From: Mari Broman Olsen (
Date: Thu Mar 26 1998 - 10:50:07 EST

As commented on by Rolf Furuli:

I used it sloppily. I don't do so in published work. Of course
'complete' is better, without the past tense. But my intent here was
not to present my technical definition (which is actually 'focus on
the coda of an event/state at reference time', so something could be
complete(d) in the future), but to highlight the difference between
imperfective and perfective. Perfective MUST be complete (and what it
means to be a complete event differs based on the lexical aspect
(aktionsart) of the verb), but imperfective need not be.

Yes, there is an association between complete and past: in languages
without tense, perfective is generally (pragmatically) used for past
time and imperfective for present (note the association with ongoing,
or 'focus on the nucleus of an event/state,' the content of the
'nucleus' again provided by lexical aspect).

Interestingly, and this hasn't been brought up (to my knowledge) here:
languages with both tense and aspect always make more aspectual
distinctions in the past tenses than in unmarked (present tense) or

I think the real difference between Rolf and me (and I think neither
of us has read the other, except on the list) is in approach: I want
individual accounts of a language to feed into an explanatory
predictive approach for describing other languages, and how kids
acquire them with ease (and simpler systems, like algebra and logic,
only with great difficulty). I think the answer lies in discovering
the universal underpinnings (hardware?) of Language, and pushing the
universal semantic categories so they cover (without being too
diluted) the range of phenomena we see, so that they can cover the kid
coming to her language at the start. I think Rolf is more concerned
that the theory cover the nuances and specifics in the individual
languages (and he no doubt knows Greek much better than I (or me ;-)
)). We agree in the building blocks (I think, namely Aktionsart,
Aspect, Tense), and that good theories of these should be explanatory.

Mari Broman Olsen
Research Associate

University of Maryland Institute for Advanced Computer Studies
3141 A.V. Williams Building
University of Maryland
College Park, MD 20742

PHONE: (301) 405-6754 FAX: (301) 314-9658

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:39:17 EDT