From: fournet.arnaud
Message: 56436
Date: 2008-04-02
----- Original Message -----
From: Anders R. Joergensen
Before we got a bit sidetracked by the whole mak- 'eat' discussion, I
responded to your (Arnaud's) list of alledged instances of *-h2G- >
PCelt. -KK-.
There were some words that required additional clarification and I
also presented some possible counter-examples to the proposed law,
that you might want to address (see below).
Anders
==========
My proposal is : For Celtic and Osco-Umbrian :
-?-k > -kk-
-?-t > -tt-
-?-p > -pp-
It does not work with all H2
only when H2 is H2.1 = *?
Eastern PIE has voiced consonants -g- -d- -b-
when Celtic has geminates.
Arnaud
============
--- In cybalist@yahoogroups.com, "Anders R. Joergensen"
<ollga_loudec@...> wrote:
> What about
> > lakk- "lax, soft"
> > lag-aros
>
> Isn't this OIr. lac `weak, feeble' once again? This had /g/ and was
> thus not relevant.
========
I agree this example is bad
*lagg-
and Latin as la(n)g
Arnaud
==========
> >
> > brecc- "speckled"
>
> Yes, along with W brych/brech etc., this points to PCelt *brikko-.
> However, what is the relevance to the present discussion?
===========
The relevance is Lit mar-g- with -g- when Celtic is -kk-
Cf. the word for badger in Celtic
black and white speckled.
Arnaud
===========
>
> W hawdd `easy, happy', MCorn. hueth `glad, joyful' may also be
> relevant, if it is the `sweet'-word. Then from PCelt. *swa:du- <
> *swah2du-, but the laryngeal isn't completely secured.
>
> OIr. saigid, -saig `searches for' etc. < PCelt. *sag-ye/o- is
usually
> derived from *s&g(^)- from the root *sah2g(^)- (sa:gio:, sokjan,
> etc.).
>
> Anders
>
==========
These examples are off topic
as LAtin has -g- and -d-
They are neither examples nor counter-examples.
Arnaud
==========