Sanskrit Tense & Aspect

[Vidhyanath Rao]

"Peter &/or Graham" <petegray at btinternet.com> wrote:

>The difference in Sanskrit is slight, at best.   I quote Stenzler, (1997):
>"In Classical Sanskrit the aorist is used in narrative as a past tense
>alongside the imperfect and perfect, without any distinct function." There
>is a slight difference in time of reference, (recent or more remote past)
>but this is not aspect.   Perhaps it is different in Vedic, although my
>Vedic grammar tells me the names refer to the formation, and not to the use
>of the tenses.

The major problem with `Classical Sanskrit' is that there differences based
on the primary language(s) of the author/audience and also on the genre. For
example, in the dramatic dialogues, past tense is expressed using only the
``past passive participle''. On the other hand in the Upanishads and in
earliest Pali, the PPP has a resultative sense. In Kavya Sanskrit,  the
various forms are used in the same slot, but in different frequencies that
correspond to the roles taught in traditional grammar. This makes it hard to
see the evolution after 300 BCE.

----------

Miguel Carrasquer Vidal <mcv at wxs.nl> wrote:

> The question isn't *if* there was something distinguishing the
> three forms.  Of course there was, or we wouldn't have three
> *forms*.

Does this mean that different stem formants must have had some distinction
in meaning too? If you assume that there was a difference in meaning at some
point, how do we know that  point was before late PIE. Perhaps, the
different stem formants were derivational affixes and collapsed into
grammatical categories after the languages split up.

> The question is *what* distinguished them.
> The imperfect vs. aorist distinction was one of (im)perfective
> aspect, that much is clear from the way it is formed (present
> stem vs. aorist stem) and the attested uses in Greek.

This is not at all clear. This argument relies on two unmentioned
assumptions, common in IE aspectology, but questionable on general grounds.
The assumptions are that ``usability in the present => atelicity'' and
``atelicity => not usable in perfective. The second is well known to be
false for many languages with aspect (see Dahl, Tense and aspect systems).
The first depends on how telic events are handled. It may apply to languages
with serial verb constructions, but there is no evidence of such for any
stage from PIE to Vedic. Otherwise, it is true only of languages with aspect
(not even then, for example Lith. jis dabar perras^o lais^ka, `he is
rewiring
a letter'; Lith also allows presents like is^be'ga, nulipa which are called
imperfective in Dambriunas' grammar, but the glosses, `runs out of', `climbs
down', suggest telicity). So this argument boils down to assuming aspect to
prove it.

And the question of the pathway to Vedic is still there. The attempts I have
read are quite weak: Gonda has nothing better than ``national character of
Indians'' for the use of the alleged imperfective in narration. I hope that
I need not belabor this any further. Hoffman posits an intermediate stage in
which aspectual distinctions were limited to narratives and remote past (not
``statements'' where aorist was used exclusively). But without other
examples of such a division, it is quite unbelievable; no language with
aspect is known to do that. Also in such a stage, the perfective would be
much more common as perfectives are the usual narrative form.Without other
examples of such an evolution, it is hard to see how the imperfect was
generalized in narration. [Recently I was referred to H. Rix, Historische
Grammatik desGriechischen. Darmstadt 1976 p. 192 sqq., and  G. Meiser,
Historische Laut- und Formenlehre der lateinischen Sprache, Darmstadt 1998,
p. 180 sq. The former is currently checked out from the local library and
the latter is not in any nearby library. I will appreciate it if any one can
summarize what these say.]

Finally, `what else could it be' could have answers other than aspect.
Cross-linguistic studies, like Dahl Tense and aspect systems and Bybee et
al, The evolution of grammar, turned up an interesting category called
`completives' by Bybee et al. These are emphatic/ highly marked forms that
are not the usual forms in narration, and often contrast with either a
simple past or a perfective that is the usual form in narration. [An
in-depth discussion of such a form in Tamil, using viDu as an auxiliary, is
found in Annamalai, International journal of Dravidian linguistics vol
11(1982),  pp 92--122.] Sanskrit usage might go back to such a distinction
rather than perfective-imperfective distinction.