Re: [tied] Stative Verbs, or Perfect Tense

--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:

> In any case, the Hittite form doesn't stand alone. If we
> compare Hitt. -ir, -er with Latin -e:re, the ending can only
> be reconstructed as *-e:r.

Certainly!

>
> >[JER:]
> >> >In Modern Greek the imperfect and aorist use the endings -a, -
> >es, -e,
> >> >-ame, -ate, -an. In this the 2/3 sg are thematic, while the

rest

> >have the
> >> >-a- of the aorist. In Old Church Slavic the aorist has

thematic

> >endings in
> >> >the 2/3 sg, as tec^e, tec^e 'ran, flowed', while the other
> >persons are
> >> >sigmatic, either 1sg têxU, pl. têxomU, têste, tês^eN, or 1sg
> >tekoxU, pl
> >> >tekoxomU, tekoste, tekos^eN. Here one could say: Where else do

we

> >find
> >> >retention of an unmarked stem only in the 2/3 sg
> >>
> >> You know that's exact opposite of what I was saying.
> >
> >Is it? Wouldn't that be just as hard to ascribe to chance? Is the
> >Greek-Slavic correspondence trivial and the Hittite-Tocharian one

so

> >dramatic that comparative grammar has to be recast in its honour?

I

> >am willing to accept anything if there is a good reason, but I

have

> >trouble seeing it here.
>
> My probem is exactly the same. I fail do see a good reason
> for the proposed Anatolian/Tocharian scenario.

It does not demand that much of a good reason to try and invent a
way to preserve the morphological system pointed to by the IE
branches whose diachronic interpretation is so much easier.

There are several options for that, even before one begins to
disqualify tradition. If Anatolian and Tocharian left the old unity
together there are no problems at all, for then whatever
correspondences one does not want to ascribe to chance can have
occurred in a common period and only once. If one does not want to
consider that or has reasons to reject it, one can find some
consolation in the differences they do present: For Tocharian one
need only assume that the inflection of the s-aorist began to assume
the endings of the perfect, but did not extend that process to the
3sg and the middle voice, which would not be unnatural at all. For
Anatolian one need only assume that the lack of marking of the 3sg
of the perfect made the corresponding form of the s-aorist
preferable in the long run. I cannot consider the superficial
correspondence that the 3sg is then found with an /s/ in the
continuation of the perfect in two languages (in Toch. in the only
continuation, in Anatolian only in one of its tenses) so dramatic
that it demands the suggested recasting of the history-book.

>
> There are no mysteries in the Greek case. Classical Greek
> had a thematic imperfect -on, -es, -e, -omen, -ete, -on, and
> an s-aorist -sa, -sas, -se, -samen, -sate, -san (k-perfect
> -ka, -kas, -ke, -kamen, -kate, -kasi). The -a- comes from a
> syllabic nasal (1sg. *-sm., 3pl. *-sn.t) c.q. from pf. 1sg.
> *-h2a. In Modern Greek, the paradigms have been levelled at
> the desinence level, but an s-aorist remains an s-aorist:
> the -s- is still there (allright, there are now also
> s-imperfects). The case is even clearer in Latin: the old
> s-aorists adopted the desinences of the perfect (vix-i:,
> vix-isti:, vix-it etc.), but they didn't lose the -s-.

I have never regarded the Greek development as particularly
problematic.

> The Slavic case is more complicated: disregarding the
> secondary -ox- aorist, we have a mix of e-imperfect,
> root-aorist, s-aorist and se-aorist: the 1st. sg.du.pl.
> person has *-só- (but long grade root, as in the s-aorist),
> the 2/3 persons have *-s- in the plural and dual, but in the
> singular we have either imperfect (nése-) or root-aorist
> (velê'-). A few athematic verbs have s-aorist in the 2/3sg
> (by(stU), da(stU)). Some roots ending in a consonant have
> the thematic non-sigmatic aorist (< root aorist/imperfect)
> in all persons. Without going into more details, the
> rationale seems to be a general confusion of the different
> types of aorist (with the imperfect thrown in after the
> creation of the characterized impf. in -êaxU), with a choice
> being made in favour of those forms with the most open
> syllables (to the point of creating the double thematic
> -oxU-aorist), which follows from the "law of open
> syllables".

I have never regarded the Slavic development as paricularly
problematic either. It is the completely fortuitous identity of the
Greek and Slavic results that gives cause to reflections.

> For Hittite and Tocharian, I just don't see a rationale. I
> don't really believe Jasanoff's account that the 3sg. of the
> hi-past (and only the 3sg.) was replaced, nor do I believe
> the more traditional versions that see the hi-past as a
> perfect that adopted the 3sg. from the s-aorist. The 3rd.
> person is usually the most resistant to change.

But does the theory demand that the 3sg -s represents change? You
can get it to be *very* change-resistent if you regard it as the
*only* member of an s-aorist paradigm that resisted the invasion of
the perfect endings.

> The version where the hi-past is an s-aorist that gradually
> incorporated forms of the perfect is more credible.

Ah, there it came; thank you.

> Still,
> if -s- was already an aspect marker in PIE, it's strange
> that it would have disappeared, except for phonetic reasons
> (the Latin development, where perfect endings are added on
> top of the -s- is more plausible, other things being equal).

Anatolian has obliterated the present/aorist dichotomy completely.
So has Germanic. I really cannot regard the creation of the s-aorist
as a post-fission innovation, seeing that it has Narten ablaut which
is otherwise only found in archaic ruins, and also because of its
relationship with the sk-presents which most of all looks like
something that cuts back to a time we cannot otherwise reach.

> Tell-tale archaisms (like Slavic dastU, bystU) seem to be
> missing from the Hittite hi-past: the 2sg. (3sg.) ending
> -sta (as if from s-aorist *-s + perfect *-th2a) is
> _Neo-Hittite_ (so it must really be analyzed as mi-past 2sg.
> -s plus hi-past -ta, c.q. 3sg. hi-past -s + 3sg. mi-past
> -t(a)). In Tocharian, if the active and the middle both go
> back to fully sigmatic forms, why was -s- preserved in the
> middle and not in the active, outside the 3sg.?

It is often claimed that the perfect did not use a middle voice on a
regular basis. But even if it did, its forms would be under no
obligation to invade the s-aorist also.

> It's not as
> if the Tocharian class III middles continue the s-aorist
> middle in pristine form: for one thing, they have added
> -a:-.

Yes, after the -s-. That looks like ending conglomerates segmented
off from the root aorist of set.-roots.

Still, the accent of the prt. III middle seems to often reflect
reduplication, so maybe this is not som simple.

Jens