ABSTRACT
This chapter traces the historical development of corpora and the shift from written corpora to the creation of spoken corpora. Initially, corpora were not seen as essential to materials development, and debate took place as to their suitability, especially regarding reliance on native-speaker data. Meanwhile, the same period saw the emergence of learner corpora. In the exploitation of corpora in materials, approaches have varied from strictly data-driven to more indirectly corpus-informed, where insights gained from corpora are more subtly reflected in the material. Problems with corpus-informed materials include the fact that corpus research often challenges the ELT canon and that spoken corpora often display forms considered incorrect. The translation of corpus insights into materials is complex and involves a methodological shift alongside changes in content. The creation of corpus-informed material that is non-threatening and transparent to teachers of all levels of experience is advocated, and consideration is given to the future relationship between corpora and learning technology.
