The use of semantic transparency and L1-L2 congruency as multi-word units selection criteria

Abstract

Multi-word units (MWUs) are defined as two or more words that commonly co-occur. There is evidence that knowledge of high-frequency MWUs is essential to language fluency, leading to growing research identifying valuable MWUs to learn and the impact of L1-L2 congruency and semantic transparency on the learning burden of MWUs. Therefore, there needs to be more research on which MWUs should be selected with these criteria. This article highlights an investigation of the role of congruency and semantic transparency using a corpus-based list that offers a sizable sample of MWUs that appear in general English. In this study, we analysed a list of 11,212 high-frequency MWUs created using a lemmatised concgramming approach to examine the role of semantic transparency and L1-L2 congruency. The list was translated into Persian, and L1-L2 congruency ratings were given to each item. The list was also classified based on Grant and Bauer’s (2004) taxonomy to explore the role of semantic transparency to determine the extent to which these two factors play a role in the learning burden of the MWUs. The results showed that 85% of items were literal, and a low number of opaque items were found in the high L1-L2 congruency rating, suggesting a positive relationship between congruency and transparency. The current research then discusses the implications of these two criteria for teaching English as a second language and considers their importance in designing teaching materials.