--- In
cybalist@yahoogroups.com, enlil@... wrote:
> Me:
> > I mean *t-ex-, not *t-e-x-, where *-ex is the feminine ending,
> > sorry. That makes a teensy bit of difference.
> >
> > To be clear, the feminine ending when it was created was not a
> > feminine ending; it was a human collective ending.
>
> Jens, warping the context of the above quote:
> > That is a theory, perhaps a true one, but still only a theory as
> > long as it has not been proved. It is not solved by a simple
decree.
I am not addressing the context, only the side-issue of the relation
between collective and feminine. If that distorts any interesting
context that had already been done before I quoted it.
> Yes, it is a theory, like yours which you "supported" by claiming
> that the mere existence of *tex- somehow proves that your "decree"
> is the correct one. Well, my theory shows that this is untrue.
No theory "shows" anything. A theory can be false.
> Since my above "decree" shows that we can produce *tex- long AFTER
> the original process that created e/o alternation in thematic
> vowels, simply by affixing *-ex to a stem perceived to be *t- on
the
> basis of other case forms, the entire basis of your theory is
> in reality non-existent.
A decree certainly does not show much. I continue to see a problem
in this, but I am not sure it is decisive. Still, it is embarrassing
if an author of a grand theory does not have any idea how his theory
worked in specific terms.
The problem is of course that the theory presupposes spread of
a "feminine marker" either *-H2- or *-eH2-. Now we are told
that "other case forms" supply the basis of it. But what case forms
could possibly have a feminine marker if there was no feminine
category? And if the category was in reality a collective, why was
the collective *teH2 not used as a feminine, which instead took the
funny form *seH2? Was there an event of replacement of a feminine
*teH2 in non-Anatolian IE, or was *seH2 already formed before
Anatolian broke off, and then as a collective?
> Simple: *tex- does NOT conclusively show that we must accept that
> IE had an original three-way gender system. The absence of
> feminine gender in Anatolian proves the opposite and it is
> this two-gender system that happens to be the consensus
> for IE.
No, the absence of a feminine category in Anatolian proves no more
then the absence of genders in English. Even considering the
relative early attestation there is ample time for it to disappear.
The question will have to be decided by the particulars. It is a
problem to the theory of the lateness of the feminine category that
the feminine marker added to consonant stems //-yeH2-// shows
ablaut. Why did it do that if it was not formed till way after the
period when the phonetic changes that caused the ablaut? And if it
was formed already while ablaut was operative or before it arose,
and then must have been a collective marker (if it is not to destroy
the theory), why does it never mark the collective, but always the
feminine? These obstacles vanish if an IE feminine category has
disappeared in Anatolian.
> So unless you have other "evidence" to support your claim, I would
> suggest discontinuing your own decrees which have no basis.
>
>
> > There is no underlying vowel in the colletive/feminine morpheme
*-
> > H2. The form is a strong case, and there is no full-grade
variant.
> >
> > Therefore, *táH2 must be analysed as *té-H2.
>
> A new decree?
No, the language tells us so. That ought to be respected. All
morphemes that have an interplæay of full-grade and zero-grade
alternants shift the accent to the following syllable. That is not
done by the collective marker, the nom.-acc.pl. neuter being a
strong case. That is proof there was no vowel in it. Therefore a
segmentation involving suffixal *-eH2 is incorrect, and */teH2/ must
be analyzed as stem /te-/ + ending /-H2/, as I suppose almost
everybody would anyway, irrespective of their views on the mechanism
of the ablaut.
> Say wha? If we don't have *-ex, where are later feminines showing
> *-a: in various IE branches coming from?? Rather, IE shows both
> *-x and *-ex, just as we see both *-r and *-or-, *-n and *-on-,
etc.
"[L]ater feminines showing *-a: in various IE branches" come from
the thematic declension. Fem. *néw-aH2 is stem *new-e- + fem./coll.
*-H2. Collectives of consonant stems just end in *-C-H2. Oddly,
feminines of consonant stems end in *-C-iH2/*-C-yeH2-, but that does
not permit an analysis that segments off the *-eH2- of the strong
variant of the latter, assigning it the status of a feminine marker
in its own right which I do not see it can be shown to be.
> If you claim that *-ex somehow doesn't exist, you employ yourself
> full-time in a futile effort to erase all instances of this well
> documented feminine ending. For what purpose other than to come
> up with some crazy counter to an inevitable conclusion?
Some stems actually do have a suffixal segment //-eH2-//, and to
cause confusion the lexeme //gWén-eH2-// 'woman' is among them. But
so are *még^-eH2- 'big' (not used in the fem. which adds //-yeH2-//,
as Ved. mahí:-, and quite probably Hitt. mekki-s which I would take
to be an old feminine stem, but that is no more than a personal
point of view), and *pónt-eH2- 'path' which is masculine. I know no
derived feminines that use a morpheme //-eH2-// to form the
feminine. The purpose is to account for the facts of the language in
an economic and consistent way.
Jens