From: Arnaud Fournet
Message: 62043
Date: 2008-12-13
----- Original Message -----
From: "gprosti" <gprosti@...>
>>
>> Is the above Armenian word typically traced to a suffix-less *wed-? (I
>> recall that the IEW traces it to *wedo:(r).) Is there any other
>> possible attestation of a root noun that the heteroclitic "water" stem
>> would have been derived from?
>>
>> ============
>>
>> I don't think there is any problem with PIE *wed-(or/-en) becoming
> LW *wed
>> in Uralic.
>> Most LWs from IE languages into Uralic languages are rearranged to
> fit into
>> Uralic morphology.
>> The alternation -or/-en makes no sense from the Uralic PofV.
>>
>> A.
>>
>
> Why would this word have been taken into Uralic without the suffix,
> rather than with one or another form of this suffix? I don't see how
> the former option is any more probable than the latter.
>
========
Anyway the oblique case form PIE *w(e)d-en- fits into the oblique case form
*wed-en in Uralic.
So we can suppose that the borrowed from *wed-en can have been
retro-analysed by Uralic itself as *wed + en (genitif morpheme).
The analysis being in a way also valid for PIE.
The result within Uralic being a naked *wed "water" for nominative case.
In other words, with -en or without -en, the result in Uralic is the same :
*wed for nominative.
Arnaud