Momentary-Durative

[Jens Elmegaard Rasmussen]

On Thu, 17 Jun 1999, petegray wrote [in discussion with Vidhyanath Rao]:

[]
> My reading of the literature is that there is a fairly wide general
> agreement, that at first a tenseless verb was used with endings -m, -s, -t
> and so on.

Pardon the intrusion, but what does "at first" mean here? That at one time
the verbal system was as small as that? You still have tenseless -m, -s,
-t in Vedic and the Gathas, but embedded in a system where tense can be
specified whenever needed. If the tense markers such as the primary *-i
and the augment *e- were once independent words, how can we know they are
younger than the person markers?

> The precise formation of it may be more argued, so thematic -
> or especially sigmatic - aorists may be later than athematic and asigmatic.

Thematic apparently yes, for they are plainly secondary and presuppose the
existence of an athematic type from which they can be reformations. But
there is no indication that the sigmatic aorist is an innovation; it must
once have had a special function, and I suppose it was inchoative (the
aorist corresponding to the sk^-presents). It became immensely productive,
sure, but that does not mean that _all_ of its examples are late. The
s-aor./sk^-prs. complex is simply a derived verb on a par with the
causative or the desiderative.

> Then the endings -mi, -si, -ti and so on develop, as part of the need to
> mark present action, as a tense-based system gets going.

Why could that need not have been there from the start?

> Various devices
> were also used to mark the stems with continuity or incohativity or various
> other things, all being non-completed.   One of those devices was accent on
> the stem, which produce full grade.

The accent rather _saves_ full grade, but perhaps that is what you mean.

> Hence the appearance in Sanksrit of
> root accented presents with full grade beside zero grade presents with
> accented thematic vowel.

The first part is true, but zero-grade thematic verbs have the same
history as the thematic aorists. They are based on weak forms of an
athematic paradigm (root prs. or root aor.) that lent themselves to
reanalysis. Thus, the 3sg middle *wid-e' "found for himself" looked just
as little like a third person as the 3sg middle prs. in *-o-r (whence, by
subtraction, non-prs. *-o) which took over the personal marker *-t- from
the active, thus creating *-to-r (non-prs. *-to). In like fashion, *wid-e'
took *-t- from the active, but put it at the end, because there already
existed a type ending in *-e-t, namely thematic formations (e.g.
subjunctives) in the active. This gave 3sg *wid-e'-t, on the basis of
which the paradigm was filled by adding *wid-o'-m, *wid-e'-s etc. - If
such an aorist form was re-intepreted as an imperfect, as e.g. Indo-Ir.
*tud-a'-t, there arose a present tud-a'-ti to go with it. The
re-interpretation presupposes that aorists and imperfects could be
functionally confused which is no big problem, since they were both
preterites and must have been equivalent whenever the aspect parameter of
punctuality vs. durativity did not matter.
[]

Jens