Korean Linguistics - Volume 13, Issue 1, 2006

Abstract. Acoustic data elicited from 34 native speakers of Korean living in the United States pro-vide evidence for diachronic change in the voice onset time (VOT) of phrase-initial aspirated and lax stop phonemes. While older speakers produce aspirated and lax stops with clearly differentiated average VOT values, many younger speakers appear to have neutralized this difference, producing VOTs for aspirated stops that are substantially shorter than those of older speakers, and comparable to those for corresponding lax stops. The data further indicate that, within each age group, older speakers manifest sex-based differences in VOT while younger speakers do not. Despite this appar-ent shift in VOT values, the acoustic evidence suggests that all speakers in this study, regardless of age, continue to mark underlying differences between aspirated and lax stops in terms of stop closure and the fundamental frequency of the following vowel. It is concluded that the data point to a recent phonetic shift in the language, whereby VOT no longer serves as the primary cue to differentiate between lax and aspirated stops. There is not, however, evidence of any reorganization of the lan-guage as the phonemic level: the language's underlying lax ~ aspirated ~ tense contrasts endure.

Abstract. This article examines the acoustic characteristics of Korean fricatives (lax /s/ and tense /s*/) in three vowel contexts (/a/, /i/, and /u/) in the speech of AE learners of Korean. AE learners fall into two groups based on their proficiency. Acoustically, Korean /s*/ is more similar to English /s/ than the latter is to Korean /s/. In terms of A/F duration ratio, the more similar sound (Korean /s*/) was easier for AE advanced learners of Korean to produce, while in terms of amplitude difference AE learners did not distinguish the Korean fricatives at all regardless of their proficiency level. In terms of the mean A/F duration ratio, the Korean /s*/ was authentically produced by AE advanced learners, but AE advanced and beginning learners of Korean inaccurately produced Korean /s/ was L2 advanced learners and L1 speakers used different strategies in the production of Korean fricatives. AE KSL advanced learners showed primacy of duration over amplitude, while AE KSL beginning learners showed no significant acoustic cue effects in differentiating the Korean fricatives /s/ and /s*/ in production. As in the speech production of Korean fricatives /s/ and /s*/ by AE learners of Korean, factors other than years spent learning L2 should be considered to explain the acquisition of L2 sounds.

Abstract. A number of modifications affect the sound structure of foreign words as they are bor-rowed into Korean. We consider specifically the adaptation of word-final stops, liquids, and voiceless as well as voiced coronal sibilants. The particular manifestation of these is shown to corre-late with the place they hold in the syllable structure of the recipient language rather than, as might seem to be the case, with either contrastive categories of the source language or allophonic qualities of the recipient. This discussion thus contributes to the continuing debate over the awareness that listeners may have of phonetic properties that are contrastive in the source language but redundant in the recipient (and hence presumably below the threshold of categorical perception), as well as vice versa, and it offers a unified view of the factors which appear to be at play in the phonological pro-cessing of both native words and loanwords. At base is a simple yet comprehensive principle of phonological perception: Phonetic representations are interpreted according to the salient perceptual categories of the listener's native language.

Abstract. It has been claimed that a focused word may project its focus to a syntactic constituent larger than the focused item, under what are known as Focus Projection principles (Selkirk 1995; Rochemont 1998). Engdahl and Vallduvi (1996) rejected this purely syntax-based approach and proposed considering the interactions between the grammatical function and the types of an argu-ment. Chung, Kim, and Sells (to appear) applied Engdahl and Valduvi's theory to Korean and claimed that in Korean only a theme argument, but not an oblique argument (1.O or Locative PP), can project its focus to the Verb Phrase. This paper examines how VP focus is realized in Korean and tests Chung et al.'s claim that the types and the order of arguments can affect the focus projec-tion (especially 'VP focus'). The results show that there is no sensitivity to argument type, word order, or the length of VP in projecting the domain of focus to VP in Korean. Regardless of these factors, VP focus was prosodically marked by boosting the prominence of all words inside the VP, with the VP-initial word being the most prominent. Our data suggest that focus projection rules can be eliminated as proposed in Buring (2003).

Abstract. This paper deals with an old, but recurrent, topic of a certain asymmetry of in situ wh-words with respect to classical extraction domains. Huang's analysis (1982) of wh-word scope tak-ing cannot offer an empirically adequate account for this important asymmetry, apart from the conceptual problem that does not fit into the minimalist assumption of computation (Chomsky 1995). I suggest two ways of scope taking at LF: movement of way 'why' vs. unselective binding of indefi-nite wh-words by the question morpheme. I will show that scope taking of in situ wh-words along this way can nicely account for the observed asymmetry of in situ wh-words with respect to islands, conforming to the minimalist thesis of movement as a last resort.

Abstract. There are three different -nun phrases in Korean: topic, contrastive, and matching. The contrastive and the matching differ in that the matching is always followed by a focus phrase, while the former never is. Following J.-R. Kim (2004b) and Y.-B. Kim (2004), this article shows that the contrastive -nun phrase can be analyzed as a focus phrase, which is incompatible with the traditional belief that -nun is a topic marker. However, it is found that there are good reasons to believe there is a hidden focus AFFIRMATION/DENIAL following the -nun phrase in the contrastive sentence. Then, the foremost difference between the matching and the contrastive disappears: both of them are always followed by a focus phrase. We can collapse them, then, into the single category the contras-tive topic. The same LF structure involved in contrastive and the matching sentences suggests that one way to look at the contrastive topic is as a focus inducer, in the sense that the contrastive topic is always followed by a focus structure.

Abstract. The Korean reflexive caki 'self is ambiguous semantically and syntactically. This reflex-ive does not always behave as an anaphor or a pronominal. It is often not bound, nor is it consistent in its distribution as would be expected from the alternative view, which regards it as a long-distance anaphor or a pronominal. It is noted in this study, on the other hand, that the behavior of the reflex-ive is ambivalent in two ways. For one, the interpretation of the reflexive is sensitive to its co-occurring predicate and allows the reading of the abstract concept of personhood and individuality along the lines with Safir (1996). For another, the reflexive seems to behave like a Near-reflexive in the sense of Lidz (2001). Given these observations, this study has largely focused its discussion on how inherent reflexivity of predicates and the abstract concept of personhood and individuality of the reflexive may interact in determining the antecedent. It is concluded that the reflexive is a refer-entially independent and lexically idiosyncratic noun when it is conceived of as a "reified" soul, and behaves like a discourse-sensitive reflexive pronoun in other contexts.

Abstract. Honorification in Korean elevates the social status of a participant in a clause with respect to the subject and/or the hearer. Honorific marking may occur as a nominal suffix, a special honor-ific form of a noun, an honorific case particle, an honorific marker on a verb, or a special honorific form of a verb. Previous accounts have proposed a specification [HON +], with unmarked forms typically being [HON -]. Our key idea is that honorific forms introduce a dimension of meaning, the expressive meaning of Potts (2005), which is privative, and hence simply absent from all non-honorific forms.All previous accounts fail with regard to three types of fact: first, the different expressions of 'honor-ification' do not mean exactly the same thing. Second, multiple expressions of honorific marking within the same clause progressively elevates the social status of the referent: the effect is cumula-tive. Third, under the traditional analysis, some nouns have to be given a spurious and ultimately inconsistent ambiguity. We further argue that it is mistaken to consider honorific marking to be 'agreement' between, say, a subject and a verb. This position puts us in contrast with all of the syn-tactic literature on Korean (and Japanese), and some of the semantic and pragmatic literature.

Abstract. This paper compares disjunction in Kannada and Korean and examines the morpho-syntactic properties of it in these languages, focusing on Kannada -oo/illa and Korean -na. Kannada and Korean are genealogically unrelated, but share SOV word order. It is thus expected that the disjunction head follow its complement. Indeed, Kannada -oo and Korean -na follow their comple-ments. What is truly intriguing is that it shows surprisingly parallel properties. For example, Kan-nada -oo and Korean -na not only yield disjunction but also invoke indirect question interpretation. That is, both of them are used as a disjunctive conjunct as well as a question marker. In this paper, we examine the parallels and differences between the disjunction in these languages and show that the conjunctive and disjunctive reading is expressed in different modes. Under the present analysis, we can readily explain the indirect question formation, wh-scope marking, and the dummy wh-phrases in the oo-complementation. We also show that the existential quantifier interpretation of wh-indefinites plus -oo can be accounted for, assuming -oo to be a wh-question marker. Unlike what Amritavalli (2003) predicts, wh-indefinites plus disjunctive morpheme yields universal quantifier interpretations in languages like English, Japanese, and Korean.

Abstract. Many of the verb endings of modern Japanese and Korean have been created by contracting structures that consist of the stem + attached strings of particles and auxiliaries. Most of the auxiliaries have been taken from free verb stems that were grammaticalized for special purposes. Though the paradigmatic systems grew independently in the two languages, many of the ingredients go back to a common source that we can reconstruct on the basis of their shapes and meanings. Korean and Japanese share certain configurations of meaning and grammar, such as the well-known marking of focus, that are realized by markers which are not directly cognate in these structures but can be seen as cognate with forms in other structures within each language. These two languages have much more in common with each other than either has with any other language. This is why we think it is possible to reconstruct a prehistoric ancestor that can be called proto Korean-Japanese.

Abstract. The last decade has seen considerable research on conversation and grammar, influenced by the conversation-analytic research of Sacks et al. (1974). Inspired by that line of research, some Korean linguists have examined conversation by adopting the assumptions and methodology of conversation analysis (CA) into discourse analysis. This study introduces basic assumptions and research topics relating to CA, and explores the possibility of adopting CA methodology into dis-course analysis in Korean linguistics. This paper first provides a brief overview of basic assumptions, methodology, and major research topics of CA and the development of conversation-analytic dis-course studies. Then it provides a brief overview of some major findings and research topics in the interaction-based studies which have dealt with conversational data in Korean linguistics in terms of: (i) turn-taking, turn-constructional units, and turn increments, (ii) interactional functions of certain clausal connectives and sentence-ending suffixes, and (iii) other interaction-based studies on such topics as repair, demonstratives, reported speech, and so on. This research discusses how interaction-based research can provide a new way of viewing language functions; it explores: (i) turn-taking and co-construction; (ii) word-order variability, turn increments, repair, and retroactive elaboration; (iii) a conversation-analytic approach to the clausal connective -nuntey, and (iv) an interactional ap-proach to the verbal affixes -ese and -nikka in conversation, among others. Overall, this paper shows what has been, and needs to be, studied regarding the relationship between conversation, social action, and grammar in conversation in Korean linguistics.