New evidence for a Mesoamerican homeland for Proto-Uto-Aztecan

Merrill et al. (1) find “untenable” the hypothesis that speakers of Proto-Uto-Aztecan (PUA) were maize cultivators in or near Mesoamerica. Linguistic evidence supporting this hypothesis was first assembled in ref. 2 and additional evidence is cited here. Of central importance to the proposal is that a word for maize, **sunu, reconstructs for PUA. Although reflexes of the latter word were not reported in ref. 2 for Uto-Aztecan languages of California, one has now been identified. This is Gabrielino şoŋ-áxe-y “tortilla” (literally, maize-put.in.mouth-nominalizing.suffix’). Two other maize-complex terms found in UA California languages are traced to PUA words: Luiseño şa:xi-š “grain, wheat” from PUA **saki “parched corn,” and Tübatulabal paca:hil “hulled pine nuts” from PUA **pa7ca “maize kernel.”

The Gabrielino word, squarely in the maize domain, is difficult to dismiss as a late loan. It displays the regular Northern UA sound change from PUA **n to *ŋ. Gabrielino was spoken by hunter-gatherers in the Los Angeles Basin and the southern Channel Islands. The closest possible sources are Hopi soŋowï Calamovilfa gigantea, not in the maize domain (2), and O'odham hu:ñ “corn,” reflecting a Proto-Tepiman sound change of PUA **s to *h. The semantic reconstruction for ** pa7ca, challenged in ref. 3, has been strengthened by the identification of maize-domain cognates in Opata and in Classical Nahuatl.

Several word roots in the domain of “pottery” can be reconstructed for PUA, also with reflexes in California UA languages. Some of the PUA ceramics vocabulary is probably of Otomanguean origin, and this is probably true of the PUA maize vocabulary as well (4). Otomanguean-speaking peoples are not historically attested north of San Luis Potosí, Mexico.

Details of the processes by which maize spread from Mesoamerica to the US Southwest cannot be fleshed out by historical linguistic methods alone. However, those methods do robustly suggest that diffusion was involved at least twice, from some Otomanguean language or languages into PUA (4), and from Proto-Northern UA into Proto-Kiowa-Tanoan (5). If maize was borrowed along a Uto-Aztecan dialect continuum of communities in western Mexico, as suggested in ref. 1, this diffusion occurred at such an early date that historical linguistic method cannot distinguish it from the spread of maize cultivated in the PUA community to the US Southwest through migration.