COSTRUZIONI CON VERBO SUPPORTO. UN'ANALISI CONTRASTIVA ITALIANO-RUSSO IN PROSPETTIVA ACQUISIZIONALE

Collocations are an important component of a native speaker’s language competence (Sinclair 1991, Wray 2002). This study is on a special type of collocation that exists in many languages–Light, or Support, Verb Constructions (LVCs). LVCs consist of a semantically reduced verb together with a noun (as the direct object or embedded in a prepositional phrase) that conveys core lexical meaning to the combination. Light verb constructions often vary cross-linguistically, with languages using different strategies to conceptualize the same denotative situations: different light verbs may be used to denote the same situation. In addition, an LVC in one language can correspond to a synthetic form in another language. Being idiosyncratic, yet semi-productive units, LVCs pose a real challenge for foreign language students. To date, no studies of students’ collocational competence in Russian as a foreign language have been conducted. We also conducted this study in order to contribute information that may be useful to the improvement of monolingual and bilingual ItalianRussian dictionaries, which often treat LVCs unsystematically. Our approach was based on the principles of the Integrated Contrastive Model (Granger 1996, Gilquin 2008), which combines Contrastive Interlanguage Analysis and Contrastive Analysis. Our goals were firstly to establish whether Russian LVCs cause difficulty for Italianspeaking students, and what the regular deviations in the use of these units are and their possible causes; and secondly to conduct a corpus-driven Contrastive Analysis of the semantic preferences of the productive Russian light verbs that give students the most difficulty, and the semantic preferences of their equivalents in Italian. After brief discussion of the main literature and presentation of a semantic-syntactic test battery to delineate different types of verb-noun constructions, we report the results of the study. The first part, exploring Italian-speaking learners’ knowledge of Russian LVCs, involved an investigation of the deviations found in a learner corpus, the quantitative comparison of learners’ and native Russian speakers’ collocation production and two elicitation tests. It revealed deficiencies in learners’ collocational knowledge resulting in misuse, underuse or overuse of support verb constructions. To conceal these deficiencies, two main strategies are used by learners: L1 transfer and avoidance. The second part of the study involved a contrastive corpus-driven cognitive-semantic analysis of LVCs with the productive Russian light verbs davat’ / dat’ ‘to give’, delat’ / sdelat’ ‘to do, to make’, brat’ / vzjat’ ‘to take’, prinimat’ / prinjat’ ‘to take’ and their Italian counterparts dare ‘to give’, fare ‘to do, to make’ and prendere ‘to take’. The Construction Grammar model (Goldberg 1995, 2006) and the notion of a family of constructions (Goldberg 1995: 140, Croft 2009:161, 162) were adopted to account for relatedness of LVCs within each language. Productive systematic metaphors that license extensions from the basic meaning of the verbs and semantic classes of nouns that combine with set verbs were identified. The analysis resulted in a structural-semantic classification of LVCs in the two languages, and detected convergences and divergences and allows for some generalizations. The findings of this study, in addition to its theoretical significance, may be useful in teaching Russian as a foreign language and could be of interest for lexicography.