Investigating the roles of feature reassembly and linguistic input in later-stage second language acquisition: a case study of aspectual development in university learners of French

Abstract

This thesis centres on the second language (L2) development of viewpoint aspect, a grammatical property definable as the contrast between perfective viewpoint, which indicates a completed eventuality, and imperfective viewpoint, which provides an ongoing or non-complete perspective (Smith 1991, 1997). This property is well-established as challenging to acquire in an L2 for native (L1) speakers of English, due to the limited grammaticalisation of viewpoint aspect in English (Montrul & Slabakova 2002, 2003; McManus 2015). Yet studies examining L2 viewpoint aspectual development in L1 English speakers (e.g. Domínguez et al. 2011, 2017) do not always find that their results align consistently with L1 influence (frequently operationalised using the Feature Reassembly Hypothesis (Lardiere 2003, 2005, 2008, 2009), a seminal generative framework). This project analyses the production and comprehension of L2 French viewpoint aspectual form-meaning mappings in n = 43 L1 English university-level learners of French, studied cross-sectionally prior to and following a “year abroad” in a French-speaking country. This L2 data is combined with a frequency-distributional analysis of L1 French from two sources: experimental production data and a >10,000-word corpus sample, the latter sourced from the freely available CFPP2000 (http://cfpp2000.univparis3.fr/index.html). The aim is to assess the extent to which both Feature Reassembly predictions and frequency distributional information from the input can respectively explain trajectories of L2 viewpoint aspectual development, and consequently evaluate whether an approach which unites generative and input-based principles can usefully contribute to our understanding of L2 acquisition as a whole. It is found that learners are indeed sensitive to frequency-distributional information – as is reflected in their production of French aspectual forms – and suggest that they are also sensitive to input changes, as shown by the unexpected lower production accuracy of the post-year abroad group. In conclusion, it appears that an approach which integrates both Feature Reassembly and input information may provide a more comprehensive picture of L2 (aspectual) development. Keywords: L2 French, L1 English, viewpoint aspect, imperfect, input, feature reassembly, study abroad, corpus analysis

More information

Identifiers

ORCID for Laura Dominguez: orcid.org/0000-0003-2701-2469

Catalogue record

Export record