University of Sheffield
Formulaic language and L2 syntactic development: A multiparadigmatic approach
Abstract
dc:description.abstractThis thesis presents a novel, multiparadigmatic approach to the role of formulaic language in the development of second language (L2 ) syntax. It adopts theoretical frameworks and concepts from generative and usage-based linguistics to analyse two longitudinal corpora documenting interlanguage development from the initial state through transitional state grammars. These datasets constitute L1 Spanish/Catalan learners of L2 English, and L1 English learners of L2 French. Similar formulaic expressions (FEs) are identified in both corpora at the initial state that learners are able to produce fluently upon contextual cues in absence of related L2 syntactic competence. I track learners’ use of these FEs and evidence of related syntactic competence across the data collection periods, as measured by knowledge of underlying computational properties (generative) and schematic patterns (usage-based). Through adopting a multiparadigmatic approach, similar developmental trends are observed in both corpora that would have otherwise been missed if relying on one framework alone. Outside of the FEs, learners first demonstrate knowledge of lexical categories only, before knowledge of L2 functional categories emerges later. Importantly, the use of FEs seems to be influential on the rate in which learners progress through this trajectory. Correlations are found between a more frequent use of these expressions at the initial state and a greater knowledge of related underlying computational properties longitudinally. Traceback analysis also reveals that the FEs have instantiated utterance schema extraction and generalisation across similar functional structures. I discuss the interplay of these concepts by drawing on processing models of SLA, suggesting that schematic learning could provide learners with more syntactic and morphological distributional evidence needed for L2 syntactic activation levels to better compete with the existing L1 ones during processing.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Sheffield
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Hammond, Thomas
- Advisor dc:contributor.advisor
-
- Gil, Kook-Hee
Rights
- Language dc:language.dcterms_iso639_2
- eng