The "state of the art" in the three fields of contrastive analysis, error analysis and interlanguage is critically examined from the point of view of evolving an explanatory theory of a second language learner's performance. Each field is discussed with respect to its outreach, theoretical assumptions, methodology, claims and empirical validations thereof, scope of practical studies, and pedagogical implications. It is argued that contrastive analysis, far from being irrelevant or obsolete as some critics have claimed, remains, both theoretically and methodologically, the most principled component of a theory of errors. (Author)