Re: [tied] o/e or reduplication

On Wed, 26 May 2004 21:18:52 +0000, elmeras2000
<jer@...> wrote:

>--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>> On Wed, 26 May 2004 02:23:33 +0200 (MET DST), Jens Elmegaard
>> Rasmussen <jer@...> wrote:
>>
>>
>> >> I simply prefer *wednós, *wedéni, because of the complete
>> >> parallel with Vedic udnás, udén(i).
>> >
>> >This will be reading an alternation into the graphics of a
>paradigm that
>> >do no show that. What is the basis of the assumption that the
>presumed
>> >cluster of /wedn-/ would be written with -te- and -ti-?
>>
>> Standard Hittite orthographic practice. The cluster is
>> written using signs that echo the preceding vowel (/linkta/
>> = li-ni-kta, /walhzi/ = wa-al-ah-zi). How else could
>> /wednas/ be written?
>
>It would be normal for it to be written *ú-(e-)id-na-(a-)as. It
>would be most unusual if an intervocalic two-consonant cluster was
>written in a way that could indicate a vowel between the consonants.
>That appears to be avoided as a general rule.

You're right. I don't know what I was thinking (too much
Linear B).

>Therefore the five
>spellings ú-wi5-te-na-as, ú-e-te-na, ú-e-ti-na-an-za, ú-i-te-ni-it,
>ú-i-te-na-as given by Rieken for case forms other than the loc.sg.
>can hardly have /dn/, but would seem to demand /-den-/ just like the
>loc.sg. ú-i-te-e-ni, ú-e-te-ni.

There is another plene spelling, given by Melchert as
we:dand(a), i.e. *wédn.d or *wédend, with accent on the
initial vowel.

It looks as if there was a near complete merger in Hittite
of the two paradigms *wódr., *wédn- and *udó:r, *udén-, with
the initial-stressed form taking the *-e- from the
end-stressed form (at least in the oblique: *wódr, *wéden-),
and the end-stressed form taking the *we- from the
initial-stressed form (*wedó:r, *wedén-).

So that gives:

NA *wódr (wá:tar)
G *wédenos ~ *wedénos (wetenas/witenas)
DL *wédeni ~ *wedéni (weteni/wité:ni)
All *wédeno ~ *wedéno (wetena)
IAb *wéden(o)t ~ *wedén(e)t (wé:dand(a)/wetenaz/witenit)
NA *wedó:r (witá:r)


=======================
Miguel Carrasquer Vidal
mcv@...