A sampling technique for worldwide comparisons of language contact scenarios

Abstract Existing sampling methods in language typology strive to control for areal biases typological datasets as a means avoid contact effects the distribution of linguistic structure. However, none these provide ways directly compare scenarios from perspective. This paper addresses this gap by introducing procedure worldwide comparisons scenarios. The unit consists sets three languages. Focus Language is whose structures we examine search effects; Neighbor genealogically unrelated Language, and counts potential source influence on Language; Benchmark relative neither with nor language, used disentangling genealogical inheritance Language. Through design, compiled sample 49 three-language (147 languages total), which present here. By switching focus individual relations between languages, our method has uncovering patterns diffusion structures, how they vary change.