E4: Specific language impairment and early second language acquisition: Differentiating Deviations in morphosyntactic acquisition

Project leader: Prof. Dr. Monika Rothweiler, FB 12, University of Bremen (www.monikarothweiler.de)

Collaborators: Dr. Manuela Schönenberger, Franziska Sterner
      
This project investigates the acquisition of syntax and morphology in early L2 German by children with L1 Turkish. These children either do or do not have SLI.

The following three questions are central to the project:

• What does the acquisition of L2 German look like in young children whose first exposure to German is at age 3?
• What are the typical features of SLI in the L2 German of children with SLI who start to acquire German at age 3?
• What are the clinical markers of SLI in Turkish?

To address these questions we examine the longitudinal data of more than 20 successive bilingual children who have been exposed to Turkish only for the first three years of their life and whose first contact with German is at age 3 when they start to attend kindergarten. We collected data from children without SLI, as well as from children who appear to have SLI, over a period of 24-30 months. We also collected data over a period of 12 months from older successive bilingual children (6-8 years of age) with the same age of onset and who have been diagnosed with SLI. Since SLI manifests itself in every language a speaker acquires, data on L1 Turkish have also been collected from time to time to gain some insight into these children's linguistic development in Turkish. The combination of languages (Turkish/German) and the type of acquisition (L1/early L2) as well as the absence vs. manifestation of SLI in the group of children we are studying enables us to attempt to answer the three questions central to the project.

The first question concerns the early successive acquisition by children without SLI and touches upon the issue of a critical period in language acquisition. The results obtained so far support the hypothesis that the linguistic development of children who start to acquire a second language before age 4 parallels that of monolingual children. The acquisition of clause structure and subject-verb agreement in German by these successive bilingual children closely resembles that of monolingual children (Rothweiler 2006). A comparison of these data with data from the Wegener corpus of Turkish-speaking children whose first contact with German is at age 6 shows that transfer from L1 Turkish does not seem to play a role in the children with a much earlier age of onset, i.e. age 3 (Chilla 2008, Kroffke & Rothweiler 2006). We are currently investigating other domains of grammar which involve lexical learning, as, for example, gender. If it turns out that there are differences in this domain between successive bilingual children and monolingual children a closer examination of vocabulary size and transfer from L1 will be required to determine whether these factors influence the acquisition of gender.

The second question is concerned with the acquisition of grammar in L2 German by children with SLI, that is, children whose language development in Turkish also differs from that of typically-developing Turkish-acquiring children. Findings on the acquisition of the German sentence structure by the successive bilingual children with SLI mirror those of monolingual German children with SLI (Chilla 2008). These results are not really surprising, since the successive bilingual children without SLI pattern with typically-developing monolingual German children in this respect as well. These findings are important for the identification of SLI in diagnostic screening. However, a detailed investigation into other grammatical domains is necessary to confirm the inference that the development in successive bilingual children with or without SLI is parallel to that of monolingual children with or without SLI.

The third question relates to the manifestation of SLI in Turkish. Since there are no studies of SLI in Turkish our project tries to identify clinical markers of SLI in Turkish. An examination of the Turkish data collected in this project suggests deviations in case morphology and verb morphology (Babur, Rothweiler & Kroffke 2007; Chilla & Babur in press).