Phonological contrast and coda saliency of sonorant assimilation in Korean

손형숙 SPRINGER 2008 Journal of East Asian Linguistics Vol.17 No.1

영어와 한국어의 CVCV 구조에서 폐쇄음의 유․무성이 모음의 지속시간에 미치는 영향

손형숙,임신영 새한영어영문학회 2019 새한영어영문학 Vol.61 No.2

This study investigates the effect of stop voicing on vowel duration in the initial stressed syllable of CVC and CVCV English words produced by two groups of native English speakers and Korean learners of English. The production experiment was also conducted for the bisyllabic Korean words. Experiment results confirmed voicing-induced vowel lengthening in preconsonantal position of monosyllabic words. Also shared by the two speaker groups is the finding that the voicing of a preceding stop significantly affects vowel duration. In bisyllabic English words, however, the two speaker groups contrast: for the native English speakers vowel duration is affected by voicing of the word-medial stop, while this is not observed for the Korean learners of English. The diverging pattern of phonetic realization is claimed to be due to ambisyllabicity within a trochaic foot in English.

Random patterning in paradigms of noun-final coronal obstruents: Palatalization, reanalysis, and analogical leveling

손형숙 한국음운론학회 2012 음성·음운·형태론 연구 Vol.18 No.1

Focusing on random variation in paradigms of /th/- and /ch/-final stems in Native Korean nouns, this study investigates a comprehensive range of paradigmatic patterns of coronals in relation to the categorical stages of leveling in progress. The paradigmatic patterns derived from the phonological system are drawn into comparison with the survey results. A survey designed to test the phonetic value of the stem-final coronal was conducted to explore the paradigmatic patterns speakers put to use for lexical items, and furthermore, to abstract away a gradient system of paradigmatic types reflecting the speakers’ collective preference in usage. Based on paradigmatic types and their frequencies, this study demonstrates that random variation of the paradigms of the /th/-final stems is motivated by analogical extension of palatalization in progress as well as skewed distribution in the lexicon. Random patterning of /ch/-final stems, however, is primarily attributed to skewed distribution since reanalysis undoing palatalization plays a limited role. The effect of the lexical frequency on the paradigm frequency supports the claim that the phonetically conditioned sound change affects the high frequency words first before the low frequency words, while analogical changes affect the less frequent items before the more frequent ones.

외래어의 음가 인식과 음운론적 구조 보존

손형숙 언어과학회 1994 언어과학연구 Vol.11 No.-

Prosodic Structure of Wh-interrogatives and Intervention Effects in North Kyungsang Korean

손형숙 한국언어학회 2017 언어 Vol.42 No.4

Sohn, Hyang-Sook. 2017. Prosodic Structure of Wh-interrogatives and Intervention Effects in North Kyungsang Korean. Korean Journal of Linguistics, 42-4, 735-760. Resorting to prosodic structure of double foci construction, this study aims to provide a prosody-based account of intervention effects, and claims that fronting of the wh-in-situ as a syntactic operation is prosodically motivated. In this approach, intervention effects arise as a consequence of prosodic search for the optimal form to resolve the conflicts among prosodic constraints induced by focus. Selection of optimal prosodic structure is accounted for in the framework of Optimality Theory and this study shows how resolution of conflicts among prosodically motivated constraints results in asymmetry in linear precedence between interveners and wh-operator. This supports the claim that prosody is a crucial factor in determining grammatical linearization. Prosodic constraints induced by wh-operator and focus are further extended to indefinite pronominals in focus construction. As a corollary to interactions of prosodic constraints induced by focus, linear precedence relation between indefinite pronominals and focus naturally falls out from the constraint ranking required anyway. Examining a set of pairwise precedence relations based on prosodic structure, we draw the generalization that indefinite pronominals are followed by wh-operator, which in turn is followed by focus-inducing items. This indicates that fronting of DPs, relative to another focus, is a necessary condition for optimal phrasing in focus construction, and corroborates the claim that syntactic structure is conditioned by prosodic structure. (Kyungpook National University)

미명세이론과 설정음의 음운적 표시

손형숙 언어과학회 1996 언어과학연구 Vol.13 No.-

미명세이론의 음운론적 의미

손형숙 한국언어학회 1997 언어 Vol.22 No.1

영어 접사첨가형에 나타난 어간말 폐쇄음의 기식성 실현과 음성적 충실성

손형숙(Sohn, Hyang-Sook) 새한영어영문학회 2023 새한영어영문학 Vol.65 No.2

Measuring the VOT of the stem-final stops in affixed forms, this paper examines the relationship between the word-final release noise and the comparable stem-final burst noise in affixed forms. The experiment indicated that the three types of level 2 suffixes are not altogether parallel with respect to their effect on aspiration of the preceding stem-final stops. Also indicated is that the gradient realization of aspiration in affixed forms is affected by the phonological context of the stem, which in turn effects gradience of the release noise in the word-final stops (Kang 2003). These results suggest that the two allophonic properties of the release noise and the corresponding burst noise are related as perceptual cues in the wake of morphological variation. In this perspective, the burst noise of stem-final stops is interpreted as an effort to preserve the word-final release noise. This type of output-to-output correspondence in allophonic release-to-burst noise between the base and the affixed forms emerges naturally in Optimality Theory through the affix-specific constraints and overapplication of Ident-BA over the normal markedness constraints.

자연 발화 코퍼스에 기반한 한국어 폐쇄음의 격음화 실현 양상

손형숙(Sohn, Hyang-Sook),임신영(Shinyoung Lim),안미애(Miae Ahn) 언어과학회 2016 언어과학연구 Vol.0 No.78

Based on a spoken corpus, this study investigates phonetic realization of aspiration merger in Korean, whereby plain stops become aspirated before the fricative /h/. Unlike phonological prediction in the wake of the canonical rule application, the VOT analysis of stops in the merger environment in the Seoul corpus confirms that phonetic realization of aspiration is not categorical but gradient. The production pattern of aspiration across the four age groups of talkers ranging from their tens to their forties in the corpus shows that the age functions as an important factor in distribution of the VOT variation: the degree of aspiration turned out to be in inverse proportion to the age, reflecting the careless speech of the older speakers. The results also support the tendency that gradient phonetic realization of aspiration merger is affected by word type and prosodic structure conditioned by stops and their following fricative in the merger environment.

운율경계와 억양하강에 따른 폐쇄음과 마찰음의 음향적 특성

손형숙(Hyang Sook Sohn) 언어과학회 2014 언어과학연구 Vol.0 No.68

This study investigates the effect of prosodic boundary on phonetic realizations of obstruents in word-initial position in North Kyungsang Korean. Based on the acoustic parameters such as VOT, frication duration and F0, it examines the way in which alveolar stops [t] and [th] and the fricative [s] are phonetically strengthened in terms of prosodic domains of U, IP, PP and Wd. Statistical analysis indicates that the effect of segment-based acoustic parameters such as VOT and frication duration is not as uniform across the target obstruents as that of suprasegmental parameter F0: VOT-[t]: U>IP, PP>Wd; VOT-[th]: U>PP>Wd, IP=U,PP; Dur-[s]: U, IP>PP>Wd; F0-[t, th, s]: U>IP, PP>Wd. A considerable degree of correlation between segmental and suprasegmental acoustic parameters was found in hierarchical phonetic strengthening, namely U>PP>Wd. However, discrepancy is observed between stops and fricatives in their degree of correlation between segmental and suprasegmental parameters. Phonetic properties of word-initial obstruents in various positions within PP and Wd reveal that phonetic strengthening within a domain is not simply proportionate with linear precedence of each loci scaled with reference to the relative phonological distance; rather, it is a reflection of the internal prosodic structure of word-initial obstruents.