Conditionals as exemplified by English if-clauses are constructions that relate an antecedent and a consequent and convey typically that all (relevant) situations that verify the antecedent verify the consequent as well. This basic meaning is compatible with many different relations between the propositions expressed by antecedent and consequent. In English, conditionals of the form if p, q are exceptionally unconstrained in terms of what type of relations they can express, while crosslinguistically many languages have a more intricate inventory of conditional expressions, which may impose unique restrictions on the types of relations they are used to express. My dissertation project contributes to this line of inquiry by addressing the following two research questions:

(i) crosslinguistically, how do conditional marking strategies contribute to an identifiable core of conditional meaning (roughly, “all antecedent cases/situations are consequent cases/situations”)?

(ii) what are the “additional” pragmatic inferences, flavors, or other types of conventional meanings carried by individual conditional expressions?

Conditionality without if : Mandarin conditionals and jiu. SURGE reading group, Rutgers University. [handout]

A major domain of my research lies in is the pragmatic effects of various linguistic expressions, as well as their interactions with speech acts/clause types. One central concern in Gricean pragmatics is how interlocutors draw inferences from each others' utterance choices. I explore this research question by investigating discourse particles and how they may affect the conventional discourse effects of their anchors.

1. Particle ne. This project investigates the behaviours of the particle -ne in the sentence-final position along with its interactions with different clause types in Mandarin. I propose that -ne signals that the speaker believes that the current discourse move she makes is not optimal for the addressee. The notion optimal discourse moves is rooted in Gricean pragmatics: cooperative interlocutors are expected to only make optimal utterances (or utterances which obey general Gricean principles) to each other (cf., Lauer 2013; Portner 2004, 2007). Moreover, the proposal challenges the view that sentence-final ne is a CT marker (Constant 2014), which is shown to make incorrect predictions of the distribution of ne.

Updating unexpected moves. SuB 27. [slides]

2. Particle ba. While traditionally, ba has been taken to contribute to the so-called “weakening effect” in assertions, it is sometimes mentioned that ba generates an opposite “enhancing effect” in questions. I demonstrate that the “enhancing effect” of ba stems from its discourse function of challenging the presuppositions of the question to which it is attached. I also explored the pragmatic effects of ba in assertions and conditionals, and propose that the seemingly contradictory inferences can be unified if we take the core discourse function of ba to be contributing an implication that the speaker does not believe that the conversational goal in certain context can be achieved.

A discourse model for Mandarin ba-interrogatives. [slides]. SuB 24 & GLOW in Asia XII

Extracting commitment: the case of Mandarin rising ba-declaratives. [paper]. SuB 25.

Establishing discourse relations: two contrastive markers in Mandarin. [slides]. CLS 58, GLOW 45, TLLM 2022.

While imperative/jussive clauses are known to have interaction with (null) subjects, verbal morphology, and clause embeddability as well as speaker/addressee projections, whether objects interact with jussive clauses is, however, less understood. This study reports such a case of interaction with objects, which is observable in a particular movement context. The core data comes from non-agreeing resumptive pronouns (NRPs) in Mandarin Chinese. The NRP exhibits a multifaceted empirical profile that involves (i) licensing by jussive clauses, (ii) patient roles of objects, and (iii) movement-derived properties. We argue that the intricate pattern can be accounted for by an Agree relation between the NRP and jussive head, coupled with interface conditions on partial Copy Deletion. This account sheds light on how clause types (i.e., jussives) interact with argument structure.

Jussive agreement with non-agreeing resumptive pronouns in Mandarin Chinese with Ka-Fai Yip. [handout]. BCGL16.

Agreement in imperative clauses: evidence from object resumptive pronouns in Mandarin Chinese with Ka-Fai Yip. [handout]. NELS 54.

In this project, we argue for the existence of defocus in our grammar, which is construed as a systematic resistance of alternative-based focus interpretation. We show that in Mandarin and Cantonese, RD chunks are structural manifestations of defocus. We conducted acoustic production experiments and found that there is a syntax-prosody mismatch in RD constructions in these two languages: they are syntactically biclausal, but prosodically monoclausal. We propose that this mismatch is due to an intricate interaction between the existence of DeFocP in syntax and the ranking of DeFoc constraint in prosody.

In the future, we plan to explore the cross-linguistic syntax-prosody patterns of RD constructions. The current defocus rephrasing view predicts a factorial typology of RD constructions with respect to two parameters: whether there is an obligatory DeFocP in syntax, and whether the DeFoc constraint is ranked higher than Match in prosody.

Defocus leads to syntax-prosody mismatches in right dislocated structures with Ka-Fai Yip. [slides]. CLS 59.

Right dislocation, defocus, and variations in syntax- prosody mapping with Ka-Fai Yip. [slides]. phex 15.

Teasing apart the prosodic effects of focus and of defocus: syntax-prosody mismatches in right dislocation with Ka-Fai Yip. LabPhon 19.

In collaboration with Yusuke Yagi, we explore the crosslinguistic picture of concessive conditional expressions (e.g., even if in English), particularly on Japanese mo and Mandarin ye. Motivated by the fact that crosslinguistically many languages adopt additive particles to express concessive meaning in conditionals, we propose to establish the connection between additivity and concessivity: concessivity is derived from a stronger additivity which requires an assertion to be not inferrable from the context. We also explored the parallel between Japanese mo and Mandarin dou/ye.

Stronger Additivity Derives Concessivity with Yusuke Yagi. [slides]. PLC 46 & TaLK 2022.

Additive Prejecent and/or Additive Alternatives: A Principle and a Parameter in Mandarin and Japanese with Yusuke Yagi. [slides]. GLOW 45.

In this ongoing project with Hiroaki Saito, we look into the behaviors of Mandarin speech verb shuo ‘say’ both in matrix clauses and embedded contexts. We showed that when shuo occurs on the edge of a matrix clause, it expresses an additional reportative evidential meaning. We thus argue that this matrix shuo is located in a syntactic position (EvidP) different from the pure complementizer shuo. Currently, we are interested in the interactions between attitude verbs and the complementizer shuo, which has been assumed to be semantically vacuous in embedded contexts (see also Huang 2018).

On matrix shuo in Mandarin with Hiroaki Saito. [paper]. NACCL 32.