Bivalent verb classes across Slavic: areal and genealogical patterns

Abstract The Slavic languages, shaped by a complex history of divergence and contact, offer an ideal setting for investigating how genealogical and areal factors influence valency class systems. These systems, marked by diachronic instability, reflect both internal linguistic developments and external pressures. Drawing on data from BivalTyp, a typological database of bivalent verbs and their coding frames ( https://bivaltyp.info/ ), this study analyzes the argument-coding properties of 130 bivalent predicates (e.g., ‘hit’, ‘know’, ‘be afraid’, ‘look’, ‘resemble’) across 11 Slavic standard varieties: Belarusian, Bulgarian, Croatian, Czech, Macedonian, Polish, Russian, Serbian, Slovak, Slovenian, and Ukrainian. The dataset is annotated for argument-coding devices, such as cases and adpositions, as well as for cognacy relations between verbs and argument-coding devices. These annotations serve as the basis for constructing distance matrices that reveal patterns of similarity and divergence among Slavic languages in various subsystems of argument coding, with syntactic dimensions partially independent of the cognacy relations of the verbs involved. The findings highlight the significant role of areal factors in shaping valency classes in Slavic languages, often surpassing genealogical influences. This is evident in the systematic convergence between Polish and East Slavic languages and in broader areal trends, where Slavic languages exhibit syntactic convergences with neighboring non-Slavic languages.