Does gender make a difference? Comparing the effect of gender on children's comprehension of relative clauses in Hebrew and Italian

In this paper we assessed the effect of gender morphology on children's comprehension of object relatives in Hebrew and Italian. We compared headed object relative clauses in which the relative head (the moved object) and the intervening embedded subject have the same or different genders. The participants were 62 children aged 3;9–5;5, 31 speakers of Hebrew and 31 speakers of Italian. The comprehension of relative clauses was assessed using a sentence–picture matching task. The main result was that whereas gender mismatch sharply improved the comprehension of object relatives in Hebrew, it did not significantly affect comprehension in Italian. In line with our previous work (Friedmann et al., 2009), we propose that the children's problem in the comprehension of headed object relatives stems from the intervention of the embedded subject between the moved relative head and its trace. We ascribe the different behavior of children in Hebrew and in Italian to the different status of the gender feature in the two languages: in Hebrew, gender is part of the featural composition of the clausal inflectional head, hence it is part of the feature set attracting the subject, whereas in Italian, where tensed verbs are not inflected for gender, it is not. Under the assumption that intervention effects are amenable to the locality principle Relativized Minimality, it is expected that only features functioning as attractors for syntactic movement will enter into the computation of intervention. We thus account for the different effect of gender mismatch in object relative comprehension in the two developing systems. Thus, the main finding of this work is comparative in nature: there is no effect of gender per se; rather, the potential effect of gender is crucially modulated by the morphosyntactic status of the feature in each language.