고대영어의 두음 자음군 분석

양선기 한국영어학학회 2000 영어학연구 Vol.- No.9

공교점규약과 보상장음화

梁선基 順天大學校 1993 論文集 人文社會科學篇 Vol.12 No.-

Ingria(1980) proposes the Emty Node Convention to account for compensatory lengthening and argues the process can adequately be treated as a metrical phenomenon. The purpose of this study is to show that the metrical analysis of compensatory lengthening utilizing the Empty Node Convention leads to some problems and that the problems can satisfactorily be settled in Moraic Phonology. Various types of compensatory lengthening are discussed in this study with special emphasis laid on compensatory lengthening triggered by onset consonant deletion as in ancient Greek kalwos > ka:los 'beautiful'. In the Moraic Phonology adopted here, onset consonants are not directly linked with the syllable node but with the leftmost mora of a syllable.

柳音 앞에서의 子音重複

梁선基 順天大學校 1991 論文集 人文社會科學篇 Vol.10 No.-

This study is a non-linear analysis of the so-called 'gemination before liquids' in West Germanic and Old English. The goal of this paper is two-fold: one is to argue that the gemination before liquids is actually not one process but three separate types of compensatory gemination triggered by vowel shortening, vowel deletion and resyllabification, respectively; the other is to show that moraic phonology can account for the gemination more adequately than CV phonology. For this analysis, the gemination before liquids will be classified into three types. It will be discussed in this paper when each type of compensatory gemination can take place.

영어의 분절음배열제약

양선기 순천대학교 어학연구소 1997 어학연구 Vol.8 No.-

The purpose of this study is to delineate phonotactic constraints in English in terms of positive conditions and negative conditions. In this study it is shown that phonotactic constraints can be adequately described without recourse to negative conditions since they can always be replaced by negative conditions, and that the basic unit for the statement of sequential constraints is the syllable, not the morpheme. The organization of this paper is as follows. Section 2 considers Hooper's (1973, 1975) claim that phonotactic constraints must be stated in terms of syllable structure conditions, with no regard to morpheme boundaries, since morpheme structure conditions are expressible as and so reducible to syllable structure conditions. In Section 3 I contend, contrary to Schachter and Fromkin(1968) who suggest that negative conditions are not needed, that if-then conditions are not needed since they can always be replaced by negative conditions. Sections 4 and 5 deal with sequential constraints of syllable-initial consonant clusters and those of syllable-final consonant clusters, respectively. In this analysis of phonotactic constraints in English, it is assumed that the segment [?] does not exist in underlying representation but is derived from /n/, and that some coronals lack the Place Node in underlying representation.

서게르만어 중첩음화 : 최적성이론의 분석 An Optimality-Theoretic Analysis

양선기 순천대학교 어학연구소 2000 어학연구 Vol.11 No.-

This paper provides an analysis of West Germanic Gemination (WGG) within the framework of Optimality Theory. I first examine the environments where WGG occurred and review both linear and nonlinear rule based approaches to point out their problems. Then, I posit some constraints and discuss how WGG can be accounted for by their interaction. During the discussion, I argue that an OT analysis can give a better account WGG than previous rule-based analyese. In this paper, I only examine WGG before j which was the most prevalent and effective conditioning factor, though w, r, and l also triggered WGG. It is assumed here that before causing the gemination the thematic element j was lost before i and vocalized before consonants.

모라음운론에서의 서게르만어 중자음화

양선기 순천대학교 어학연구소 1991 어학연구 Vol.3 No.-

This study proposes a version of moraic theory and argues that so-called West Germanic gemination is not simply an insertion of consonants as it has been traditionally considered but a special type of compensatory lengthening. the moraic theory proposed here is some-what different from that of Hayes (1989) in that onset consonants are not directly linked to the syllable node but to the leftmost mora of a syllable. This makes it possible to represent the length of both consonants and vowels uniformly. In the CV-theory, resyllabification leaves and empty slot behind only when it applies to t도 input having an empty C slot in the syllable-initial position. In the moraic theory, however, and empty mora is left behind after the application of resyllabification if the second mora of the left syllable is linked exhaustively with only one segment. Due to the different effects of resyllabification, any CV-theory cannot account for the gemination in West Germanic as compensatory lengthening triggered by resyllabification, but the moraic theory can do and further predict when the process can occur.

모라표시

양선기 순천대학교 어학연구소 1992 어학연구 Vol.4 No.-

This study aims to propose a representation of moras, where onset consonants are linked with the leftmost mora of a syllable rather than with a syllable node and the word - final consonant is assumed to be extrprosodic, and to provide evidence that onset consonants should be syllabified under a mora node. Evidence is drawn from the language game Shiritori 'hip-taking', a traditional game widely played among children throughout Japan. Three saparate types of mora representations are examined one by one and their problems are pointed out, respectively.

古代英語에서의 再音節化에 의한 補償長音化

梁선基 順天大學校 1992 論文集 人文社會科學篇 Vol.11 No.-

The purpose of this paper is to analyze the compensatory lengthening process triggered by resyllabification within the framework of Morale Phonology and to account for when this process can take place. For this purpose, two problemes are dealt with in this study: 1) Dose the application of resyllabification rule always provide the context in which compensatory lengthening can coour? 2) If not, when can resliabification trigger compensatory lengthen_ing? Examined in this paper are Steriade (1982)'s analysis of compensatory lengthening by resyllabification, and onset consonant deletions in Old English. Here it is shown that com- pensatory lengthening by resyllabification can take place only when the application of resyllabification rule leaves an empty mora behind and that an empty mora is left behind after restllabification rule is applied only to the words where the second mora of the preced- ing syllable is exhaustively associated with a coda consonant.

영어 SMS 약어의 철자변이에 대한 음운분석

양선기,강석근 국제언어인문학회 2017 인문언어 Vol.19 No.2

This paper considers the phonological characteristics and type frequency of spelling variation in English SMS abbreviations. We argue that spelling variation is not random but phonologically motivated. Showing that vowels are more likely to be deleted than consonants, we claim that word-intial vowel deletion is closely related to the trochaic rhythm of English, while word-medial vowel deletion is associated with word-initial prominence. We also show that word-final e is deleted only when the preceding vowel is short. Among consonants, gh and h are most frequently omitted, and r is deleted only in coda position, reflecting the /r/-deletion rule in Modern English. Meanwhile, geminate consonants are degeminated after a tense vowel, but degemination is optional after a lax vowel. We also show that ‘8’ is the most frequently used number homophone, and that only th and ks are subject to substitution. With regard to frequency, we assert that deletion (57.42%) occurs most frequently, which is followed by number homophone (13.23%), letter homophone (12.58%), substitution (9.68%) and alphanumeric homophone (1.94%) in that order. Besides, we also argue that spelling variation may result from assimilation and word-final e-insertion, though rare (1.29% each).

최적성이론에 의한 고대영어의 h-탈락 분석

양선기 한국영어학학회 2002 영어학연구 Vol.- No.13

Yang, Seon-Ki. 2002. An Optimality-Theoretic Analysis of h-Deletion in Old Englsh. The History of English, 13. The purpose of this study is to provide a constraint-based analysis of h-deletion in Old English within the framework of Correspondence Theory (McCarthy and prince 1995), and to argue that h is deleted in onset of unstressed syllables rather than in coda. I show that previous rule-based analyses (Lass & Anderson 1975, Keyser 1975, Kiparsky &O Neil 1976, and Peinovich 1979)are not even observationally adequate in that there are too many exceptions to the rule by which h is deleted between sonorants. This study also treats compensatory lengthening and vowel contraction concomitant with h-deletion. Key words: H-deletion, compensatory lengthening, vowel contraction