Complex networks reveal a glottochronological classification of natural languages

Abstract

The success of humans cannot be attributed to language, but it is certainly true that language and modern humans are inseparable. This work focuses on revealing the structure of 20 Indo-European languages belonging to three sub-families (Romance, Germanic, and Slavic) from a chronological perspective. In order to find the chronological characteristic features of these languages, we use (1) Heaps’ law, which describes the growth of vocabulary (distinct words) in a corpora for each language to the total number of words in the same corpora and (2) structural properties of networks created from word co-occurrence in corpora of 20 written languages. Using clustering approaches and entanglement, we show that in spite of differences from years of being used separately and differences in alphabets, one can find language characteristics that lead to cluster of languages resembling the organization according to historical sub-families and chronological relations.