ABSTRACT
This chapter deals with how researchers and practitioners have exploited written learner corpora to inform language teaching. It considers both direct and indirect uses of learner corpora, with a focus on their contribution to dictionaries, grammars and textbooks, as well as data-driven learning. It is shown that the analysis of error-tagged learner corpora or the comparison of learner corpora with native corpora can highlight aspects of learner language which deserve pedagogical attention and that the metadata of learner corpora can help with the customisation of teaching resources. The chapter also highlights certain issues that may have prevented the wider use of learner corpora for pedagogical purposes and it offers some recommendations for practice, including more transparency with regard to information derived from learner corpora. Finally, it points to avenues for further developments, in particular, with respect to computer-assisted language learning applications and the use of process learner corpora, which represent learners’ writing process through the addition of keystroke logs and/or screencast videos.
