The Genetic Unity of Southern Uto-Aztecan

Abstract

The internal structure of the Uto-Aztecan language family has been debated since the late 19th century, when the historical relationships among all of its major subdivisions were first recognized. Alexis Manaster Ramer’s identification in 1992 of a phonological innovation shared by languages belonging to the four northernmost subfamilies led to the acceptance of these languages as a genetic linguistic unit called Northern Uto-Aztecan, but no consensus has emerged regarding the organization into higher-level subgroups of the remaining five subfamilies. In this essay, I argue in support of a perspective, originally developed by Terrence Kaufman, that the languages in these subfamilies also constitute a genetic unit, Southern Uto-Aztecan, based on two shared, sequential innovations: *-n- > *-r- and *-ŋ- > *-n-. Key to my argument is the reconstruction of a Proto-Uto-Aztecan liquid phoneme with **[-r-] and **[-l-] as its allophones, which clarifies the diachronic relationships among reflexes of **-n-, **-ŋ-, and **-r- in the daughter languages. The model that I propose offers a parsimonious solution to several perennial issues in Uto-Aztecan historical phonology and a possible explanation for the absence of a liquid phoneme in the Numic languages.