A number of recent experimental works on the nature of island effects have focused on the role of syntactic and processing factors, but little attention has been paid to pragmatic factors. The main purpose of this paper is to test the role of a pragmatic constraint, called the Characterization Constraint (CC), in English relativization. Based on an acceptability judgment task conducted on two types of relative clause constructions (RCCs), it is claimed that CC plays a significant role in constraining relativization in English. In addition, the paper further discusses two interesting issues: (1) difference between the two types of tested RCCs in terms of acceptability and (2) difference between Korean and English with respect to the degree of influence of CC in relativization.

Projection has been extensively used as a diagnostic for presupposition, but many utterance classes generate projection even if they do not have standard features of presupposition. The essential property of all projective meanings is that they are not-at-issue, characterized as a connection to the question under discussion. Tonhauser et al.(2013) propose to classify projective content based on two criteria: (i) whether they are subject to the strong contextual felicity (SCF) constraint, i.e., whether they must have been part of the interlocutors' common ground prior to the utterance, and (ii) whether they have the obligatory local effect (OLE), i.e., whether their projection is necessarily filtered by a belief predicate. Instead of OLE, I propose the independence as a better criterion: a projective content p of a sentence S is independent from the at-issue content of S iff p's truth or falsity has no effect on the at-issue content of S. The Potts' CIs share with the prejacent of only and the polar implications of approximatives the feature of logical independence from the at-issue content. In addition, it seems sensible to assume that projective meanings and their classifying criteria display a gradience, with some projecting much more than others.

This paper investigates a certain variability of island-(in)sensitivity in Right Dislocation (RD) in Korean. I suggest that the variability arises due to the processing mechanism that demands the parser process the sentence as soon as possible to minimize processing load in consistent with the Minimal Attachment Principle (cf. Frazier and Fodor 1978; Fodor and Inoue 1998, 2000; Bae and Park 2018). Extending the principle, I argue that the appendix tends to be processed with the linearly closer, potentially matching predicate, and that in certain contexts it results in wrong processing, giving rise to the island effect. I also suggest that the processing account can be compatible with the ellipsis approach to RD by showing that when the processing mechanism becomes ineffective, the ellipsis effect emerges. This in turn supports the claim that island violations can be repaired by ellipsis.

Null argument phenomena have been discussed roughly in two aspects: distribution and interpretation. A pro analysis of null arguments has been challenged because of null argument examples that may exhibit sloppy interpretation. In this paper, we pay attention to the fact that in the environment where sloppy(-like) interpretation of null arguments is observed, overt pronoun can also yield the same interpretation. More specifically, we show that our version of pro analysis is desirable on both conceptual and empirical grounds. Our analysis is conceptually simple because under the assumption that pro is equal to overt pronoun, we can account for the interpretation of both null arguments and overt pronouns in a parallel way. Our analysis is also empirically superior in explaining null CP asymmetries and resolving wh-puzzles.

In the standard Korean subject honorification, an honorific si honors a human surface subject. However, the existence of si, which is now being interpreted to honor a nonhuman subject, has been causing controversy in recent years. Quite a few Korean linguists analyze this as a strategic usage and call it the hearer honorification, which is to honor the hearer, not the subject, whereas the National Institute of Korean Language and the media define it as a serious misuse. In this situation, based on the results of a survey on the nonhuman surface subject honorification (NSSH), we have proposed that a clear intuition on the contrast between proper and improper uses of NSSH is maintained (Yoon 2017). Given this, one purpose of this paper is to investigate various ways of proper indirect licensing of si in NSSH based on experimental results that confirm the validity of the proposed rules of indirect licensing of si by both older and younger generations. More importantly, another purpose is to demonstrate the scalarity of indirect licensing of si, which will be proposed to be the main cause of the current and slightly confusing situation regarding the use of si.