Stop Consonant Typology: Contrast, Laryngeal Activity, and Enhancement
Lucas Annear & Joseph Salmons, Language Sciences
How does contrast relate to the realization of voice onset time in stops cross-linguistically? How do languages enhance these contrasts? Various typological surveys of laryngeal features to date rely on a distinction between ‘voiced’ and ‘voiceless’ stops, showing voiceless stops as by far the most common type cross-linguistically, with aspirated stops mostly found in languages having multiple laryngeal contrasts (three or more series of stops). We present a survey of such patterns built on the role of phonological contrast, drawing on work in Laryngeal Realism and related traditions. On this view, languages with two series of stops will have one laryngeally unspecified series, and one series specified for an active laryngeal gesture, such as [spread glottis] for aspiration, or [voice] for pre-voicing/voicing. We draw on a dataset from a survey of voice onset time (Chodroff et al., 2019), and present data from 93 language varieties. We show first that VOTs cluster in three distributions: lead-lag (pre-voiced), short-lag (plain), and long-lag (aspirated) stops. Plain, short-lag stops were the most common type of stop in the data set (present in 79 of 93 varieties), and these plain stops were the most common stop type in languages with one series of stops. Active laryngeal gestures for voicing (lead-lag stops) and aspiration (long-lag stops) were associated with the presence of a laryngeal contrast. Thus, greater laryngeal activity in stops was associated with the presence of 2-or-more series of stops, and plain, short-lag stops were the most common stop type in languages with no laryngeal contrast. We further explored the realization of short-lag stops in 31 languages with an aspiration contrast (long-lag ~ short-lag) and 26 languages with a voicing contrast (lead-lag ~ short-lag), finding that languages enhance these contrasts by producing short-lag stops in a direction opposite the contrasting series (e.g., languages with a lead-lag ~ short-lag contrast tend to produce short-lag stops with slightly longer VOT than languages with a long-lag ~ short-lag contrast). Together, these results indicate that greater laryngeal activity in stops is associated with the presence of a laryngeal contrast, but also that languages enhance these contrasts through adjustments to laryngeally plain stops.
