The Conceptualization and Integration of an E-Collocation Trainer

Collocations, broadly defined as affinity-bound, recurrent lexical chunks, represent one of the primary domains reflecting differences in language usage between native and non-native speakers. Their inherent sense of illusiveness, stemming primarily from tendencies to deviate from any quantitatively-sound lexical or syntactic explanatory parameters, has proven to be a stumbling block for L2 language users. This difficulty is exacerbated by the fact that collocations cannot be internalized in a second language following a concrete, linear, systematic methodology.

Along these lines, this research proposes and outlines an e-collocation trainer, E-COR (Electronic Collocation Organization and Retrieval), a dynamic, individualized database adhering to the objectives of the translator working with English as a second language. In line with a CALL-based framework, user-friendly corpus linguistics guidelines and incidental learning theories, this research addresses existing shortcomings in terms of L2 collocation resources at the user's disposal, and exemplifies how naturally-occurring collocations and collocational data can be obtained, verified, stored and ultimately retrieved at multiple levels. Collocation verification is foregrounded as most prevalent in terms of meeting the translator's immediate, task-based needs, yet channels for pursuing collocation production are also integrated as a way of enhancing longer-term collocational competence.