Prompt literacy in english language teaching: rethinking communicative competence in the age of generative artificial intelligence

Abstract

Generative artificial intelligence entered higher education English language teaching (ELT) faster than the field could theorize it, and the scholarly vocabulary available for describing what learners do when they write to a generative AI system has largely been borrowed from human-computer interaction, educational policy, and general AIliteracy research rather than built for the specific case of a second-language (L2) English user addressing a generative AI interlocutor in natural language. This chapter addresses that gap through a theoretical and conceptual analysis. Building on communicative competence theory (Hymes, 1972; Canale & Swain, 1980; Bachman & Palmer, 1996) and on the emerging literature that has begun to name and describe prompt literacy for language learners specifically – most directly Hwang et al. (2023) and Tour and Zadorozhnyy (2025) – the chapter argues that prompt-related competence for L2 English users is best theorized not as a subtype of general AI literacy or of prompt engineering, but as a linguistically grounded form of L2 communicative competence. It distinguishes this reading from Digital Literacy, Digital Competence, AI Literacy, and Prompt Engineering; proposes a four-dimension taxonomy (linguisticformational, pragmatic-interactional, metacognitive-strategic, critical-epistemic); integrates seven learning-theoretic traditions into that taxonomy while preserving a genuine, unresolved tension between cognitive load theory and learner autonomy theory; and develops a conceptually derived developmental continuum, an application across the skill areas of ELT, and a proposed assessment rubric grounded in validity theory (Messick, 1995). The central contribution is theoretical: a reconceptualization of prompt literacy, in continuity with rather than in place of the term's originators, as an applied-linguistic construct. The taxonomy, developmental continuum, and rubric are offered as theoretically derived, conceptually testable proposals rather than as validated instruments. The chapter is a theoretical and conceptual study; it does not report original empirical data collection, and it closes by naming the empirical work – piloting, inter-rater reliability testing, and validation research – that this proposal now requires. Keywords: prompt literacy; L2 communicative competence; generative AI; English language teaching; AI literacy; applied linguistics; assessment validity; higher education.