PhD in Second Language Studies Dissertation Defense
Idioms in the bilingual mental lexicon
Chair: Thom Hudson
Friday April 29, 12:00 p.m.–2:00 p.m.
Moore Hall, Room 155A
This study investigates how native and non-native speakers of English understand English idioms by investigating how context and other factors affect idiom processing and how idioms are represented in speakers’ mental lexicons. The study utilizes Abel’s (2003) dual idiom representation model (DIRM), which differs from previous idiom processing models in that it is based on second language (L2) data as well as first language (L1) data.
In order to address the questions of how L1 and L2 speakers process and conceive of idioms, the study conducted an experiment in which L1 English speakers and L2 English speakers completed a self-paced reading (SPR) task. Both groups also completed a semantic decomposability survey and an idiom familiarity survey, and the L2 group additionally completed a metaphoric equivalence survey.
The participants were 36 native speakers of Chinese and 40 native speakers of English, all either undergraduate or graduate students at the University of Hawai‘i at Mānoa (UHM). For the study materials, 24 English idioms from a list of idioms utilized by Abel (2003) were selected on the basis of word frequency, currency, degree of familiarity, semantic decomposability, and metaphoric equivalence. The SPR task used all of these idioms, and included 48 passages (12 figurative contexts, 12 neutral contexts, and 24 fillers), which participants read on a computer screen. Their reaction times (RT) were recorded by the computer. The surveys asked the participants to rate the 24 idioms in terms of familiarity, semantic decomposability, and metaphoric equivalence (L2 group only) on a 5-point Likert scale.
The research provides several important results. First, both the L1 and the L2 language users processed the idioms faster in figurative contexts than in neutral contexts. The dissertation discusses this finding in terms of the direct access view (Gibbs, 1994) and the modular view (Fodor, 1983) of idiom processing. Second, the L1 language users processed the idioms faster than the L2 language users, a difference related to degree of familiarity. Third, both the L1 and the L2 language users tended to conceive of the idioms as semantically decomposable rather than semantically nondecomposable, and this tendency was stronger in the L1 group; this finding differs from previous studies’ reports. Drawing on the DIRM, the dissertation provides possible reasons for this difference. The analysis of the L1 group’s idiom processing found an interaction effect between familiarity and semantic decomposability: As the degrees of familiarity and semantic decomposability increased, the RTs in the figurative contexts decreased greatly, but the RTs in the neutral contexts did not. In the L2 participants’ idiom processing, metaphoric equivalence was a significant factor: As the degree of metaphoric equivalence between L1 and L2 increased, the RTs in both context types decreased equally.
This study contributes to research on second language acquisition and idiom processing by using the DIRM to investigate both L1 and L2 language users’ idiom comprehension. In addition, it provides insights into how to select idioms and create reliable rating surveys for idiom research. Furthermore, its results offer L2 language teachers some pedagogical implications by providing a better understanding of how L2 learners conceive of and process L2 idioms.