Long-range comparison between genes and languages based on syntactic distances

Objective: To propose a new approach for genetic/linguistic comparison of populations belonging to distantly-related groups. Background: Lexical comparison among languages has been used to approximately reconstruct demographic history in many genetic studies. The models tested assume that linguistic similarities between populations reflect either descent from common ancestors, or the balance between isolation and gene flow. However, precise measures of linguistic divergence based on lexical comparisons are difficult and procedures to neutralize chance similarity are reliable only for closely related languages; hence, analyses of genetic and linguistic diversity at the worldwide scale appeared so far impossible or unreliable. Methods: Syntax is more measurable, universally comparable, and stable than the lexicon (Longobardi and Guardiano 2009), hence certain syntactic similarities might reflect deeper linguistic relationships, such as those between more distant languages and families. In this study, for the first time we compared genetic data to a matrix of linguistic differences between selected populations of three continents inferred from syntactic comparisons. Results: Using a database of 835 microsatellite (STR) markers and 4140 Single Nucleotides Polymorphisms (SNPs), compared to the values of 62 grammatical parameters, we show that there is indeed a correlation of syntactic and genetic distances, we identify a few outliers and we suggest a possible interpretation of the overall pattern. Conclusions: These results strongly support the possibility of better investigating population history by combining genetic data with linguistic information of a new type, building upon a theoretically more sophisticated assessment of the relationships between distantly-related languages and language families.