Acquiring distributional patterns in a closely related second language : Referent introduction, maintenance and prefield constituents in Dutch and Swedish

Abstract: This paper presents an explorative study of informal written narratives and oral picture descriptions in the native language as well as in the second language (L2) of six intermediate/advanced Dutch L1 learners of Swedish. Data from six Swedish L1 native controls, performing the same tasks as the learners, were also collected. Quantifications forthe following aspects of information structure were made for the transcribed oral descriptions: 1) reference introduction, including occurrences of spatial/existential constructions, 2) reference maintenance per type and finally 3) elements in the prefield. For the written material, elements in the prefield were analysed. Pronouns in all clausal positions andadverbials in the prefield were analysed in detail for both oral and written material. The following research questions were asked: Do Dutch L1 and Swedish L1 differ in their patterns of information structure within the area of referent introductions and maintenance and prefield constituents in picture descriptions? Do the L2 learners adhere to the target language(Swedish) patterns, the patterns of their mother tongue (Dutch), or does the interlanguage exhibit different patterns? In the written data, differences in percentages for prefield constituents emerged where Dutch L1 showed a more adverbial-initial pattern. No major differences were found in the oral data, where both languages exhibited a pattern of two thirds subject-initial clauses and one third adverbial-initial clauses. Some further differences were found concerning the distribution of subtypes of prefield constituents (such as place adverbials, subject pronouns, subject NP’s). Concerning the perspectives taken for referent introduction, Dutch L1 was slightly more spatial than Swedish L1. However, both languages exhibited a predominantly existential perspective. Substantial individual variation was foundin all three language varieties and for all features investigated. The learners in their Swedish L2 mainly followed the norms of the L2, but also showed patterns similar to those of their L1 as well as patterns neither connected to the L1 or the L2. The strict dichotomy ofspatial/existential perspective of referent introduction proposed by e.g. Carroll et al. (2000) was found to exclude many cases of referent introduction. Drawbacks of the present study are the small quantity of data and the lack of comparable studies; thus, this subject and language combination need to be researched further.