This study investigates the lexical low tone (L) and its tone sandhi patterns in Laoling, a Northern Mandarin dialect. Cao (2007) observed that disyllabic L+L sequences consistently map the initial L to LM, while trisyllabic sequences only do so when the sequence is right-aligned, regardless of morphosyntactic boundaries. Traditionally, tonal dissimilation has been cited as the motivation for L+L tone sandhi in Mandarin dialects. However, this explanation fails to account for the exclusive phrase-final dissimilation in Laoling. Based on F0 data, this study proposes that stress-conditioned tonal assimilation underlies this phenomenon, that is, the elevated tonal onset of a stressed L tone raises the tonal offset of the preceding L tone. A constraint-based analysis is proposed to formally capture the interaction between prosody and the L tone sandhi in Laoling.
Ettehkey 'how' in Korean can be interpreted as either causal, manner, or resultative how according to its "height of interpretation" (Hacquard 2006). Interestingly, it is immune to strong island effects when interpreted as manner or resultative, whereas it is island-sensitive when interpreted as causal. This observation presents a challenge to the Noun versus Adverb Generalization of wh-in-situ construals (Tsai 1999a; Fujii et al. 2014), since ettehkey does not seem to contain a nominal head. In this paper, it is argued that the difficulty is only apparent, and the unselective binding analysis can be extended to Korean how-questions with manner and resultative ettehkey being decomposed into a couple of functional elements such as an indeterminate noun (Kuroda 1965), whereas Korean causal how has evolved into a genuine adverb in the left periphery. We also draw cross-linguistic evidence from Chinese and Japanese to further support this novel approach in combining the insights from Syntactic Cartography and Distributed Morphology.
This study investigates the bidirectional second language acquisition (L2) of the Applicative Double Object Construction (ADOC) in Chinese and Spanish. Although Chinese and Spanish ADOCs differ morphologically-Spanish features a concrete applicative head and a dative marker for the indirect object, while Chinese lacks these-both languages allow the Source as an applicative argument. L2 acquisition research generally suggests that L1 structures similar to the target language (TL) facilitate L2 acquisition, while differences hinder it. However, this study shows that the Source applicative argument in the Chinese ADOC only partially transfers to L2 Spanish and that the structural similarity between Chinese and Spanish offers minimal facilitation for Chinese speakers learning the Spanish ADOC. In contrast, the Source applicative argument in Spanish fully transfers to L2 Chinese for Spanish speakers. These findings support Yuan's (2001) claim that L1 transfer is relative. Drawing on Yuan's (2015) derivational complexity measure for Chinese, this study extends Jakubowicz's (2005, 2011) Derivational Complexity Hypothesis to L2 acquisition, illustrating how computational complexity influences L1 transfer patterns.
This study employs conversation analysis to investigate how tutees initiate repairs during Mandarin Chinese tutoring sessions, focusing on their use of candidate other-initiated repairs (OIRs) to address intersubjective issues. By using candidate OIRs, tutees actively propose solutions, showing greater agency in overcoming interactional challenges effectively while collaborating to complete tutoring tasks. The analysis identifies three main types of interactional issues addressed by candidate OIRs: (1) task details, (2) the scope of questions, and (3) specific linguistic elements. Each type of issue is managed using distinct turn designs within the local context. Ultimately, candidate OIRs allow tutees to reshape the trajectory, fostering a more collaborative exchange. This process highlights how participants co-construct sequences to address interactional issues, providing insights into learner-initiated repairs, intersubjectivity, and active participation in Mandarin Chinese tutoring sessions.
This study investigates the distinction between English middle constructions in determinative and appositive relative clauses. The objective is to examine the behavior of middles within subordinate clauses by analyzing their sentential positions and structural characteristics to identify influencing factors. The presence of an arbitrary agent and the pragmatically oriented structure of middles were found to influence their sentential positions. Data from the COCA corpus reveals frequent occurrences of middles in both types of relative clauses, particularly sentence-initially and sentence-finally, with a higher frequency at final positions. Determinatives, likely indefinite, are more common in final positions, while appositives, likely definite, occur frequently in both initial and final positions. The study concludes that the appearance of these constructions is primarily influenced by information structure, definiteness, informativeness, at-issueness, and clausal structure. Additionally, informativeness and at-issueness were found to affect the obligatoriness of determinative relative clauses. The distinctive feature of middles lies in the arbitrary agent, which not only facilitates their appearance sentence-initially but also shapes the context. This study is the first comparative study of relative clauses with a specific sentence structure.
