The construction of an annotated corpus for the analysis of Italian transfer predicates

Abstract

I have presented in this paper some research carried out with the help of the electronic dictionaries and grammars that I built for the automatically recognition and annotation of Italian transfer predicates with a software called NooJ. The analysis of the nominal and verbal dative constructions has been carried out on the basis of the lexicon-grammar classification of Italian verbs, focusing on the various constructions produced by the application of transformations. Furthermore, particular attention has been given to the ambiguity of some dative nominal sequences. I have proposed a hypothesis that accounts for the fact that these structures, though in principle ambiguous, usually have a preferential reading, due to the adjacency of the complements. This hypothesis has been verified by applying the electronic dictionaries and grammars to an Italian corpus and the results drawn up with the automatic corpus analysis show that the most frequent dative constructions are those involving only the object. Finally, the results of this research have been compared to those drawn up by T. De Mauro (1965) in his study on the dative in classical Greek. Both results not only confirm the linguistic intuition on the preferred vs. the non-preferred interpretation, but also show the discrepancy between the theoretical dative constructions and the real usage.