The encoding of definiteness in L2 Norwegian: A study of L1 effects and universals in narratives written by L1 Russian and L1 English learners
Abstract
Norwegian expresses definiteness grammatically partly with inflection (definite NPs) and partly with articles (indefinite NPs and some definite NPs). The present study investigates the use of grammatical definiteness in Norwegian as a second language among adult Russian learners (N=7) and English learners (N=4) living in Norway. The data consist of written Pear Story retellings (Chafe, 1980a) collected at three data points within approximately one year. The overall assumption guiding the research design is that the linguistic structure of the first language is a major force in second language learning. In the specific case of L2 acquisition of definiteness, most research conducted on L2 English has documented deviant patterns in the article use of learners with an L1 not possessing any such category (Austin, Pongpairoj, & Trenkic, 2015; Butler, 2002; Huebner, 1983, 1985; Chaudron & Parker, 1990; Ionin, Ko, & Wexler, 2004, Ionin, Zubizarreta, & Maldonado, 2008; Liu & Gleason, 2002; Master, 1987; Parrish, 1987; Robertson, 2000; Tarone & Parrish, 1988; Thomas, 1989; Trenkic, 2007, 2008, 2009; Trenkic & Pongpairoj, 2013; Trenkic, Mirkovic, & Altmann, 2014; Young 1996). However, an interaction between the L1 and universal principles of language and second language learning is often hypothesized. In the present study learners with an L1 not exhibiting a category of grammatical definiteness (Russian) are therefore compared to learners with an L1 exhibiting a category of grammatical definiteness (English) that is partly similar to and partly diverge from that of the target language (Norwegian). Following previous research (e.g., Butler, 2002; Huebner, 1983, 1985; Master, 1987, Parrish, 1987; Tarone & Parrish, 1988; Thomas, 1989, Trenkic, 2002b; Young, 1996) and in particular Sharma (2005a), the present study employs two models of analysis: The Semantic Wheel of NP Reference (Bickerton, 1981; Huebner, 1983) and The Taxonomy of Assumed Familiarity (Prince, 1981). The first model classifies NPs in terms of two binary categories: [± specific referent] and [± assumed hearer knowledge], whereas the second model distinguishes discourse referents from each other on the basis of degree of givenness. Finally, the study performs a post-hoc analysis of a set of specific NP constructions in Norwegian in order to arrive at a deeper understanding of the impact of structural features of grammatical definiteness. Through these analyses the study seeks to answer how L1 related effects and universal principles of discourse interact in the learners’ encoding of grammatical definiteness in Norwegian. How the crosslinguistic relations between Norwegian, English and Russian affect the L2 outcome are accounted for by the incorporation of two different frameworks for cross-linguistic relations: one originating from Ringbom (2007) and the other from Tokowicz and MacWhinney (2006). The data sample of the present study is limited and its strengths are thus mainly found in the fact that it highlights and explores complexity and diversity rather than creating a basis for statistical generalization. It further capitalizes on the strengths of longitudinal data by performing several in-depth analyses that report on both similarities and differences in individual profiles and between the two L1 groups. The results of the analyses are, as mentioned above, interpreted in light of the research questions and prediction. The main findings of the study have emerged from an explorative process of conducting several analyses. Chiefly, it is documented that omission of the indefinite article is a characteristic feature of the L1 Russian learners’ development only, while the L1 English learners, unexpectedly, seems to be prone to omit the definite inflection at initial stages. The observation of these patterns appears to be reinforced when specific NP constructions are investigated more closely. Yet, when all three data points are examined, it becomes clear that the development towards the target norm is more rapid within the L1 English learner group than within the L1 Russian learner group. This is particularly salient in the encoding of definites for the L1 Russian learners, where only minimal change occurs during the time of the data collection. In short, the behavior of the L1 Russian learners is largely predictable from the literature on L2 English article acquisition, whereas additional frameworks are required in order to explain the behavior of the L1 English learners. Finally, it is clear from the study as a whole that the multiple approaches taken in the analysis, which includes both the semantic/pragmatic level, the discourse level, and the level of syntax, have given access to insights that would not have been obtained by one analytical approach alone.