A Case of Tense Acquisition: Is 90% Enough?

Abstract

This paper is a case study to analyze divergences in the L2 English past tense that a Korean learner of English has produced. The learner supplies appropriate past tense verbs in about 90% of their obligatory contexts, which strongly suggests that he has acquired the past tense morpheme. The present study, however, identifies the following aspects of systematicity in the 10% residue. First, the divergences mostly occur in embedded clauses. Second, most divergences occurring in matrix contexts can be related to a past tense morpheme nearby. We propose that this systematic divergence mainly reflects an English-Korean difference in the tense system: Unlike in English, the [-past] tense can be interpreted as denoting a time or interval in the past in embedded clauses in Korean (Han 1996, Lee 2001). Further, We also note that some cases of divergence are not attested in either of the learner's L1 and L2. We suggest that the "wild grammar" feature can arguably be derived from an interaction of a calcitrant macro-parameter "efficiency-oriented" that Korean has and the relevant input from the target language, and be compatible with the thesis that acquisition of EFL is constrained by Universal Grammar.