Oral and nasal airflow in nasalized laryngeal consonants

While it has long been known that laryngeal consonants do not block nasal harmony, it has been debated whether the velum actually remains lowered during the production of glottals in nasal harmony contexts. Using newly collected nasal and oral airflow data from two Amazonian languages, Maihki (Tukanoan, Peru) and Piaroa (Jodi¨-Sáliban, Venezuela and Colombia), we demonstrate that the laryngeals [h, ʔ] are, in fact, produced as nasalized [h̃ , ̃ ʔ] in these languages when they occur in nasal harmony spans. We observe that nasal flow is greater for both intervocalic [h] and [ʔ] in a nasal harmony context than in a non-harmony (oral) context, suggesting a lowered velum for both segments conditioned by the nasal harmony span. The glottal stops examined lack complete closure, as indicated by their non-zero oral flow; we consider the implications of this for the relationship between glottal stricture and measured nasality. [Work supported by U.S. National Science Foundation Award #1918064.]