Agreement and Interference in Direct Object Clitic Production in Italian Monolingual Children

Italian speaking typically developing children optionally omit third person direct object clitics (3DO) until at least age 4 years. In a set of two studies, we investigated whether clitic omissions depend on intervention phenomena. We discussed a model of syntactic intervention in 3DO clitic production in which the subject of a sentence might intervene in clitic production at an intermediate stage of the sentence derivation, when the clitic moves from its initial merging position to a higher position above the subject. We argued that omissions and errors in 3DO clitic productions arise when retrieving from memory the clitic to achieve an agreement relation in a context of syntactic intervention. We argued that working memory limitations impact on the clitic retrieving operation cued by lexical and representational features. In Study 1, we elicited the production of third singular clitics in sentences with a full lexical subject. The third singular subject and the third singular clitic were matched or not in gender. Results of Study 1 showed that when there is a gender feature mismatch between the subject and the 3DO clitic children optionally make clitic gender errors or even replace the 3DO clitic with a post-verbal full DP. We argued that these results could be explained in an intervention model in which the external verb lexical argument (the lexical subject) is erroneously retrieved for achieving the agreement operation involving the movement of the clitic to its surface position. In Study 2 we investigated whether the problems in clitic production found in Study 1 depend on phonological priming or structural intervention. We elicited the production of third singular clitics in sentences with a silent pro subject. As in Study 1, the third singular subject and the third singular clitic were matched or not in gender. Results of Study 2 showed that in sentences with a mismatch in gender features between the null subject and the clitic, children tend to produce a clitic with the incorrect gender or to optionally replace it with a full lexical post-verbal DP, regardless of the gender of the target clitic. This suggests that a null subject intervenes in the same way a lexical subject does in the derivation of clitics and, consequently, that the gender features inherited by a null subject via its anaphoric link with its antecedent have the same grammatical status of gender features conveyed at a lexical level. Overall results indicate that the interference errors are not dependent on phonological attraction but rather have a structural nature and are modulated by short-term memory resources.