Past Event

On the importance of register in learner corpus research: A multi-dimensional approach

Dr. Tove Larsson
Northern Arizona University / UCLouvain
(in collaboration with Magali Paquot and Doug Biber)

The question of which features of EFL learner writing are shared rather than first-language (L1) specific has received a great deal of attention in learner corpus research (LCR) in recent years. High involvement (e.g. frequent use of questions and first-person pronouns) is an oft-cited example of one such shared feature (e.g. Petch-Tyson, 1998). However, the reliability and generalizability of such claims about “learner language” have since been called into question, as they are almost exclusively made based on argumentative learner writing (e.g. Larsson & Kaatari, 2019). Instead, it seems that we need to recognize the existence of register variability in learner language too.

Paquot & Biber (2015) used Biber’s (1988) dimensions in a multidimensional (MD) analysis to compare learner (L1 French and Norwegian) and native-speaker student writing from two different registers (argumentative texts vs. academic prose) to published expert writing. However, in using dimensions that were built on corpora of NS English, the study did not take advantage of the richness of the learner data in its own right (Hardy & Römer, 2013). In the present study, therefore, we develop and extend previous research in the following ways: first, we use native-speaker and learner corpora to identify frequently co-occurring features and build corresponding dimensions; second, we include a larger set of L1s and more native-speaker material to attain a more complete picture of learner writing. The data used come from two learner corpora (ICLE and VESPA), three native-speaker-student corpora (LOCNESS, BAWE and MICUSP), and one expert corpus (LOCRA). The results of the study testify that it is imperative to include register as a predictor in LCR studies.

About the Presenter

Dr. Tove Larsson is currently a visiting researcher at Northern Arizona University. She is also affiliated with the Centre for English Corpus Linguistics at Université catholique de Louvain. Her research centers primarily around corpus-based studies of learner English with a particular focus on second-language writing, lexico-grammar and register variation. She also has a keen interest in research methodology.

This event is sponsored by the School of Linguistics and Language Studies.