Voice Onset Time merger and development of tonal contrast in Seoul Korean stops: A corpus study

Abstract This paper is an apparent-time study of sound change in a three-way laryngeal contrast of Korean stops. The data are drawn from a read speech corpus distributed by the National Institute of the Korean Language. Voice Onset Time (VOT) of sentence initial stops and f0 of the first two vowels of the sentence produced by 117 Seoul Korean speakers were measured to determine how VOT and f0 realization of stop categories vary by speakers' age (range: 19–71) and gender. This is the first large-scale study of its kind based on data gathered from Seoul residents. The results replicate previous findings that the VOT values of aspirated stops are shortening in younger speakers' speech with the VOT difference between aspirated and lenis stops reducing accordingly and that this change is more advanced in female speech than in male speech. The novel finding of the study is that there is a trend of enhanced f0 distinction between aspirated and lenis stops, whereby the f0 distinction is amplified and extends further into the phrase in the speech of younger compared to older speakers and in the speech of female compared to male speakers. The results confirm that the establishment of tonal contrast and the loss of VOT distinction are taking place in tandem rather than in separate stages. The result also indicates that f0 enhancement as sound change is mediated by structural categories, namely all [+spread glottis] sounds including /h/ rather than narrowly targeting threatened segmental contrast in stop consonants.