Universiteit van Amsterdam


Institute for Logic, Language and Computation

Please note that this newsitem has been archived, and may contain outdated information or links.

7 July 2000, DIP Colloquium (final session of season)

7 July 2000, DIP Colloquium (final session of season)
Title: Puzzles of UNTIL: stativity, negation, and the perfect
Speaker: Anastasia Giannakidou, University of Groningen
Location: Philosophy Department, MFR Ground Floor
Date and Time: Friday 7th July 2000, 15.15-17.00

The goal if this talk is to re-examine the durativity/stativity of the family of meanings associated with UNTIL in the light of the hypotheses that (a) negation is an aspectuality operator tranforming events into states (Verkuyl 1993, de Swart 1996, de Swart and Molendijk 1990,among others), and (b) the perfect is likewise an state-yielding aspectuality operator.

It will be shown that the former hypothesis cannot be proposed as a general property of negation across languages, and is in need of revision. I will conclude (with Kamp and Reyle 1993) that negation need not change the eventuality type of the predicate: states remain states under negation, and events remain events. To this end, I examine contrastively the family of meanings associated with until and its Greek counterpart mexri. Though mexri is durative, like until, it cannot be used with negated eventive predicates (signalled by perfective aspect). In these cases, a lexically distinct NPI is usedópara monon. With negations of states or processes (and imperfective aspect) the expected ambiguity arises between wide and narrow scope readings of mexri wrt negation.

The perfect, on the other hand, does indeed yield states (Vlach 1993, Kamp and Reyle 1993). Yet it is shown that present perfect states are incompatible with mexri and until. This will be explained as a semantic clash between the present perfect semantics, which requires that the state include the utterence time (n), and the meaning of UNTIL which sets an endpoint prior or including n. The past perfect is not affected because the duration of the state in this case does not reach n, hence an endpoint prior to n can be set. Negation, similarly, voids the problem by cancelling n as the expected endpoint.

As a result of the proposed analysis, Karttunenís (1974) theses that there is an NPI-until, and that this until is punctual rather than durative is confirmed by analogy to the Greek data. Data from Dutch and German will also be shown to confirm this result.

More information can be found on the DIP (Discourse Processing) homepage, or by contacting the DIP Colloquium organizing committee at DIP@hum.uva.nl

Please note that this newsitem has been archived, and may contain outdated information or links.