Rethinking Motivation and Epistemic Justice in AI-Mediated Language Learning: A Decolonial Research Agenda for Post-Conflict Indigenous Learners

This concept paper interrogates the intersection of Artificial Intelligence (AI), epistemic equity, and language acquisition within marginalized communities. While contemporary scholarship often characterizes AI-mediated language learning (including automated feedback and large language models) as a democratizing force , this paper argues that such technologies risk reproducing colonial hierarchies by privileging high-resource languages and Western knowledge systems. The author identifies a significant theoretical gap: traditional L2 motivation frameworks—such as the L2 Motivational Self-System and Self-Determination Theory—frequently overlook the sociopolitical and historical dimensions of learning in post-conflict Indigenous contexts. Drawing on Fricker’s (2007) conceptualization of epistemic injustice and decolonial theory , the paper proposes a transformative research agenda. Key themes include: Critiquing the "testimonial" and "hermeneutical" injustices embedded in current AI training data. Reconceptualizing motivation as a political and epistemic phenomenon rather than a purely individual psychological construct. Advocating for the integration of Indigenous knowledge systems into the design of future AI-supported language environments. Ultimately, this work serves as a foundational call to center the epistemic participation of marginalized learners to ensure that the AI transition in education does not reinforce historical exclusion.