Stavrakaki, S., Vogindroukas, I. Chelas, E., Ghousi, S., (in press), Subject-Verb Agreement in SLI and Child L2: A Comparative Approach to Greek SLI and (un) Impaired Child L2, in Language Acquisition and Development: Proceedings of GALA 2007, Cambridge University Press
Recent studies of Specific Language Impairment (SLI) and child L2 have raised two interrelated questions. The first one concerns the characteristics of language development in SLI compared to unimpaired L2 learners whereas the second one concerns the differential diagnosis of impaired children learning a second language in a multilingual society (Grueter, 2005; Hakansson, 2001; Paradis, in press).
This study addresses the question of whether performance on subject-verb agreement marking in Greek can distinguish Greek speaking children with SLI, Albanian speaking children with SLI learning Greek as L2, and unimpaired Albanian children learning Greek as L2. It also explores the implications for theories of SLI and child L2 acquisition.
Recent studies of Greek preschool children with SLI indicated incomplete acquisition of the agreement paradigm with problems focussing on the 2nd person singular and plural (Clahsen & Dalalakis, 1999, among others). Despite the improvement in subject-verb agreement at school age (Stavrakaki, 2001), for preschoolers with SLI difficulties with verb endings constitute a marker for SLI.
To investigate whether such difficulties exist across other learners at certain stages of their language development at preschool age and compare with findings from SLI children speaking Greek, we studied the spontaneous production of subject-verb agreement in Greek by (i) five unimpaired Albanian children aged 4;6-5;6 who attended Greek nursery schools in Thessaloniki; at the time of data collection their exposure to Greek varied between 1-2 years (ii) one 5 and a half-year-old Albanian SLI child who attended a Greek nursery school in Thessaloniki. The child showed severe difficulties with language development in Greek and Albanian and fulfilled the standard criteria for SLI (Stark & Tallal, 1981). At the time of data collection he had been exposed to Greek systematically for two years at the nursery school and received speech and language therapy services in Greek (iii) six Greek preschool children with SLI aged 4;4-6. Spontaneous production included elicitation of different verbal forms by the means of a semi-structured interview that was recorded on audiotape.
Data analysis indicated that unimpaired L2 learners of Greek demonstrated ceiling and near ceiling performance and showed evidence for the acquisition of verbal inflection in Greek. By contrast, the Greek SLI children showed selective problems with subject-verb agreement limited in particular to the 2nd person singular and especially to the 2nd person plural confirming the pattern reported in the literature. The Albanian speaking SLI child-L2 learner of Greek showed the same pattern of performance as monolingual Greek speaking SLI children in Greek, that is, selective problems with subject-verb agreement especially with the 2nd person plural.
Based on these results, we argue that performance on subject-verb agreement can successfully distinguish SLI children from children learning Greek as L2. In particular, we argue that the full verbal paradigm is available in children learning Greek as a second language (cf. Haznedar & Schwartz, 1997; Varlokosta, 2001) while selective deficits in subject-verb agreement characterize Greek speaking SLI children (with Greek being L1 or L2) at preschool age. Since the SLI children are not completely unable to produce the correct verb suffix, we suggest that the selective deficits in subject-verb agreement in preschool children with SLI indicate unsuccessful performance of feature checking operations (cf. Stavrakaki 2006).