[tied] Re: Syncope

--- In cybalist@yahoogroups.com, enlil@... wrote:

> If the word is postSyncope, *wertmn could be based on
> another word with the same pattern that DID survive
> Syncope, a word like *kwo:ns, so I don't know how you're
> saying that I've been proven false.

What features of *k^wó:n can *wértm.n possibly have copied?

> > I get my head blown off if I take the liberty to be more
> > specific than our actual knowledge permits. With "something
> > like *´-mon-s" I mean something so *very* like *´-mon-s that
> > we cannot tell the difference.
>
> Alright. So we agree to an extent: *-m&n-s?

Yes, to the extent this involves a reduced form of the earlier /e/
later to surface with o-timbre, this is what I have been saying for
a very long time.

> > It is VERY unfair that you stoop to misusing this in an
> > effort to make a fool of me.
>
> I'm not interested in making fools of anyone. I'm only
> interesting (obsessed, if you will) in pursuing ideas to
> their extreme limit and questioning to the end. By the way,
> happy fools day. I know that I'll be celebrating <:)

Hey, you actually got me laughing this time.

>
> > You get a much more natural explanation of that if you
> > leave out the unmotivated vowel of the ending. Of course a
> > vowelless ending does not undergo syncopation of anything.
>
> Natural is evidently in the eye of the beholder.
>
> The most optimal etymology for *-s happens to be from a
> demonstrative. I see no other etymology that is as satisfying.

I see it but remain very dissatisfied, for it just is not a way this
language can be seen to operate. Nor can I feel it makes such good
sense as you apparently estimate.

> The genitive-nominative connection is, afaic, too heavily
> dependent on the mere phonetic resemblance of the two suffixes
> and doesn't do anything to explain the animacy contrast
> between *-s and *-d in pronominal stems. It can't just be
> pushed aside with more weak phonetic invokings of ablative
> *-od perhaps. Since *-s/*-d mirrors the contrast of *so and
> *tod in both form AND function, the conclusion is that *-s
> must indeed derive from *-s& (an affixed demonstrative *sa
> > IE *so) just as the inanimate *-d is derived from *-t&.
> The vowel is necessary because its origin demands this.

Yes, it's circular. I must have protested against this a million
times by now: I cannot accept that a stem is used as a case-marking
desinence, and for *to-d even the same stem as the stem it is glued
on to. I understand you will not hear of deriving *so from older *so-
s, so you have a system which was *so, *to-d, that is, pretty
much, "this", but "that-that". Does that really make sense? Why not
*so-s, *to-d, i.e. "this-this", "that-that", which would at least be
consistent? And if we got you that far, why not take the final step
and derive it all from *to-s, *to-d, which would be regular for the
language as we know it? And of course then jettison the presumed
identity of the *-s with the pronominal form *so which offers only a
fortuitous and quite partial similarity and never obeyed any rules
anyway?

> This creates a snowball of further conclusions that explain
> better IE.

That is evidently also in the eye of the beholder.

> For example, since it's safe to say that
> inanimate *-d < *-t&, then the ablative *-od may also derive
> from *-Vt&. In fact, QAR and the accentuation of *-od
> doubly supports this conclusion. We then solve a problem that
> Bomhard briefly laments in "IndoEuropean and the Nostratic
> Hypothesis" concerning the precise relationship, if any, of
> IE *-od and Uralic *-ta. With the above, we see that there
> is a relationship and their phonetic differences can now be
> completely resolved. The snowball after this involves more
> accurate connections between IE and Uralic.

If you call this "completely resolved" you are not demanding much.
They have a dental stop in common, oh yes, and the ablative meaning.
You cannot use this to show that a curtailed *-tV leaves IE *-d, for
there is no evidence that the consonant was a /d/ and not a /t/.

> We now have a transparent and more natural solution than
> what can be arrived at by your point of departure.
>
>
> > Do you have any underlyingly asyllabic endings in your
> > pre-PIE morphology at all?
>
> No. Is that bad? Does Japanese have asyllabic endings?

No, come to think of it, nor does Slavic. But these languages only
have open syllables if I am correctly informed, and that is not the
case with your pre-ablaut IE. I couldn't say it's necessarily fatal,
but it strikes me as strange and unnecessarily complicated.

> > I see nothing unexpected here for I do not fill the
> > preforms full of unobserved vowels.
>
> The vowels are necessary to explain the origin of these
> suffixes properly. If we ignore them, it gets silly.
> The etymologies that I propose are also transparent
> to see.

No, you have not even shown that the minimum form of the presumed
stems *so- and *to- when used as desinences is a full syllable and
not merely the consonant, and, frankly, that was the core of the
issue.

> > I do not think the loss of the nominative marker after
> > long vowel + certain sonants (n,r,y) is irregular.
>
> Well I suppose but it is an irregularity in relation to
> the expectation of having *-s everywhere. Since we don't
> find *-s everywhere, it is irregular in that manner of
> speaking.

Then rules are irregular. Your language gets harder and harder to
use.

> Actually, let's talk about it, because I have yet to
> understand where this loss of *-s is coming from. It only
> disappears after certain voiced sounds but not all.

That's the nature of conditioned sound rules.

> I don't see a natural phonemic class out of {*n, *r, *y}
> other than that they are continuants, but then why
> isn't *w in this? How do you make it any more regular
> with *-z?

I make it more regular in the sense that using *-z, not *-s, I can
avoid a clash with the lack of lengthening before the marker of the
2sg which is also retained everywhere. A voiced *-z would indeed be
more prone to exert a lengthening influence than a voiceless *-s,
and *-z would also be more likely to be absorbed by a preceding
sequence of long vowel + sonant. But in that light I do not quite
like the lengthening effect of *-H2. Still, who am I to dictate the
language what to do? I am not sure what is so special about /w/.
Still, we do have root beginning with *wl-, but not *yl-, *nl- or
*rl-, so maybe there is something natural about it. We also have *ml-
, but I am not sure we have *-s in the nominative of m-stems; Latin
hiems may indicate that we do, but Greek khthó:n, Hittite tekan
speaks against it.

>
> > The thematic class is not restricted to words of a
> > particular phonotactic built, so there is no basis for this.
>
> This is the resulting conclusion based on the rules I so
> far have which I know you don't accept. Hmm. This is
> going to be tough.
>
>
> >> So *-mon- < *-m&n- is an animate variant of *-mn.
> >
> > It does not make sense to say that of a particular
> > allomorph.
>
> Why not?

It's a stage in a process of phonetic reduction, and so not
different from *-mn which shows a later stage of the same reduction.

>
> > All of this was perfectly well acounted for already.
>
> I don't think it was. There's too much mathematics in
> what you say, and not enough linguistics.

Oh, you can say nice things too.

Jens