--- In
cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
> On Wed, 11 Aug 2004 10:39:49 +0000, tgpedersen
> <tgpedersen@...> wrote:
>
> >--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...>
wrote:
> >> On Wed, 11 Aug 2004 09:39:03 +0000, tgpedersen
> >> <tgpedersen@...> wrote:
> >>
> >> >--- In cybalist@yahoogroups.com, "petegray" <petegray@...>
wrote:
> >> >> > Odd thing is, medio-passives in other IE languages
(Italic,
> >> >Celtic,
> >> >> > Tocharian, Hittite) use the same *(t)er as a suffix in
medio-
> >> >passive
> >> >> > standing for impersonal statements.
> >> >>
> >> >> Latin uses -tur < *-to-r, not -ter.
> >> >>
> >> >
> >> >I should have written *-te/or. Tocharian has -tär and -tar.
> >>
> >> Both from *-tor. (*-ter would have given *-cär).
> >>
> >
> >Hm. At least Dutch 'daar', Engl. 'there' shows it must have been
> >*te/or.
>
> 1. Dutch 'daar', Engl. 'there' are not 3sg. middle endings.
My point is: they might be.
> 2. OE þa:r/þæ:r has a long vowel (*te:r).
>
Ablaut.
I was about to guess that the inchoative verbal suffix *-nu-/*-new-
(only in present stems) was identical to the sentence connective *nu-
when I saw that Curtius had proposed that back in the 19th century.
In other words, we get <verbal root><sentence connective><person and
number suffix>.
Now the question is: how the heck did the sentence connective *nu-
manage to get itself wedged in between the verbal stem and the
personal endings? Logically, there are two possibilities:
1.
<root> + <connective> + <ending> ->
<root><connective> + <ending> ->
<root><connective><ending>
and
2.
<root> + <connective> + <ending> ->
<root> + <connective><ending> ->
<root><connective><ending>
The advantage of 2. is that the <connective><ending> merger could
take place at the beginning of the sentence, as it does in Hittite.
The result of that might end up behind the verbal stem as a result
of fronting of the verb.
One more thing: if this is the way for the sentence connective *nu-
to get itself wedged insidde the verb, other sentence connectives
could do the same, eg. *sV-. Sanskrit has, besides the *sa-/ta-
demonstrative (the *s- and *t- of which are actually sentence
connectives) also sya-/tya- (and might have had *sva-/*tva-,
according to Burrow). Now if the y- (presumably that of *yos?) is
also a "detachable" sentence connective (no one has explained the y-
of 'yon' and German 'jener') we might have two sentence connectives,
*s- and *sy-, which, if they can follow the course of *nu- into the
verb, will be an alternative to explaining *-s-/*-sy- suffixes by
the verbal root *es- "to be". Since the semantics of *s- is temporal
sequence (I'm reminded of a narrative style in Danish we were told
not to use in school: 'og så ... og så ... og så' "and then ... and
then ... and then" (in the present), reminiscent of the Hittite
inane 'nu ... nu ... nu') it would match a use as future or
subjunctive.
(Actually, although cumbersome, you might use 'og da ... og da ...
og da' in Danish, meaning "and then ... and then ... and then" (in
the past). Perhaps all the deixis, all the temporal and local
reference were once restricted to the sentence (sequential)
connectives)?
Torsten