TyDEE linguistic dataset

TyDEE (Typological Dataset of Eastern Eurasia) is a cross-family linguistic typological database covering 574 language varieties across 11 language families in Eastern Eurasia — Turkic, Mongolic, Tungusic, Sinitic, Tibeto-Burman, Tai-Kadai, Hmong-Mien, Austroasiatic, Austronesian, Japonic, and Koreanic — plus 15 Indo-European varieties as an outgroup reference. Each variety is coded for 52 binary typological features spanning phonology, phonotactics, semantics, colexification, and morphosyntax, with no missing values. The database builds on earlier work on typological variation and convergence in North and Southeast Asia (Szeto, Ansaldo & Matthews, 2018; Yurayong & Szeto, 2020) and has been substantially expanded in geographic and genealogical coverage. Language sampling ensures representation of all major sub-branches within each family, following the variety sampling method (Miestamo et al., 2016). Feature selection prioritizes traits with meaningful cross-family variation in Eastern Eurasia and documented relevance to contact-induced change, with each feature supported by typological and areal-linguistic literature. All coding was carried out collaboratively with linguistic specialists based on reference grammars and grammatical descriptions.