Dynamic and static aspects of laryngeal co-occurrence restrictions in Korean

Hijo Kang,Mira Oh 한국음운론학회 2016 음성·음운·형태론 연구 Vol.22 No.1

This study investigates the dynamic aspects of laryngeal cooccurrence restrictions whose static aspects have been evaluated. By examining word-initial tensification and adverbializer allomorphy in Korean, we show that a laryngeally marked feature tends to co-occur with the same features in word formations. A tense consonant in the second syllable facilitates the tensification of the word-initial consonant, and an aspirated consonant in an adjectival stem requires –hi as an adverbializer rather than –i. The analysis of the entire lexicon shows that these patterns are correlated with those in the lexicon. Based on these results, we argue that Korean has an assimilation type restriction (Gallagher 2010), contrary to Ito (2014). This study provides insight about how the dynamic and static aspects of the restrictions interact to form specific patterns in the lexicon. (Chosun University and Chonnam National University)

Segmental and suprasegmental effects on Korean listeners’ English stress perception

Hijo Kang,Hyun-ju Kim 한국언어학회 2019 언어 Vol.44 No.4

English stress is realized by suprasegmental features such as duration, pitch, and intensity as well as by vowel quality. It has been known that native speakers of English are more attentive to vowel reduction than to suprasegmental cues because it offers considerable processing advantages (Fear et al. 1995, Cutler 2015). However, it has been shown that this is challenging for L2 learners of English, who depend on suprasegmental features in their production (Hammond 1986, Lee et al. 2006). Assuming that this difficulty is ascribed to their misperception, we examined how Korean listeners perceive English stress by modulating vowels at the segmental and the suprasegmental levels. Consistent with previous studies (Chung 2016a, 2016b, Kang & Kim 2016), Korean listeners relied heavily on pitch, but were also sensitive to vowel reduction, whose effect was stronger in the high proficiency group. For English native listeners, vowel reduction was the most reliable cue, but those who have been exposed to Korean accents for long time did not exploit vowel reduction for stress placement. The results suggest that phonetic/phonological patterns at the segmental level are more easily acquired than those at the suprasegmental level (Saito 2018).

The Realizations of /h/ in Seoul and Gwangju Koreans

Hijo Kang,Hyo-Young Lee 사단법인 한국언어학회 2019 언어학 Vol.0 No.84

In Korean /h/ appears in the initial position of accentual phrase but it rarely does elsewhere. After obstruents, it merges with the preceding obstruent into an aspirated stop and after sonorants, it may be voiced or deleted in Seoul Korean. Based on the fact that /h/ may be deleted even after obstruents in Gwangju (and South-Western) Korean, we hypothesized that /h/ would delete more often after sonorants in Gwangju than in Seoul. In addition, we aimed to examine the dialectal difference at the phonetic and the phonological levels. We analyzed underlying and derived aspirated stops and /h/ in a small reading corpus, which consists of five stories read by three Seoul speakers and three Gwangju speakers in their 50"s. The duration of /h/ was measured in assistance with the center of gravity values of every 10ms sections. The results showed that /h/ was deleted after obstruents in Gwangju as expected, but the VOT of aspirated stops were not significantly longer in Seoul than in Gwangju. Similarly, /h/ was weaked or deleted more often in Gwangju than in Seoul, but the duration of voiceless and voiced /h/s was not significantly longer in Seoul. These results suggest that the dialectal difference is construed as a matter of optionality, which takes place at the phonological level. The Maxent Grammar Tool gave a much higher weight to Max(h) in the Seoul data than in the Gwangju data. Accordingly, we argue that the difference is contributed to the relatively strong requirement to preserve /h/ in Seoul.

Vowels of Beryozovka Ewen: An acoustic phonetic study

( Kang Hijo ),( Yun Jiwon ),( Ko Seongyeon ) 한국알타이학회 2017 알타이학보 Vol.0 No.27

This study acoustically analyzes 13,540 vowel tokens of Beryozovka Ewen with the aid of automated post-transcriptional processing technique. The focus of the analysis is on the acoustic correlates of [RTR], which is the harmonic feature of the language. In addition to the first three formants, acoustic values representing spectral tilt such as H1-H2, H1-A2, and B1 are measured as potential acoustic cues of [RTR]. The results show that F1, F3, and B1 are the most reliable cues of the feature and that H1-H2 and H1-A2 are nearly reliable. These acoustic cues are also shown to interact with length and position. In general, the acoustic distance between [-RTR] and [+RTR] vowels are farther in long and word-initial vowels than in short and non-initial vowels, respectively. We claim that [RTR] is more appropriate for the harmonic feature of Ewen than [ATR], and that the greater perceptibility of word-initial vowels is understood as a means to facilitate the lexical access in a language with vowel harmony.

Acoustic Correlates of English Stress in Korean L2 Learners’ Perception

Hijo Kang,Hyun-ju Kim 한국영어학회 2016 영어학 Vol.16 No.2

The Asymmetric tense consonant effects in compound and word-initial tensifications in Korean

Hijo Kang,Mira Oh 한국음운론학회 2019 음성·음운·형태론 연구 Vol.25 No.1

A tense consonant blocks compound tensification (Ito 2014, S. Kim 2016) but facilitates word-initial tensification in Korean (Kang and Oh 2016, H. Kim 2016). Previous studies examined such a tense consonant effect as dissimilatory in compounds but as assimiliatory in simplex words separately. The aims of this paper are twofold. First, we will test if a tense consonant is affected differently depending on the morphosyntactic domains by studying both compound and word-initial tensification processes, unlike previous studies. Second, we will investigate what motivates the asymmetry of the tense consonant effect between compound and word-initial tensification processes. We conducted a survey where 40 Korean speakers selected their own pronunciations for 120 compounds and 100 simplex words. We found that a tense consonant in fact behaves differently in compound and in word-initial tensification processes. We present two possible reasons why such a difference is observed. The perception view assumes that each tensification has its own perceptual function. Compound tensification is intended to make a compound perceptually distinguishable from each of its components. Compound tensification is then avoided before another tense consonant since a tense-tense sequence of consonants could make them perceptibly similar to simplex words. The production view of the tense consonant effect emphasizes the avoidance of consecutive long vowel-to-vowel intervals. Laryngeal cooccurrence restrictions attested in tensification in Korean will be discussed in light of both production and perception.

Phonetic grounding of position and height asymmetries in hiatus resolution: An acoustic analysis of Korean VV sequences

Hijo Kang 한국음운론학회 2013 음성·음운·형태론 연구 Vol.19 No.2

This paper deals with two typological patterns in hiatus resolution, position and height asymmetries, from the perspective of phonetics and phonology interface. In languages where hiatus is disfavored in the output, the weakening of a vowel is more likely to take place in V1 and high vowels than V2 and non-high vowels, respectively. Assuming that these typological asymmetries are rooted in the acoustic and auditory mechanisms of human speech (Ohala 1993), I conducted an acoustic analysis of Korean VV sequences, which were recorded by six Korean speakers in two different speech rates. The results showed that the steady state of V1 is shorter than that of V2 in CV1V2 words but not in CVCV words. High vowels turned out to have shorter steady states than non-high vowels and the proportional difference increased in fast speech. These results suggest that the typological patterns are not only morphologically motivated (e.g., Casali 1996) but phonetically grounded. (Seoul National University)

음운 현상의 동기적 비일관성과 음운 이론

강희조(Hijo Kang) 사단법인 한국언어학회 2014 언어학 Vol.0 No.70

This paper investigates inconsistency in phonology and aims to find an appropriate model to incorporate it into linguistic theory. Phonological theories have been focusing on the phenomena that show consistent results through various types of observations and experiments. These results have served as the basis of ‘one’ grammar, which is represented as ‘one’ ranking for the phonology of a language. However, recent studies show that a language can produce different results, depending on whether the stimuli are real or nonce words or whether the stimuli are roots or inflected words. This paper presents three cases; vowel harmony in Korean verbal conjugation, accent patterns in North Gyeongsang Korean, and Segmental OCP in languages. In the first two cases, native speakers show different behaviors with real words from with nonce words. In the last case, the patterns found in the lexicon have little impact on the phonological processes in the languages. These cases are not explained by one grammar or one ranking of constraints. I argue that we should separate two processes: lexicon and phonology. When it comes to real words, speakers depend on the stored tokens rather than the phonological process related to the words, while they should rely only on the phonology for the processing of nonce words. The separation of the two processes can explain not only the inconsistency in synchronic data, but diachronic changes by means of the interaction of the two processes.

한국어 용언활용에서 보이는 모음조화의 변이와 변화

강희조(Hijo Kang) 사단법인 한국언어학회 2016 언어학 Vol.0 No.76

This paper investigates the variation and change in the vowel harmony of verbal conjugation in Korean. By comparing the results of two judgment surveys, which were carried out in 2008 and in 2016, I show that not only the synchronic variation but the diachronic change is affected by morphophonological features such as regularity of stem, trigger vowel, and type of suffix. The innovation is faster in p-irregular stems, /a/-stems, and sentence-final suffixes than regular, /o/-stems, and other suffixes, respectively. With these results, I argue that the restructuring of stem forms and the perceptibility of morphemes play a crucial role in the acceleration or retardation of change. In addition, this study also shows that variation does not always result from the instability of change.

The Prosodic Effect of Compound Tensification in Korean

Yeonju Kim,Hijo Kang 한국중원언어학회 2017 언어학연구 Vol.0 No.45

This study investigates the prosodic characteristics of compounds in Korean, compared to those of simplex words and phrases. In the production experiment, 45 pseudo minimal pairs or triplets that were produced by 12 Korean speakers from Seoul and Gwangju were acoustically analyzed. Duration of target segment (closure duration of lax vs. tense), preceding vowel, preceding consonant, and following vowel are measured and compared. Pitch contour of preceding vowel was also examined. First, we show whether and how words are prosodically different from phrases. The results show that the durations of whole expression and of closure are longer in phrases than in words. The pitch pattern of phrase also differs from that of word, though it was not consistent through dialects and genders. Second, compounds are compared with simplex words. Duration shows similar differences in compounds vs. simplex words, but pitch does not show significant differences, which leads us to the conclusion that duration plays a crucial role in distinguishing morphologically and/or syntactically different structures such as simplex words, compounds, and phrases in Korean.