Department of Social Psychology, Sookmyung Women's University, Seoul, Korea.
Department of English Education, Hongik University, Seoul, Korea.
PLoS One. 2020 Oct 22;15(10):e0240682. doi: 10.1371/journal.pone.0240682. eCollection 2020.
In present-day Seoul Korean, the primary phonetic feature for the lenis-aspirated stop distinction is shifting from VOT to F0. Some previous studies have considered this sound change to be a tonogenesis, whereby the low-level F0 perturbation has developed into tonal features (L for the lenis and H for the aspirated) in the segmental phonology. They, however, have examined the stop distinction only at a phrase- or utterance-initial position. We newly explore the sound change in relation to various prosodic structural factors (position and prominence). Apparent-time production data were recorded from four speaker groups: young female, young male, old female, old male. The way the speakers use VOT versus F0 indeed varies as a function of position and prominence. Crucially, in all groups, VOT is still used for the lenis-aspirated distinction phrase-medially due to the lenis stop voicing. This role of VOT, however, is found only in the non-prominent (unfocused) condition, in which the F0 difference is reduced to a low-level perturbation effect. In the prominent (focused) context in which tones come into play, the role of VOT diminishes, led by young female speakers. These can be interpreted as a prosodically-conditioned, complementary use of the features to maintain sufficient contrast. Importantly, however, the tonal difference under focus is not bidirectionally polarized, so that F0 is not lowered for the lenis stop. A lack of direct enhancement of the distinctive L tone weakens a possibility that F0 is transphonologized to the phonemic feature system of the language. As an alternative to the view that tonal features are newly introduced in the segmental phonology, we propose a prosodic account: the sound change is best characterized as a prosodically-conditioned change in the use of the segmental voicing feature (implemented by VOT) versus already available post-lexical tones in the intonational phonology of Korean.
在当今的首尔韩语中，软音-送气音区别的主要语音特征正从 VOT 向 F0 转移。一些先前的研究将这种音变视为声调生成，其中低水平的 F0 干扰已发展成为音段音系中的声调特征（L 代表软音，H 代表送气音）。然而，它们仅在短语或话语起始位置检查了停止区别。我们新探索了与各种韵律结构因素（位置和突出度）相关的音变。从四个说话人组（年轻女性、年轻男性、老年女性、老年男性）记录了明显时间的产生数据。说话人使用 VOT 与 F0 的方式确实因位置和突出度而异。至关重要的是，在所有组中，由于软音停止发声，VOT 仍用于短语中间的软音-送气音区别。然而，这种 VOT 的作用仅在非突出（非焦点）条件下找到，在这种条件下，F0 差异减少到低水平的干扰效应。在声调发挥作用的突出（焦点）语境中，VOT 的作用减弱，以年轻女性说话者为首。这可以解释为韵律条件下使用特征的互补作用，以维持足够的对比。重要的是，然而，焦点下的声调差异不是双向极化的，因此 F0 不会降低软音停止。独特的 L 音的直接增强不足削弱了 F0 被转译为语言音系特征系统的可能性。作为声调特征在音段音系中被新引入的观点的替代，我们提出了一种韵律解释：这种音变最好被描述为音段发声特征（通过 VOT 实现）与韩语韵律音系中已经存在的词后声调的韵律条件下使用的变化。
