About the semantic and grammatical features of text 9A2-I, ...

          The words “theory” and “theoretical” evoke a variety of responses in language teaching and research circles, many of them, for different reasons, negative. For some, the negative response they feel is due to their having sat through conference presentations or read journal articles labelled “theoretical” which have consisted of a good deal of rhetoric, however eloquent, and very little substance. “Theoretical” here is mis-used, and just means “data-free”. For others, the particular theories that have received most “air-time” in the second language acquisition (SLA) literature until now have been uninteresting, wrong, or vacuous, leading them to be potentially hostile to any new ones. Still, others have no problem with theory in general (or think they don’t), but simply feel that work in SLA has not advanced far enough yet for theorizing to be productive.

         Those who subscribe to the last view — and they include several prominent figures in SLA — hold that because, in their opinion, we know relatively little about SLA, any theory we come up with at this stage is likely to be wrong. Hence, it will be counter-productive, in that many people will waste their time working on a theoretical red herring instead of discovering more facts about acquisition. In our view, while superficially reasonable, this shows that the purpose and value of theories in (social) science are still not widely understood in our field.

Diane Larsen-Freeman and Michal H. Long. An introduction to second language acquisition research. New York: Longman, 1991.