Language and Linguistics - Volume 27, Issue 1, 2026

This paper has four goals. (1) We argue that the non-local antecedent of the long-distance reflexive mình is either an attitude holder bearing a de se belief, or an empathy locus in the event/state that the utterance speaker identifies with (Kuno & Kaburaki 1977; Kuno 1987; Huang & Liu 2001; Wang & Pan 2015; Charnavel 2019b). (2) We show that Huang & Liu’s (2001) analysis based on direct discourse representations fares better than the generalized discourse requirement in Cole et al. (2006) to characterize the distribution of the blocking effect of mình because first/second-person pronouns induce blocking only when they occur within the relevant direct discourse representation for the intended long-distance interpretation of mình. (3) We propose that the blocking effect of mình can be lifted with proper contextual support, invoking the Topic Empathy Hierarchy (Kuno & Kaburaki 1977; Kuno 1987) to override the Speech-Act Empathy Hierarchy and neutralize the perspective conflicts induced by a first/second-person pronoun. (4) We posit that the dependency between mình and its non-local antecedent is mediated by an implicit logophoric operator/pronoun in Charnavel’s (2019b) two-stage model of binding of long-distance anaphora (Nishigauchi 2014; Sundaresan 2018; Ikawa 2024; Baker & Ikawa 2024). We show that the existence of the implicit logophoric operator/pronoun not only captures the hybrid syntactico-pragmatic properties of Vietnamese mình, but also plays a pivotal role in explicating the interpretative possibilities of multiple occurrences of mình.

The phonetic radicals embedded in Chinese ideophonetic compounds serve as indicators of the ideophonetic compounds’ pronunciations. The mainstream research, therefore, has focused on the phonological activation of the phonetic radicals and their contribution to the host ideophonetic compounds’ phonological processing. The present study, however, by employing a primed part-of-speech judgment task and manipulating the phonetic regularity variable, examined phonetic radicals’ semantic, instead of phonological, activation as well as the role of phonetic regularity in the processing of the host ideophonetic compounds. These results endorsed the presence of the semantic activation of phonetic radicals, refuting the influence of phonetic regularity on their activation. Conversely, they affirm that phonetic regularity could modulate the semantic activation of the entire ideophonetic compound.

In this experimental study, we explore the syntactic constructions of left-node-raising (LNR) and right-node-raising (RNR) in Korean, examining whether LNR functions as a mirror image of RNR, as suggested in previous work by Yatabe (2001), Nakao (2009; 2010), and Chung (2010). Our findings from formal experiments indicate the existence of a unique left/right-node-raising (LRNR) construction in Korean, which we propose as a true mirror of RNR. We analyze RNR and LRNR through the framework of multidominance theory, drawing specifically on Wilder’s (1999; 2008) in situ linearization approach and Citko’s (2005) movement-based linearization approach, which provide insights into how shared elements (i.e. pivots) in these constructions are interpreted syntactically and semantically across conjuncts. For the derivation of LNR, however, we propose an alternative approach, namely, a scrambling-plus-pro analysis. Our experiment centers on acceptability judgments of sentences featuring dependent-plural markers (DPMs) within pivots across these constructions, revealing that DPMs influence acceptability differently in each structure. Specifically, DPMs mildly impact the acceptability of RNR but more strongly affect LRNR, which we attribute to distinct linearization schemes in each construction. In contrast, the presence of DPMs within pivots severely degrades the acceptability of LNR. Our study concludes that RNR and LRNR are more effectively explained by multidominance theory with in situ and/or movement-based linearization, while LNR aligns better with the scrambling-plus-pro analysis due to its unique derivational properties. Thus, our findings support the claim by Kim et al. (2023) that LNR is not a direct mirror of RNR.

This paper reviews Yang’s (2024) analysis of the sentence-final negative WHAT in Chinese and argues that the sentence-final negative WHAT, which can be classified into two types — emphatic negation and quotative negation — should be analyzed as a suffix attached to the predicate in morphology. Specifically, the affixation of the sentence-final negative WHAT is a morphosyntactic operation that occurs in l-syntax, creating a complex word with the predicate. Additionally, both sentence-final negative WHAT and its reduplicated form should be analyzed in a unified way within morphology, with the base being as short as possible phonologically. The study also emphasizes the need for a negative force projection at the sentence periphery.

The purpose of the paper is twofold: (1) to argue that one type of the Mandarin shou construction can be treated as a new form of passive construction (the shou-passive); and (2) to explore how it differs from the Chinese bei-passive. In previous studies, the shou construction has been treated as a transitive structure in which shou ‘receive, endure’ takes an event nominal as its complement. This study reveals that another type of the shou construction — the shou passive — has emerged, in which shou takes a verbal complement involving movement. Comparison with bei-passives shows that the shou-passive can also be divided into two types — short or long passives — while exhibiting two distinct properties. Shou-passives reject idiom chunk subjects and do not select an Experiencer. These distinctions result from the lexical composition of shou (i.e., Become and Have/Undergo) and the early stage of grammaticalization, which requires every component to be projected. The similarities and differences suggest that the grammaticalization of shou-passives is underway, but not yet completed.