|Título||Resolving a Classic Case of Lexical Ambiguity|
|Autoría||Bryan Leferman (University of the Basque Country (UPV/EHU))|
|Resumo||The lexical aspect evaluative adjectives (EAs) (e.g. brave, smart, rude) is a puzzle because EAs appear to shift from stative to eventive/activity predications. On the one hand, EAs in the present simple ((1a)) are stative and they pass standard diagnostics classifying them as individual-level predicates. On the other, EAs appear in the progressive aspect ((1b)) and in pseudo-clefts ((1c)).|
(1) a. Eva is rude = Stative: individual-level predication
b. Eva is being rude = Eventive: activity predication
c. ?What Eva did was be rude = Eventive: activity predication
This apparent split between stative and eventive/activity EA predications is frequently analyzed as coercion or as polysemy.
Objective: this talk aims to show that coercion and polysemy are unwarranted in the analysis of the data in (1) because close consideration of EA lexical aspect reveals that it remains stable across them. The objective is to show that all of these examples follow from an EA denotation specifying its lexical aspect as stative causative.
I propose that the pattern in (1) follows from the denotation in (2): EAs select an individual external argument, two state arguments in a causal relation, and a propositional complement. It will be shown that EAs are factive, thus presupposing their complement.
(2) ║EA║ = λqλs2λs1λx. [EA(x, s1, s2, q) & CAUSE(s1, s2)], only defined if q = 1
Results: EAs are an interesting case study in the context of coercion and polysemy because they highlight how intuitions of ambiguity can follow from a unified lexical entry, underspecification, and independent factors. The talk concludes with discussion of the nature of semantic content regarding the stative/eventive divide, supporting the position that truth-conditions are semantically underdetermined.
|Horario||Mércores 13 de xuño | 18:00 - 18:30 | Aula: C3|