Research Training Group “Dimensions of constructional space”

The Research Training Group will explore a relatively new paradigm in linguistics that has become known under the name of Construction Grammar (CxG). The fundamental premise of this approach is that the totality of speakers’ linguistic knowledge is represented in a network of form-meaning pairings called constructions. Constructions differ in size (ranging from morphemes through phrasal and clausal templates to discourse conventions), abstractness (fully elaborated, partially schematic, fully schematic) and entrenchment, as well as in the way they interconnect with other constructions in the network. These properties define a multidimensional space we call ‘constructional space’.

The project will address central theoretical questions in CxG (e.g. criteria for identifying constructions and establishing links between them), apply the framework to various languages (including less studied languages such as Persian, Arabic and Haitian Creole) and different historical stages / language contact situations, and test predictions derived from CxG using a variety of methods, including traditional and ‘big data’ corpus methods, behavioural experiments and neuroimaging techniques. In addition, the project will develop an open-access database for the academic community (linguists, psychologists, cognitive neuroscientists, computational linguists, etc.) in the form of a research constructicon that brings together and interconnects constructional descriptions as well as experimental results obtained in the various research projects on individual constructions and particular types of constructions. The project is interdisciplinary in that it combines insights from theoretical linguistics (both synchronic and diachronic), computational linguistics, neuroscience and psycholinguistics. What brings the individual projects together is a shared theoretical framework (CxG), which itself is to be subject of critical analysis, shared research questions and a radical commitment to empirical research.