Limits of variability in complexity of valency class systems

From May 2023, Sergey Say and Ilja Seržant work on the project “Limits of variability in complexity of valency class systems”. This project is planned for 2023-2025 and is envisaged as a preparatory stage for the application for the third phase of CRC (Collaborative Research Centre) 1287 “Limits of Variability in Language: Cognitive, Computational and Grammatical Aspects”.

The germ of the project is the idea that languages greatly vary in terms of the complexity of their valency class systems (ranging from simple systems with just a few valency classes to super-complex systems with dozens of classes), but that any actual valency class system probes a point of equilibrium between efficiency in learning and lexicon composition vs. maximal informativeness. As a consequence, the observed variation in the complexity of valency class systems, which can be operationalized in terms of e.g. entropy as observed in the lexicon (type frequency) and/or in discourse (token frequencies), is expected to be constrained by factors related to learnability and communicative efficiency and to interact with typological parameters such as word order and referential density. In particular, the goal of the project is to test the hypothesis that languages that put more weight on their argument-coding devices employ fewer informative cues in other parts of their grammar that serve linking referents to argument roles in discourse.