EAPSI: Investigating the Chinese Bei Construction using Cross Linguistic Priming

Stabile Claire M, Honolulu HI

Investigators

Abstract

This award supports research that helps develop understanding of bilingual language acquisition ? a phenomenon that is growing as globalization increases. It extends the use of a relatively new technique (cross-language priming) to solve a troublesome theoretical problem. This research will help solve a long-lasting problem in Chinese syntax ? the structure of the bei construction. Many researchers believe that the abstract structure of the bei construction is the same as the English passive, while others believe it is a separate structure. Understanding the nature of this construction will aid in our understanding of bilingual language acquisition, and may improve language teaching methods. This research will be conducted at National Taiwan Normal University under the guidance of Professor Dong-Bo Hsu, a noted expert on priming methodology. Previous research has shown that the disputed Mandarin bei construction primes, and is primed by English passives; suggesting that the two patterns share an abstract structural representation. However, it is possible that this priming effect is due to priming of noncanonical word order (theme-before-agent instead of agent-before-theme). This research will test whether speakers of English and Chinese who are primed by sentences with noncanonical word order produce passives or bei constructions because of the effect of that word order. If the results of this experiment show no priming, then we can conclude that previous results showing priming of bei of and by the English passive do indeed indicate a shared abstract syntactic structure between the two constructions. This project will test 60 Mandarin-English bilingual children between the ages of three and five years, and a similar number of adults. Participants will be shown a picture which will be described using either an active sentence or on in which the argument order is noncanonical (e.g., topicalization or clefting). Participants will then describe a subsequent picture. Participants will be tested on both Mandarin prime-English target, and English prime-Mandarin target. If participants produce more passives in response to noncanonical sentences (rather than in response to actives), we can conclude that the priming effect found in previous work was indeed due to priming of a shared abstract representation - the Mandarin bei construction does indeed share an abstract structural representation with the English passive. These results can help to settle a long-standing debate by way of new psycholinguistic experimental methods. This NSF EAPSI award is funded in collaboration with the Ministry of Science and Technology of Taiwan.