Differently from English, Italian has a rich morpho logical system and a relative free word-order. For these reasons, the suitability of a "full-scope" constructional approach to Italian is not given. Although Goldberg's (1995, 2006) version of Construction grammar language is constructions all the way down (or up), one could still argue that in Italian, i.e. a language rich in morphology, abstract, grammatical constructions do not play a role, thus weakening the constructionist view. One of the strong points in favour of goldberg's approach is that argument structure constructions in English have been found to be highly predictive of sentence meaning (Goldberg et al. 2005), which provides a motivationfor their early acquisition by children. Many of such studies and evidences are still missing for Italian. This contribution will therefore attempt to start filling this gap by testing the predictive power of the Italian Caused Motion Construction. Data is taken from the CHILDES database (MacWhinney 2000) and annotated according to constructional properties and verb meaning. The annotation is then used to calculate the Cue and Category Validity (Murphy 2002) of both the Construction and the main verbs, which measures their predictive power (i.e respectively their reliability and availability) in relation to the overall sentence meaning. Results show that the Italian Caused Motion Construction is not only more reliable than verbs as a predictor of overall sentence meaning, but it is also more available.
How predictive are grammatical constructions in Italian? The case of the caused-motion construction
Quochi;Valeria
2012
Abstract
Differently from English, Italian has a rich morpho logical system and a relative free word-order. For these reasons, the suitability of a "full-scope" constructional approach to Italian is not given. Although Goldberg's (1995, 2006) version of Construction grammar language is constructions all the way down (or up), one could still argue that in Italian, i.e. a language rich in morphology, abstract, grammatical constructions do not play a role, thus weakening the constructionist view. One of the strong points in favour of goldberg's approach is that argument structure constructions in English have been found to be highly predictive of sentence meaning (Goldberg et al. 2005), which provides a motivationfor their early acquisition by children. Many of such studies and evidences are still missing for Italian. This contribution will therefore attempt to start filling this gap by testing the predictive power of the Italian Caused Motion Construction. Data is taken from the CHILDES database (MacWhinney 2000) and annotated according to constructional properties and verb meaning. The annotation is then used to calculate the Cue and Category Validity (Murphy 2002) of both the Construction and the main verbs, which measures their predictive power (i.e respectively their reliability and availability) in relation to the overall sentence meaning. Results show that the Italian Caused Motion Construction is not only more reliable than verbs as a predictor of overall sentence meaning, but it is also more available.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
