun
suffix *-tlom and its Variants" from the Danish Videnskabernes Selskab
in 1988, and now the utilization for the clarification of the Latin
suffix -idus in Historische Sprachforschung 116 from 2003. I may be
under suspicion of being biased, and I am certainly not so biased
against it as the belligerant phalanx that wants to monopolize
opinions, but let me just report the theory: In a prestage of the IE
protolanguage, the sequences *-H1-T- and *-H2-T- (T being a plosive)
were fused to aspirates *-Th-. Practically all examples involve H + t
yielding IE *th, a phoneme that constitutes a red rag on the
opposition. It is not quite clear to what extent the laryngeal was
lost in the process; there are clear examples both of the structure *-
eH1t- > *-eH1th- > -e:th- with a laryngeal-induced long vowel and of
the structure *-eH1t- > *-eth- with loss of the laryngeal and
therefore a short vowel in the result. Some of the long vowels are no
doubt due to secondary restoration of the laryngeal, but it is not yet
clear if that can be extended to all cases. But suffice it to say for
our present purpose that *re:dh- can indeed reflect original *reH1d-
with UNaspirated *d.

<snip>
 
***
Patrick:
 
Jens, I hope this will be a helpful thought.
 
I had a similar problem with long and short vowels in my approach to these problems.
 
Without going into detail as to how I think they were lengthened, I hypothesize a group of roots having the form *Ce:C-, *Ca:C, and *Co:C, _without_ 'laryngeals' being involved.
 
The problem I found with this is that where I expected Ca:C, I frequently found *CaC, etc.
 
After some thought, I applied what I think is a principle of language development: economy of effort.
 
Where there was no homonymous *CaC, *Ca:C could be shortened to *CaC with no loss of root integrity; and frequently was so shortened; when there was such a root, resistance to shortening the vowel was muc stronger.