The development of aspect in a second language

This thesis investigates the second language (L2) acquisition of aspect. Aspect is considered to be a universal property of language (Chung and Timberlake, 1985; Comrie, 1976; Klein, 1994, 1995; Smith, 1991, 1997). Therefore, all natural languages are thought to be able to convey the same aspectual meanings. However, languages do not always convey these meanings in the same ways. For example, although French and English are able to convey viewpoint aspect by tense, they differ from each other in the particular aspectual meanings they map to individual tenses. In other words, English and French differ in how they pair form with meaning for viewpoint aspect. In German, viewpoint cannot be conveyed by tense alone and semantics and pragmatics are required for viewpoint interpretation (Bohnemeyer and Swift, 2004). So whilst languages are able to convey the same meanings, there are differences in how they go about doing this. This raises the question of the role of learners‘ L1 in the L2 development of aspect (e.g. Domínguez, Arche and Myles, 2011; Gabriele, 2005, 2009; Montrul and Slabakova, 2002, 2003; Slabakova, 2000, 2002, 2008). In other words, do differences in how aspect is expressed in the L1 affect how it develops in the L2? The role of prototypes in the L2 development of aspect has been widely documented as an influencing factor (e.g. Andersen and Shirai, 1994, 1996; Bardovi-Harlig and Bergström, 1996; Bardovi-Harlig, 2000; Labeau, 2005; Salaberry, 1998, 2000). The Aspect Hypothesis (Andersen and Shirai, 1994, 1996) indicates that learners are sensitive to prototypes: L2 development is characterized by initially pairing prototypes of viewpoint with situation type. These form-meaning relationships then become less restricted as L2 proficiency increases. Central to this thesis is the effect to which L1 form-meaning pairings and prototypes affect the L2 development of aspect. This study‘s research questions are as follows: How do learners express perfective and imperfective viewpoint aspect? What role do L1 form-meaning pairings have in the L2 development of viewpoint aspect?  What role do semantic prototypes have in the L2 development of viewpoint aspect? What are the theoretical implications of the role of L1 background and semantic prototypes on L2 development more generally? Participants are English- and German-speaking university learners of French L2 (n=75) and a control group of French native speakers (n=6). C-test results established two significantly different learner groups: a low group and an advanced group. Learners were further divided into groups based on L1 background, resulting in: English low group (n=19), German low group (n=19), English advanced group (n=19), German advanced group (n=18). Participants undertook three tasks: two picture-based spoken narratives and a Sentence Interpretation task. Results show significant differences between learners in production and interpretation. Differences are attributable to both proficiency level and L1 background. English low group learners are significantly different to German low group learners for viewpoint marking, especially in imperfective contexts, whereas English and German advanced group learners are not significantly different from each other. Furthermore, tense selection is subject to a semantic prototype influence, with advanced group learners influenced more than low group learners. It is argued that L1 form-meaning pairings for viewpoint aspect significantly influence L2 development at the early stages of L2 development. However, as L2 proficiency increases L1 influence begins to recede and learners develop L2 form-meaning pairings. At the more advanced stages of L2 development, semantic prototypes significantly affect tense use. Furthermore, prototypical effects appear to increase with proficiency, contrary to the Aspect Hypothesis.