Language acquisition has often been approached as an isolated domain subject to its own laws and driven by its own mechanisms. The working hypothesis of this book is that language acquisition is simply learning, subject to the same laws as learning in other domains, and well-described by models of associative learning. ‘Changing minds changing tools’ connects findings in language acquisition to comparable findings elsewhere, including research on error-driven predictive learning, Hebbian learning, chunking, category learning, and learned selective attention. In the process, ‘Changing minds cha ... More

Language acquisition has often been approached as an isolated domain subject to its own laws and driven by its own mechanisms. The working hypothesis of this book is that language acquisition is simply learning, subject to the same laws as learning in other domains, and well-described by models of associative learning. ‘Changing minds changing tools’ connects findings in language acquisition to comparable findings elsewhere, including research on error-driven predictive learning, Hebbian learning, chunking, category learning, and learned selective attention. In the process, ‘Changing minds changing tools’ provides a domain-general associationist framework for some of the central issues in language acquisition, from phonetics to phonology to morphology to the lexicon / constructicon. This perspective is argued to provide plausible explanations for several recurrent pathways of language change. From sound change to changes in construction productivity to grammaticalization, languages change in predictable ways. Some directions of change are frequent while others are rare or unattested. Following the usage-based approach to linguistics, ‘Changing minds changing tools’ argues that explaining these “diachronic universals” is the most promising way to approach the central question of linguistic theory, “why languages are the way they are.” Synchronically, languages are incredibly diverse but they change in predictable ways. Once applied to the task of language acquisition, domain-general learning mechanisms provide ready explanations for many diachronic universals. Therefore, approaching language acquisition should bring the field closer to its ultimate goal of explaining both what languages share and the ways in which they vary.