Current project

Exploring the Interface Effects on Pronoun and Reflexive Resolution: A Cross-Linguistic Bi-Directional L2 Acquisition Study

(Co-PI: Prof Antonella Sorace at the University of Edinburgh)

The goal of this three-year project is to conduct a bi-directional L2 acquisition study on the interpretation of pronouns and reflexives by native speakers of Mandarin, Japanese, Italian, and Korean, the L2 learners of Mandarin whose L1 is Japanese, Italian, or Korean, and the L2 learners of Japanese, Italian, and Korean whose L1 is Mandarin.  This project is an extension of my former MOST project of 2015-2017 on Mandarin/English bi-directional L2 acquisition.

On the part of reflexives, both Korean caki and Japanese zibun allow long-distance antecedent, which are similar to Mandarin ziji but differ from English and Italian where only the local c-commanding subject NP can serve as the antecedent for the reflexive.  In addition, Korean and Japanese also pattern like Mandarin in terms of disallowing the object NP to be the antecedent of the reflexive, in contrast to English and Italian where both the local subject and object NP can be the antecedent of the reflexive.  For pronouns, Mandarin, Japanese, Korean and Italian are all null-subject languages, but English is not.  However, there are some differences among the four null-subject languages.  First, Mandarin, Japanese, and Korean allow not only null subject but also null object, whereas Italian does not allow null object.  Second, Italian allows the backward anaphora interpretation but that reading is prohibited in some Mandarin and Japanese subordinate constructions but not others, and the pattern in Korean has not been investigated yet.  Third, although Italian has been claimed to show the effect of the Overt Pronoun Constraint (OPC) and the Position of Antecedent Strategy (PAS), Mandarin does not show those effects, and the patterns in Japanese as found in some previous studies were inconsistent with what OPC predicted.  As for Korean, there has not been systematic examination of its OPC effect yet, and whether Japanese and Korean show the PAS effect or not is unknown.

Using similar methodology as the English/Mandarin bi-directional L2 acquisition study, the project adopts the Truth Value Judgment Task for sentence comprehension to examine the interpretive patterns of the above-mentioned sentence types by native speakers and L2 learners in order to scrutinize the effects of structural and discourse constraints.  In addition, a working memory test and an attention test are also employed to investigate the interactions between the interpretation of anaphoric forms and general cognitive abilities related to cognitive control.

Recently completed project

Felicity condition and children’s knowledge of restrictive focus

The goal of this study is to investigate the development of the knowledge for restrictive focus through experiments.  Previous studies on children’s interpretations of sentences with restrictive focus predominantly found children assign non-adult interpretations to these sentences, regardless of the languages.  However, the error patterns and the degree of variation from adult interpretations were not consistent among those studies.  Moreover, the only study that examined children’s use of restrictive focus found Cantonese children started to use it as early as age 3.  This raises the possibility that the non-adult patterns found in those comprehension studies were in fact resulted from inappropriate experimental design which did not fulfill the felicity condition required for the use of restrictive focus.  Therefore, the study will include two parts.  The experiments were designed to incorporate the felicity condition in order to inspect children’s knowledge of restrictive focus for sentences with pre-subject ‘zhiyou’.  The findings from this study will provide important information in our understanding of relevant syntactic /semantic interface regarding restrictive focus in child language.