Using corpus linguistics in formulaic language learning

Abstract

This chapter leverages the IdiomsTube project to illustrate how corpus linguistics enhances research and tool development for formulaic language acquisition. Formulaic language, encompassing idioms, proverbs, and sayings, is common in everyday communication. However, English as a foreign language (EFL) learners often struggle with these conventionalised expressions due to limited exposure to authentic spoken contexts. To address this challenge, the IdiomsTube project conducted corpus studies to uncover patterns in formulaic language use, including prosodic features and distribution across internet television genres. Corpus linguistic methods have also enabled the development of the IdiomsTube app, a specialised tool for computer-assisted formulaic language learning. Informed by corpus-derived frequency data and innovative concordancer design, the app uniquely prioritises user experience. Unlike conventional concordancers, the IdiomsTube app dynamically compiles a corpus from captions retrieved in real-time from YouTube videos based on the user’s search word, allowing users to read concordance lines from current, trending videos. This design makes concordancing engaging and motivating for learners. This chapter demonstrates how modernising concordancer designs with a focus on learner accessibility and real-time content can significantly advance formulaic language acquisition.