Re: [tied] Re: *Twah-

----- Original Message -----
From: tgpedersen
To: cybalist@yahoogroups.com
Sent: Wednesday, September 29, 2004 4:37 PM
Subject: [tied] Re: *Twah-
>
>> >I just discovered South German 'zwagen'. This contrasts with
>> >OHG 'dwahan', 'twahan'. I checked with other roots in *þw-, and it
>> >seems to be a pattern, PGrm. *þw- > OHG dw-, German zw-, rather
>> >unexpected. Perhaps one should look at a German connection after
>all?
>> >
>> >Torsten
>>
>> Hm, yes, perhaps, it isn't impossible, but German zw- always > Cz
>cv- and
>> never c^v-, and as for G -g- > Cz -x-, I must find some more cases
>of this
>> change (if it had been adopted before the g > h change in Cz-Sk, it
>would
>> have looked like cvah-, not cvax- and certainly not c^vax-, if
>followed by a
>> vowel. However, if this -h- is followed by a consonant or if it is
>> word-final, it gets devoiced and does change to -x-).
>
>*þwahan is an irregular verb (at least in ON) and therefore should
>vary in accordandance with Verner. One would expect different
>outcomes in the process of regularisaton (cf the h ~ g of
>OHG 'dwahan' ~ South German 'zwagen').
>BTW which written Czech letter corresponds to what you write as -x- ?

In Czech, /x/ can be written as "ch" (then, it is an independent phoneme) or
as "h" (then, it is an allophone of /h/ before most consonants and at the
end of a word, where all voiced consonants become unvoiced in Cz; /h/ is
special
as it not only gets unvoiced, but even its articulation moves to the velum)

>>We would thus have to
>> expect, that it was *zwag or *zwag-t- (does anything similar occur
>in
>> German?)
>
>I found South German 'zwagen' in the list of cognates for 'to' (vb.)
>in my Dansk Etymologisk Ordbog. The entry for 'tvætte' does not
>contain a German cognate, but I can't exclude that it might exist.
>
>
>that once might have been borrowed from German. Yet still, there is
>> c^v- in Czech, which is so confusing! :-)
>>
>Absolutely! The question is whether the þw- on its way to zw- passed
>a stage that would be similar to c^v-. But all the loans have cv- as
>you pointed out.
>The idea that the sense "wash" of c^va:xat comes from ca:kat comes
>from where? What is the evidence that it wasn't original in c^va:xat?

There is no evidence, unfortunatelly, no proof of anything, I'm afraid. I've
found nothing. Oh, I should have written that most Czech etymological
dictionaries claim that all these words are expressive onomat. without
linking them to possible cognates in other languages. I wanted to say that,
as ca:kat and (the assumed precursor of) c^vaxat are related semantically
(they are both connected to water, bathing etc.), they might have got
contaminated. But I am getting to abandon this idea, since it seems too
constrained to me in the end.

>> Another possibility is a contamination of the two words: one being a
>> German(ic) loan, the other being a genuine Cz onomat. verb. That is
>*twah-
>> (> *xxxx- ) > cvax- plus Cz c^a:kat "to splash". I don't know, both
>zwagen
>> and c^va(:)x(t)at are puzzling and need much more further
>investigation.
>>
>
>Perhaps I should also point out that there is a development þw- >
>Germanic kW- (English 'twig', Danish 'kvist'; English '(a)thwart',
>German 'quer', German 'Zwerchfell', Swiss German 'Dwarr-(brücke)'
>(bridge on the old Srt. Gotthard route)).
>
>Torsten

Ok, I'll do some research and we'll see if I find something :-) I wish I had
an electronic version of my Czech etymological dictionaries :-(.