VARIABLE PRODUCTION OF ENGLISH PAST TENSE MARKING BY L1 THAI LEARNERS: AN APPLICATION OF THE FAILED FUNCTIONAL FEATURES HYPOTHESIS

Abstract

In second language acquisition research, it is well-attested that production of functional morphology, e.g. tense and agreement, by L2 learners is variable. English past tense marking is a functional morphology that is frequently variably produced by L2 learners (e.g. Bayley, 1991; Lardiere, 1998; Tajika, 1999; Hawkins and Liszka, 2003; among others). The present study examined whether variability exists in production of English past tense marking by L1 Thai speakers. It was hypothesized that variable use of English past tense marking will be observed and that the phenomenon can be accounted for by the Failed Functional Features Hypothesis (FFFH) (e.g. Franceschina, 2001; Hawkins and Chan, 1997; Hawkins and Liszka, 2003), but not by the Missing Surface Inflection Hypothesis (MSIH) (e.g. Lardiere, 1998; Prévost and White, 2000; White, 2003). Three tests: Grammaticality Judgment Test (GJT), Cloze Test, and Story-telling, were administered to 40 L1 Thai learners: 20 intermediate and 20 advanced learners. All of the participants were freshmen at Chulalongkorn University. The results showed that L1 Thai speakers exhibited variability in their production of English past tense marking in both the representation test: the GJT, and the production tests: the Cloze Test and the story-telling test. The two L1 Thai proficiency groups displayed a low suppliance rate of English past tense marking across the three tests. Additionally, an asymmetric rate of suppliance of past tense marking was observed. It was found that regular verbs were past-marked less frequently than irregular verbs by both proficiency groups. The suppliance rate of English past tense marking by the two L1 Thai proficiency groups was also higher when adverbial phrases of time indicating pastness were present. The low suppliances of past morphemes including both the representation and the production tasks, and the asymmetric phenomena confirmed the two hypotheses, hence, supporting the FFFH but confounding the MSIH.