On the specificities of L1 and L2 (dis)fluencies and the interactional multimodal strategies of L2 speakers in tandem interactions

‘Disfluencies’, defined as a suspension of speech, are usually more frequent in L2 than in L1, which is said to be related to the learners’ proficiency level. ‘Disfluency’ or ‘fluency’ have thus often been associated with language proficiency and temporal characteristics. Grounded in a functionally ambivalent view of (dis)fluency captured in situated multimodal discourse, our research program aims to measure the degree of (dis)fluency found in tandem interactions by comparing their specificities in L1 and L2 productions of French and American English. In this paper, we stress out the need to regard (dis) fluency as a multimodal and multilevel phenomenon, which takes into account vocal and visual-gestural modalities of discourse. While our quantitative results indicate a higher rate of complex (dis)fluencies in L2 than in L1, which supports previous studies, and a higher gestural activity in L2, the qualitative analyses conducted on the data illustrate their multimodal and interactional dimension. This paper presents new methods to evaluate the degree of (dis)fluency, by combining quantitative and qualitative methods, based on an interactional scale, which takes into account the multimodal communication strategies of L2 speakers. We thus argue that (dis)fluency is not only the result of communication breakdowns as it is highly contextualized. It should thus not only be measured with temporal variables and overall frequency rates, but also with regards to the individual speaker’s use in multimodal interactional and intersubjective contexts.