Variable production of voiceless sonorants in Hakha La

Research on Tibeto-Burman voiceless sonorants often centers on nasals, which are standardly described as having two language-specific phonetic realizations: “voiceless unaspirated” nasals have a period of voicelessness and nasal airflow during closure followed by voicing prior to oral release ([m̥m]), while “voiceless aspirated” nasals have a period of voicelessness and nasal airflow optionally preceded by a voiced period, and voicing following oral release ([(m)m̥ʰ]) (e.g., Bhaskararao and Ladefoged, 1991; Chirkova et al., 2019). Little language-internal inter- or intra-speaker variation has been reported. An exception is South Central Tibeto-Burman (Chin) languages, such as Hakha Lai, in which voiceless nasals reportedly exhibit considerable inter- and intra-speaker variation (Hoffmann, 2018; Ziegler et al., 2022). As thorough investigation of this variation is absent from the existing literature, this paper investigates acoustic data from 9 native speakers of Hakha Lai (3M, 6F). Data for all voiceless sonorants (nasals, lateral, and rhotic) is presented, revealing that: (1) voiceless nasals in Hakha Lai indeed show a high degree of variation in the phasing of laryngeal and supralaryngeal gestures, defying the two-type approach; and (2) this variation extends to voiceless laterals and rhotics, which show similar bi-phasic and even tri-phasic variation.