Dawn Knight

Dawn Knight’s research interests lie in the areas of corpus linguistics, discourse analysis, digital interaction and non-verbal communication. The main contribution of her work has been to pioneer the development of a new research area in applied linguistics: multimodal corpus-based discourse analysis. This has included the introduction of a novel methodological approach to the analysis of the relationships between language and gesture-in-use based on large-scale real-life corpora. She also has an interest in examining the use of wearable technologies as a means of capturing language, gesture and embodied actions in naturally occurring interaction (‘in the wild’) and on developing methodological and technical frameworks for crowdsourcing data collection for corpus compilation.

Since March 2016 Dawn has been leading and managing the £1.8m Economic and Social Research Council (ESRC) and Arts and Humanities Research Council (AHRC)-funded CorCenCC project (Corpws Cenedlaethol Cymraeg Cyfoes – National Corpus of Contemporary Welsh). CorCenCC is breaking new ground, in that it is creating the first large-scale corpus of Welsh and using pioneering community-driven methods, and in that pedagogical corpus tools are part of CorCenCC’s design from the very outset.

Dawn has produced 26 solo/joint authored peer-reviewed articles, chapters and proceedings in internationally recognised publications and is currently co-editing the Handbook of English Language and Digital Humanities (Routledge).

Abstract

Multimodal Corpus Linguistics: Looking back and thinking forward

A priority for researchers investigating natural language and gesture-in-use is to develop better approaches for constructing multimodal records of real-life communication that are not only naturalistic but accurate, detailed and high quality. This presentation reports on some of the methodological, practical and ethical challenges faced in the creation and use of various forms of emergent multi-modal corpora.
The presentation will take stock of previous developments and the current state-of-the-art in the field of multimodal corpus linguistic research and will reflect on possible future directions for work of this nature. First, a discussion of some of the optimum ways in which corpora from a range of different contexts might be recorded, processed, stored, accessed and analysed by the user will be provided. Then, attention will be paid to the ways in which results from such analyses can be utilised and applied to a range of different contexts. Specific attention will be paid to how multimodal corpus work can contribute to/enhance multimodal literacy.

The presentation draws on a range of case studies including the:

• Analysis of spoken and non-verbal backchannel behaviour, gaze and deictic in multimodal corpora.
• Use of wearable sensors in multimodal corpus development (as a means of recording audio tracks and accurate, detailed graphical representations of hand movements as they are performed ‘in the wild’).
• Crowdsourcing linguistic data collection and coding.