Nodes and Networks in Diachronic Construction Grammar: (27 Constructional Approaches to Language)

This volume brings together ten contributions by leading experts who present their current usage-based research in Diachronic Construction Grammar. All papers contribute to the discussion of how to conceptualize constructional networks best and how to model changes in the constructicon, as for example node creation or loss, node-external reconfiguration of the network or in/decrease in productivity and schematicity. The authors discuss the theoretical status of allostructions, homostructions, constructional families and constructional paradigms. The terminological distinction between constructionalization and constructional change is revisited. It is shown how constructional competition but also general cognitive abilities like analogical thinking and schematization relate to the structure and reorganization of the constructional network. Most contributions focus on the nature of vertical and horizontal links. Finally, contributions to the volume also discuss how existing network models should be enriched or reconceptualized in order to integrate theoretical, psychological and neurological aspects missing so far.

Review :
The volume has provided new insights into the modeling of constructional networks and is an important contribution to DCxG. Furthermore, it is a very stimulating and thought-provoking book that challenges the reader to think about how to best model the constructional network. As such, it can be expected to feed future work in (Diachronic) Construction Grammar. The volume is an innovative and thought-provoking contribution to the CxG/DCxG community. Not only does it raise open questions and highlight further directions for further research, but also offers a consistent terminological system to avoid ambiguities in the DCxG field. It also gathers state-of-the-art work on DCxG into one place, which may be especially useful for readers who do not readily have access to earlier publications. The present volume brings together an exciting range of proposals on how a dynamic network model of language can contribute to the analysis of diachronic change. The corpus methods used by the authors include state-of-the art techniques like collostructional analysis, distributional semantics and even more advanced computational tools like artificial neural networks, which have yet to become more widely applied in (historical) linguistic research. The contributions illustrate not only how a cognitively oriented network perspective can provide diachronic scholars with a new conceptual framework in which constructional change can be modelled as are configuration of linking patterns between nodes, but also how the careful analysis of language change can in turn inform network models which have so far been largely posited based on synchronic observations. The volume thus provides strong evidence that historical corpus data can complement psycholinguistic experiments in assessing the psychological plausibility of network structures, and the way in which these are shaped by speakers’ general cognitive abilities such as analogical reasoning.