Diversity as a Core Feature of Language Acquisition: A Commentary on Scaff et al. (2025)

This commentary builds on Scaff et al.'s (2025) systematic review of the CHILDES database, highlighting persistent biases in child language corpora and research. We expand the discussion, emphasizing three key areas: (1) the need to diversify naturalistic data across languages to strengthen language acquisition theories; (2) the importance of including diverse child and parent demographics within specific language environments; and (3) the underrepresentation of bilingual samples from non-WEIRD, non-Indo-European contexts. We argue that these limitations not only hinder generalizability but also shape prevalent theoretical assumptions. Promoting inclusive, globally representative corpora is important for advancing a fair and accurate understanding of child language acquisition. SUMMARY: Diversification of naturalistic data across languages strengthens language acquisition theories. Child and parent characteristics within specific language environments should be included in child language research. Bilingual samples in CHILDES corpora should be evaluated on their generalizability.