Languages are considered inherently rhythmic, belonging to one of the following rhythmic classes: stress-, syllable- or mora-timed. Interestingly, this assumption corresponds well to patterns of infant speech perception and adult speech processing. Yet, accounting for the shared experiences of linguistic rhythm by language users and validating the rhythm class hypothesis has been proven extremely difficult, and to-date, several decades after its first introduction as a term, linguistic rhythm remains elusive and unsupported by quantitative data. Here, we use acoustic data from a wide range of languages to examine the hypothesis that linguistic rhythm results from the intersections of a language’s prosodic, morphological and phonological structures.

The list of languages included in the analysis to-date includes: 

A’ingae (South America), British English (Eurasia), Castilian Spanish (Eurasia), Estonian (Eurosia), Central Arrernte (Australia), Central Sama (Oceania), Check (Eurasia), Chukchansi Yokuts (North America), Cwyzhy Abkhaz (Eurasia), Gitskan (North America), Hebrew (Oriental and Non-Oriental; Eurasia), Hungaria (Eurasia), Japanese (Eurasia), Japhug (Eurasia), Kalasha, Kedayan (Pacific), Malagasy (Africa), Mapudungun (Isla Huapi; South America), Modern Greek (Eurasia), Munji (Eurasia), Nen (Pacific), Nivaĉle (South America), Spokane (North America), Northwest Sahaptin (North America), Northern Tepehuan (North America), Salasaca Quichua (South America), Seri (North America), Shiwilu (South America), Tamambo (Oceania), Thai (Eurasia), Tausug (Oceania), Yine (South America), Zwara Berber (Africa).