Re: [tied] Water: -g, -d, -r

--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:

> On Tue, 29 Jun 2004 12:31:48 +0000, tgpedersen
> <tgpedersen@...> wrote:
>
> >
> >Basque <ur> "water" has the combining forms <ur->, <uh->, <ug->.
>
> And why not ub-?

Because I was quoting Löpelmann ;-)

> In fact, the only combining forms are u- (old) and ur-
> (recent), with regular loss of final -r (not -rr!) in the
> old compunds.
>
> If a vowel follows, the hiatus in the sequence u-V- is
> resolved as *uwV- > ubV-, or uhV- [> ugV- in h-less
> dialects], depending on the dialect and perhaps the
> following vowel (e.g. *ur + -alte => *u-alte "flood" =>
> uhalte (Z), uhalde (L), ubalde (AN), ugalde (B,G)).
>

Hm! Do you have other examples of this general (I assume?) hiatus
rule producing similarly dissimilar combining forms?

The rule in itself looks plausible enough, you've proposed something
similar for Slavic adjectives. Here are some exampls from Swedish:

frugan "the missus" cf. Danish fruen (no -g-, and none in Germanic in
general), and Swedish/Danish pairs such as båge/bue "bow",
allmoge/almue "public, peasantry". Apparently as -g- vs. -G- became a
Swedish/Danish shibboleth pair, all "suspected" -G-'s were replaced
in Swedish with -g-, including some which were just hiatus'es
cf also
Sw. stuga vs. Da. stue "cot/living room", cognate with other Germanic
words with -w-, German Stube, English stove.

> According to my rules, the pre-pre-Basque form would have
> been *(w)ud-.

With early loss of -d- in the combining form?

There's another one for your list of Basque-Nostratic cognates (or
loans, or whatever).

Torsten