Evelina Leivada: Where do Adjective Ordering Restrictions Come From?
Seminari del CLT
Where do Adjective Ordering Restrictions Come From?
EVELINA LEIVADA (U. of Tromsø)
Divendres, 24 de maig de 2019
Aula 202, 15:30
Linguists have often noted that adjective ordering shows a cross-linguisti
cally uniform character (Sproat & Shih 1991, Cinque 1999, Scott 2002). Two types of origin have been proposed for the attested ordering constraints: (i) syntactically imposed restrictions that posit a rigid hierarchy and (ii) restrictions of a cognitive basis that are sensitive to many factors such as the encoding of absoluteness and subjectivity, and the morphophonologi cal weight. Under the first approach, the universal order is encoded in Universal Grammar in the form of a rigid syntactic hierarchy (Rizzi & Cinque 2016, Panayidou 2013). The second approach suggests that the ordering is less rigid, permitting for flexibility that is the outcome of various clashing factors.
The present study aims to tap into the universality and the rigidity of adjective ordering restrictions from a psycholinguisti
c point of view. We have run a timed forced-choice task with monolingual speakers of Greek (n=140). The task involved three conditions: 1. size-nationalit y, 2. color-shape, 3. subjective comment-materia l. Each condition had two orders. In the congruent order, the size adjective preceded the nationality adjective, in agreement with what is taken to be the order imposed by the hierarchy (Scott 2002). In the incongruent order, the hierarchy was violated and the adjective ordering was reversed. Since reaction times were measured, all test structures were carefully matched for length and syntactic structure across orders and conditions.
Our subjects largely accepted the incongruent orders as correct across conditions.The absence of any context-given information in our task entails that speakers could not have interpreted the incongruent orders under a contrastive focus reading that would legitimize the violation of the hierarchy. We discuss our results in the light of different hypotheses, both within and outside the cartographic approach, ultimately suggesting that they grant support to the second approach to ordering restrictions.