Working Papers in Linguistics - 2010

Permanent URI for this collection


Recent Submissions

Now showing 1 - 7 of 7
  • Item
    Acquisition of Korean Disjunction under Negation
    (University of Hawai'i at Mānoa Department of Linguistics, 2010-12-01) Lee, On-Soon
    This study investigates Korean-speaking children’s and Korean child heritage learners’ comprehension of sentences containing a disjunction operator with a negated verb. Two experiments were conducted. The first shows that children robustly reject the disjunctive interpretation (81.6%) but accept the conjunctive interpretation (78.9%). Surprisingly, Korean adults marginally accept the disjunctive interpretation (34.2%). This raises the possibility that Korean -(i)na behaves differently from disjunction in languages like English or Japanese. Experiment 2 examines this possibility in four groups. The stories used as experimental materials were modified by reducing the number of characters so as to reduce demands on working memory. The structure of target sentences was also revised to avoid a highly biased interpretation (conjunctive interpretation). The results of Experiment 2 show that Korean-speaking adults and children and Korean child heritage learners accept target sentences around 33% of the time in the context favoring the disjunctive interpretation, whereas only Korean adult heritage learners reject target sentences 100% of the time. The findings indicate that the Korean disjunction -(i)na ‘or’ under negation behaves differently than English or or Japanese ka.
  • Item
    Acquisition of Entailment Relations in Korean Causatives
    (University of Hawai'i at Mānoa Department of Linguistics, 2010-12-01) Choe, Jinsun
    This study investigates whether two Korean causative types (morphological and syntactic) have the same entailment properties in adult and child Korean. Patterson (1974) claims that only the morphological causative entails the occurrence of the caused event, while Kim’s (2005) experimental study found no such entailment for either type of causatives. In this study, a Truth Value Judgment Task (Crain and McKee 1985; Crain and Thornton 1998) was conducted with sixteen Korean-speaking adults and showed that the entailment relation is required for the morphological causative. Twenty-five Korean-speaking children participated in the same task and behaved similarly to adults in that they rejected the morphological causative when the caused event did not take place. On the other hand, it was revealed that some children were sensitive to the type of causation depicted in the task, independent of the entailment properties. They showed a tendency to link the morphological causative only when it was associated with direct causation, but not with indirect causation. This observed difference between adults and children may be explained by the Iconicity Principle (Haiman 1983), which predicts the morphological causative to be associated with direct causation, and the syntactic causative with indirect causation.
  • Item
    Baseline Results from a Psycholinguistic Tool for the Assessment of Language Shift in Truku Seediq
    (University of Hawai'i at Mānoa Department of Linguistics, 2010-05-01) Tang, Ai-Yu (Apay)
    A major obstacle to the early diagnosis of language shift is the absence of an easy-to-use measure of lan-guage strength. The body-part naming task being developed as part of the Hawai‘i Assessment of Language Access (HALA) project exploits the fact that the speed with which bilingual speakers access lexical items in their two languages offers a sensitive measure of relative language strength. It also allows the evaluation of label accuracy, another indicator of language strength. The purposes of this study are to (i) further assess the HALA approach, (ii) assess Truku strength (vs. Mandarin) across age groups, and (iii) establish base-line data as a starting point for developing conservation programs. A total of sixty-eight participants in four age cohorts were tested (ages: 10–15, 16–25, 26–40, and 41–65). All were ethnic Truku. As predicted, Truku (non-dominant language) produced longer response times and lower accuracy than Mandarin (domi-nant language) across all speakers suggesting a cross-generational decline of Truku. Moreover, response time offers a more precise and superior measure of language access than accuracy does. Young adults and youth score equally well on vocabulary access in Truku, but the youth have slower response times˜suggesting further decline and the need for urgent remedial action if Truku is to survive for another generation.
  • Item
    Tone-Melody Relationship in Cantonese
    (University of Hawai'i at Mānoa Department of Linguistics, 2010-04-01) Lau, Elaine
    Cantonese lyrics are found to adhere to the pitch contours and relative pitch heights of the tones, in or-der to preserve in the lyrics the tonal values of the words. A series of meta-analyses over a selection of Cantonese children’s songs was performed; it investigated (i) the distribution of lexical tones in correlation with musical melody; (ii) the relationship between tonal pitch and melody; and (iii) how contour tones are dealt with in mapping lyrics with music. This study showed that the writing of lyrics in tonal languages is not as free as in non-tonal languages, and takes into account the lexical tones, in order to preserve the tonal distinctions between words.
  • Item
    Early Sensitivity to Telicity: The Role of the Count/Mass Distinction in Event Individuation
    (University of Hawai'i at Mānoa Department of Linguistics, 2010-04-01) Hara, Yukie
    This paper presents evidence that English-speaking children are sensitive to telicity based on the count/mass distinction of the object noun in verb phrases such as eat an apple (telic) vs. eat ice cream (atelic). Previous work (Wagner and Carey 2003) has demonstrated that children use the presence/absence of the object in a verb phrase to recognize the end point of an event, though children continued to show a bias for a spatio-temporally defined individuation compared to adults. In the present experiment, telicity was specified by the count/mass distinction of the object noun. The results both replicated and extended the general findings of Wagner and Carey: children showed knowledge of telicity, but the spatio-temporal bias was still dominant in their event individuation. It is concluded that children’s sensitivity to telicity stems from sources ranging from syntactic to semantic.
  • Item
    The Interpretation and Transitivity of -un/-an in Truku
    (University of Hawai'i at Mānoa Department of Linguistics, 2010-03-01) Tang, Ai-Yu (Apay)
    The analysis of the verbal suffixes -un and -an in Seejiq morphology is controversial. Two distinctive ap-proaches are found in the previous studies. In the first, they are analyzed as “focus” markers and as separate morphemes. In the second, they are separate “voice” morphemes, under the cover term Goal voice mor-pheme. This paper presents an alternative analysis and shows that neither the focus- nor the voice-based approach successfully clarifies the difference between these two morphemes; careful attention to the use of -un and -an reveals that it is problematic to conclude that -un is patient focus, whereas -an is locative focus. In addition, the previous analyses are insufficient in terms of interpreting distribution. I (i) agree that -un and -an are separate voice morphemes under the cover term Goal voice; (ii) propose that their difference is not strictly one of voice/focus, but of the degree of transitivity in the sense of Hopper and Thompson 1980; and (iii) conclude that -un has high transitivity and -an low transitivity.
  • Item
    English Irregular Past Tense Verb Data from Three Korean Returnee Children
    (University of Hawai'i at Mānoa Department of Linguistics, 2010-03-01) Kang, Sang-Gu
    This study analyzes how three Korean children who returned to Korea from the U.S. use English ir-regular past tense verbs. When children from an English-speaking environment move to a non-English-speaking one, their exposure to English is expected to decrease. As a result, some attrition can be expected. Experimental (judgment and elicitation tasks) and naturalistic data from the children were collected for a year after their return. The results indicate that the children maintained their proficiency during their first year of return. Also, error analysis of naturalistic data shows that their error rates approximate those for na-tive English-speaking children.