Nick C. Ellis, Ute Römer, and Matthew Brook O'Donnell present a view of language as a complex adaptive system that is learned through usage. In a series of research studies, they analyze Verb-Argument Constructions (VACs) in first and second language learning, processing, and use. Drawing on diverse epistemological and methodological perspectives, they show how language emerges out of multiple experiences of meaning-making. In the development of both mother tongue and additional languages, each usage experience affects construction knowledge following general principles of learning relating to frequency, contingency, and semantic prototypicality. The implications of this work will be of value to students and scholars from a wide range of disciplinary interests in language and learning.
"This is an impressive volume that will inspire researchers for generations to come. Focusing on the construction and acquisition of language, it combines a comprehensive synthesis of theory with a detailed account of extensive empirical work."
--Susan Hunston, University of Birmingham
"This book is a phenomenal synthesis of a formidable research program. In a feast of corpus, psycholinguistic, acquisitional, and simulation evidence, the authors’ bold theoretical insights advance knowledge about human language to unprecedented levels."
--Lourdes Ortega, Georgetown University
"The authors present a superb synthesis of approaches to verb-argument constructions and convincingly demonstrate the close links between lexical patterning and constructional meaning. An absolute must-read for anyone interested in usage-based approaches to language learning."
--Ewa Dabrowska, University of Northumbria at Newcastle
"This book represents an outstanding achievement. The authors illustrate why the most exciting work in the language sciences today is conducted across disciplinary boundaries. Working at the intersection of experimental, computational, and corpus-based approaches, their research inspires us to look beyond our own disciplines to observe language data from all angles."
--Patrick Rebuschat, Lancaster University
