Translanguaging and data-driven learning: How corpora can help leverage learners’ multilingual repertoires

Abstract

Translanguaging, i.e., the use of multiple languages to make and negotiate meaning, has been shown to be beneficial for language learning (see, e.g., García & Kleifgen, 2020). Although it is a fairly natural and spontaneous phenomenon in the lives of many multilingual speakers, its role is not well established in the language classroom, where the use of learners’ mother tongues (L1s) or some additional languages (L2s) besides the target language (TL) is often frowned upon (by school authorities, teachers, and even students themselves). For this reason, there have been calls to “explore what ‘teachable’ pedagogic resources are available in flexible, concurrent approaches to learning and teaching languages bilingually” (Creese & Blackledge, 2010, p. 113). One such possible resource is the corpus, an electronic database of naturally-occurring language that can be investigated by means of special tools and techniques to gain insights into the language(s) or language variety (varieties) represented in the corpus. What is particularly interesting about corpora from a pedagogical perspective is that they can be used by students to make their own discoveries about language, a pedagogical approach known as ‘data-driven learning’, or DDL for short (see Gilquin & Granger, 2022). In what follows, I review some of the corpora that could be useful in translanguaging pedagogy and briefly show how they could help language learners leverage their own multilingual repertoires.