Morphosyntactic Transfer from a Spoken Heritage Dialect to the Standard Variety
This study examines cases of morphosyntactic transfer from Syrian Arabic to Modern Standard Arabic (MSA) in the production of heritage speakers who are not only bilingual (L1 Syrian Arabic, L2 English), but also diglossic; Syrian Arabic is their heritage language (HL), and MSA is the form they learned in school. Two control groups, native speakers of Syrian Arabic and learners of MSA, were also included. The proposal presented here is that adolescent heritage speakers of Syrian Arabic have a more developed Syrian Arabic grammar, which results in- transfer to MSA, and that degree and duration of input-output and exposure to both varieties impact the type and number of non-target forms in the production of the studied heritage group. The goal is to find the extent of such transfer, how it is manifested, and whether it is also related to sentence and subject type or other factors. The focus of this study is verbs in SV and VS sentences in MSA, where the subject is a nominal DP and the verb is in the third person. The agreement patterns in VS and SV sentences are asymmetrical in MSA but they are not in Syrian Arabic. The SV order in MSA reflects different agreement patterns with both genders and all three numbers, whereas in Syrian Arabic there is one default non-singular verb form. In this paper I provide a formal account of the differences among the agreement patterns in MSA and Syrian Arabic within the Minimalist framework. Using this approach, a morphosyntactic transfer of agreement features from Syrian Arabic to MSA is argued to be a transfer of T0 features. The results demonstrate that errors in the MSA verb produced by the heritage speakers differ from those of MSA learners and that more than half of the heritage speakers’ errors are compatible with morphosyntactic agreement forms in Syrian Arabic. These findings provide evidence for transfer from Syrian Arabic to MSA. It is possible that when three linguistic systems are competing (L1, L2, and L3) and where L1 is the most dominant, L2 is less developed than L1 but more developed than L3, and L2 and L3 are typologically close, transfer takes place from L2 to L3. More research to address this question is needed. The study contributes to the understanding of agreement in heritage speakers’ production and the phenomenon of transfer in bilingual and diglossic situations.