Abstract

The production of clitics is an excellent clinical marker of SLI in Italian at 5 years (Bortolini et al., 2006) and at 7 years (Arosio et al., 2014) identifying more than 90% of the children affected with SLI at both ages. However, clitic production is problematic in other conditions (Leonini, 2006; Belletti & Haman, 2009), among which early L2 acquisition, acquisition by cochlear implanted (CI) children, acquisitionby children with dyslexia. This is a serious problem for the identification of L2 children with SLI , of children with SLI among CI children. It also raises the question of why clitics are vulnerable in all these acquisition modes. Clitics are less of challenging for Greek children with SLI , although they are also vulnerable in this language, with differences between direct and indirect object clitics. Finally, reflexive clitics are not challenging at all for children with SLI . This paper aims at providing a viable solution for the identification of SLI among L2 and CI children, an explanation for why some kind of clitics are vulnerable in all the conditions in Italian and some are not, an explanation for the crosslinguistic differences. I show that quantitative and qualitative differences are observed among the various populations and that different profiles distinguishes these populations. Finally, I will offer a proposal whereby the difficulties with clitics stem from various intervention effects, which are not featured in those constructions in which clitics are not problematic.