Re: [tied] Re: The role of analogy, alliteration and sandhi in coun

Richard Wordingham wrote:

> The Gothic is <fidwor>.

yeap.

>> Kymrish : petru- - kW > kp > p
>
> Do you mean Welsh _pedwar_ ?

yes from an *petru; at last so my dictionary.

>> Avestan : c^ature - kW > kp > k
>> Greek : tetra - kW > kp >k > t
> What's your evidence for an intermediate step /k/ in Greek? Greek
> does not have a merger of /k/ and /t/.

I know about a change of ke > kind of "t" (in Albanian), so this change is
possible. Any evidence of "p" > "t"?

> What do you mean by /kp/? I don't see any need for there to have
> been a true labiovelar stop, if that is what you mean.

I mean the kluster which resulted from "kW" due consonification of "W". This
cluster get simplified to "p" or "k".

>
> None of your examples above relate to kW > w ! As an example of an
> intermediate step, we have PIE *kWod > Germanic *xWat > OE _hwat_ >
> Early Modern English /hwat/ > /hwOt/ > /wOt/ 'what'.
>
> Richard.

because there is no "kW " > "w" !!! . Once "W" get consonificated to v,b,p,f
how would you expect to have again "w"?
Your example of Germanic is not properly here since apparently all languages
presents some trouble regarding the interogative pronoun and the labiovelar
question? Any better example?

Alex