JK ワークショップ

Workshop in Japanese/Korean Syntax and Semantics: What constitutes data in linguistics

--17^th Korea-Japan workshop on linguistics and language processing

日時: 2015年5月9日～10日

場所: 京都大学文学研究科 地下大会議室

会議使用言語：英語

主催：

京都大学文学研究科言語学専修

科学研究費補助金：基盤研究C 基盤研究（C) 2014~2016 「「視点」とモダリティの 言語現象―「意識」、エンパシー、阻止効果―」（代表 西垣内泰介 神戸松蔭女子学院大学教授）

写真のページ

プログラム

5月9日

May 9th

1．1.30-2:20 Daisuke Bekki (Ochanomizu Univ.)

Testability in Dependent Type Semantics

2．2:20-3:10 Suzushi Hayata (University of Tokyo)

Manchu Appositive Genitive Structure ハンドアウト

Coffee Break

3 3:40:4:30 Jae-Woong Choe (Korea Univ. and JSPS fellow at Waseda Univ.) ハンドアウト

A quantitative covarying-collexeme analysis of the adnominal constructions in Korean and Japanese

4. 4:30:5:20 Jae-il Yeom (Hongik University) ハンドアウト (改訂版）

Copula in Korean pseudoclefts and intensionality

Reception

5月10日

5. 10:20～11:10 Yurie Hara and Eric McCready (City University of Hong Kong and Aoyama Gakuin Univ.)

A Default-Logic Analysis of the Unexpectedness Expressed by the Cantonese Wo3 ハンドアウト

6. 11:20~12:10 Takeo Kurafuji (Ritsummeikan Univ.)

Alternative-induced plurality

Lunch Break

7．13:10～14:00 Taisuke Nishigauchi (Shoin Women’s University) ハンドアウト

The blocking effect as conflicting Agreement

8. 14:00 ~14:50 Sanae Tamura (Hokusei Gakuen)

Modality and point-of-view phrases in Japanese

Coffee Break

9. 15:10-16:00 Heedon Ahn and Youngjoon Cho (Konkuk University)

On the variation of acceptability judgments:

Naïve vs. expert intuitions on the multiple accusative construction in Korean ハンドアウト

10. 16:00-17:00 Hajime Hoji (USC)

Testability, reproducibility, and the significance of experimental results

Abstracts

Jae-il Yeom

There are languages like Spanish, Russian and Washo in which a copula shows a permanent vs. temporary or individual-level vs. stage-level distinction. Korean is also such a language. In Korean pseudoclefts, adnominal clauses always modify kes, which is supposed to denote things. However, a pseudocleft is used even in cases where the kes-phrase is supposed to denote (a) person(s). To account for these, I claim that when the kes-phrase looks like it denotes (a) person(s), it actually denotes a concept or, more generally an intensional semantic entity. To support this claim, I show that when the kes-phrase is supposed to denote (a) person(s), it is not pluralized, and two kes-phrases are not conjoined. This implies that what the kes-phrase denotes is not (a) person(s) in a pseudocleft. On the other hand, when the kes-phrase denotes (a) thing(s), it can be pluralized and two post-copular phrases can be conjoined. This shows that such a pseudocleft is predicational, and that when the kes-phrase denotes (a) person(s), the pseudocleft is an identity statement. Therefore the two uses of pseudoclefts are attributed to two different meanings of the copula. This leads to two different restrictions on the tenses of the two uses of pseudoclefts.

Jae-Woong Choe

Adnominal constructions in Korean and Japanese appear to be the same in their syntactic patterns, having an N(P)-ui/no N(P) format. But seen from a larger distributional perspective, they seem to show some distinctive characteristics (Choe, Shudo and Harada, 2012). This presentation takes an indepth look at the semantic/collocational characteristics of the construction, and show how we can characterize the similarities and differences between Japanese and Korean on the basis of a medium sized bilingual parallel corpus (the Sejong Korean-Japanese parallel corpus). We adopt a construction based approach to the issue. In particular, we make use of the 'covarying-collexeme analysis' (Stefanowitsch and Gries, 2005) in order to represent and compare in statistical terms the association strength for each pair of the two N(P) involved.

Yurie Hara and Eric McCready

In the framework of Conversation Analysis, Luke (1990) shows that the

meaning of Cantonese sentence-final particle /wo/ involves the

violation of expectations. In particular, /wo/ can indicate not only

the unexpectedness of the content but also that of the current

discourse move that the speaker makes. This paper takes Luke’s insight

and provides a default-logic analysis of /wo/. The notion of

unexpectedness is formally characterized by normality conditionals.

The analysis has a further implication on the Gricean Cooperative

Principle in that the use of /wo/ makes reference to the general

knowledge which includes conditions on how the discourse should

normally proceed.

Hee-Don Ahn and Youngjoon Cho

Native speakers’ introspective judgments of sentence acceptability have been an essential tool for linguistic research. However, this traditional method of data collection has been criticized in various respects, and the crucial criticism concerns its methodological ‘informality’; Namely, such judgment data cannot evade inter- and intra-speaker variation, as it is not gathered by means of objective and rigid procedures. A number of alternatives have been suggested such as corpus data, but judgment data still functions as the primary source of empirical evidence in syntactic research, due to its efficacy. Since the work of Bard et al. (1996) and Cowart (1997), more formal experimental methods have been gaining popularity, in a trend known as ‘experimental syntax’. Our study explores the utility of experimental syntax in the area of Korean syntax via a direct comparison of the results of informal judgment collection methods with the results of formal judgment collection methods. We investigate the utility of experimental syntax in the area of Korean linguistics, focusing on the issue of comparison between linguists versus non-linguists as the source of judgment data.

For these purposes, the multiple accusative construction in Korean provides a testing ground in this study. This construction occurs where more than one accusative phrase is permitted in a single clause. The construction provides an ideal opportunity to test whether reliance upon linguists as judges is reliable. The present study reports the findings among 160 participants (33 naive vs. 127 experts) performing 7-point Likert scale judgment tasks on five types of the multiple accusative construction.

The results suggest that judgment data from linguists and non-linguists generally converge. However, the judgments of generative syntacticians differ significantly from those of other groups (such as functionalists and non-linguists). Our results partially support the findings in D?browska (2010) and Sprouse et al. (2013), while ours also reveal nontrivial differences in some respects. For example, our results pattern with D?browska (2010) in that generative syntacticians show different judgments from functionalists/non-linguists contra Sprouse et al. (2013). However, unlike D?browska (2010) the two groups’ judgments are statistically different in well-formed sentences as well as ill-formed data. This calls for thorough empirical scrutiny through rigid experimentation in the field of Korean syntax. On the other hand, our results partially support Sprouse et al. (2013) in that a significant difference between acceptability judgments of pairwise phenomena is more meaningful than absolute acceptability ratings of a single sentence. Furthermore, contrary to Hofmann’s (2013) suggestions, the results for the present experimental stimuli do not necessarily need to be graphed against grammatical and ungrammatical fillers.

References

Bard, E.G., Robertson, D., Sorace, A., 1996. Magnitude estimation of linguistic acceptability.Language 72, 32-68.

Cowart, W., 1997. Experimental syntax: Applying objective methods to sentence judgments. Sage Publications, Inc.

Dąbrowska, E., 2010. Naïve vs. expert intuitions: an empirical study of acceptability judgments.Linguistic Review 27, 1-23.

Hofmann, T., 2013. Obtaining introspective acceptability judgments. M. Krug & J. Schlüter (eds.),Research methods in language variation and change. Cambridge, UK: Cambridge University Press.

Sprouse, J., Schütze, C., Almeida, D., 2013. A comparison of informal and formal acceptability judgments using a random sample from Linguistic Inquiry 2001-2010. Lingua 134, 219-248.