It is well known that the binding domain of Korean reflexives is ambiguous. A few psycholinguistic studies have shown that caki and casin tend to prefer a long-distance antecedent while caki-casin strongly prefers a local antecedent. However, it should be noted that this pattern of binding preference has been obtained only using comprehension tasks. To assess the generalizability of the preference pattern, the present study examined the antecedent preference of Korean reflexives in a comprehension task, a production task, and a corpus analysis. The comprehension task showed similar binding patterns for caki, casin, and caki-casin to those of previous studies. Conversely, the production task and the corpus analysis revealed considerably different antecedent preferences for caki and casin. The task-related differences can be explained by the different pragmatic and morphosyntactic factors involved in comprehension and production. Modality (speaking vs. writing) can also make differences in the binding preference patterns.

This study aims to investigate the relationships among writing self-efficacy, affective factors, and writing proficiency of Chinese EFL high school students. A total of 187 subjects responded to a questionnaire with 43 items, which surveyed linguistic self-efficacy, composition self-regulatory efficacy, motivation, goal orientation, task value, and anxiety held by the subjects. These subjects had also contributed their essays that were scored by two raters. The correlations among the six constructs were statistically significant (p<0.01). In addition, the results of the multiple linear regression revealed that 55.2% variance in writing proficiency was explained by the linear combination of linguistic self-efficacy, composition self-regulatory efficacy, motivation, goal orientation, task value, and anxiety. Also, it was found that the group of students with extrinsic goal orientation outperformed the other group of students with extrinsic goal orientation. Finally, some implications for enhancing English writing strategies for high school students were made.

This study aims to provide the grammatical and discourse properties of that-complement clause extraposition (hereafter CCE) constructions with the help of a total of 3,791 sentences from COCA. As a starting point, we lend weight to the argument that the extraposed clause of the head noun appears to be more akin to appositive than relative clause constructions. In corpus findings, the singular form as well as the definite article are clearly more frequent with respect to use of the head noun characterized with [ANIMATE -] and [THEME +] values. As well, the present tense predominantly comes into use, whereas the presentational verbs take the extraposed that-complement clause fairly often. Of particular relevance here is that we provide several types of modality of the head noun, so the epistemic value making up most of the occurrences is further classified into five groups: evaluation, argument, evidence, idea, and perception. The final question addressed in our investigation is what illocutionary force a speaker or writer conveys within the that-complement clause. Self-assertion shows the highest rise in the use of illocutionary force, while prior context is highly favored over no discourse-based information in CCE construction.

This study focuses on the emergence of the unmarked (TETU) concerning syllabic structure and segmental features in the base of two Chinese dialects like Shunping and Pingyao. Partial reduplication generally exhibits intriguing TETUs in the reduplicant. In Shunping, the onglide and a vowel simplify into a simple vowel while a coda consonant deletes in the base. This shows structural TETU in the base: no diphthong and a closed syllable. The reduplicant in Shunping implies the prevocalic glide belongs to the rime. The segmental content of the rime in the reduplicant comes from the input. This shows a direct correspondent relation between the input and the reduplicant. In Pingyao, the rime excluding a prevocalic glide in the base is replaced by a fixed [-Λʔ]. The rime including the prevocalic glide in the input is copied; the copied part appears after the fixed onset in the reduplicant. The base segments display two TETUs. One is related to the consonantal place features of the coda, and the other is with a reduced fixed vowel. However, the prevocalic glide seems to affiliate neither with the onset nor the rime. The Pingyao dialects also demonstrate direct correspondence relations between the input and the reduplicant. To account for the variant implementations of TETU, we adopt the full reduplication model proposed by McCarthy and Prince (1995), and put forth the constraint ranking schema for the TETU in two Chinese dialects.