Fluency and disfluency : a corpus study of non-native and native speaker (dis)fluency profiles
(2018)
Files
AmandineDumont2018Fluencyanddisfluency.pdf
Open Access - Adobe PDF
- 12.26 MB
Details
- Authors
- Supervisors
- Abstract
- The fluency of a spoken discourse in a first or foreign language depends on numerous factors such as speech rate, the use of filled and unfilled pauses, repairs, or discourse markers. While previous research has mainly focused on the analysis of specific features, to date, little is known about the interrelationships between these factors or the possible variations between speakers. This thesis aims to get better insights into the notion of (dis)fluency by comparing and contrasting the spoken language of French-speaking learners of English and that of native speakers of English. The study is based on two corpora of transcribed interviews (c. 30 hours of speech in total). After having identified the elements contributing to (dis)fluency in the two spoken corpora, this study examined these (dis)fluency features quantitatively, first by contrasting learners and native speakers, then by focusing more specifically on two groups of learners at different levels (B2 and C1). The results of this study revealed that (dis)fluency is a complex phenomenon that includes different underlying dimensions. These subtly vary between the group of learners and the group of native speakers – a case in point is the use of filled pauses. Moreover, this study has shown that a fluent discourse in both a first and foreign language is absolutely not without hesitations and can take very different forms. Several learner and native "(dis)fluency profiles" have been delineated. It is noteworthy that the learner profiles do not correspond to the levels of the Common European Framework of Reference (CEFR), which calls into question the use of this type of grid for the assessment of learner fluency.
- Affiliations
Citations
Dumont, A. (2018). Fluency and disfluency : a corpus study of non-native and native speaker (dis)fluency profiles. https://hdl.handle.net/2078.5/97052