Native language identification

The main objective of this chapter is to examine the automatic techniques that can be applied to learner corpora to identify learners’ mother tongue backgrounds from their patterns of production in the target language. The chapter focuses particularly on aspects of learner corpus design that have a direct bearing on the results (size, topic homogeneity, type and level of annotation). Special attention is also paid to the contribution of NLI research to SLA, more particularly in the form of the detection-based approach to transfer. NLI is a relatively new line of enquiry: research so far has focused on identifying the native language of writers using English as a foreign language and the chapter will thus deal with English as a target language