On 05-03-07 08:20, Piotr Gasiorowski wrote:
> On 05-03-04 17:22, João Simões Lopes Filho wrote:
>
>>But *usro- > *ustra- in Germanic.
>
> Not with Verner's Law.
At any rate the EIEC article suggests that -sr- > -zr- > -:r- (with
compensatory lengthening) is a possible development. I haven't given
much thought to it before, but it's an interesting problem, since if
Verner's Law applies to *-sr-, this means that the insertion of -t- is
_very_ young in Germanic (younger than VL); otherwise, any *-sr- would
have ended up as *-str- whatever the stress pattern. What a shame the
sequence is so rare! Can anyone propose any examples or counterexamples?
The only comparable case I can think of off hand is *wesró-m > *wezraN >
*we:ra > ON vár 'spring'.
Piotr