Judges’ Queries and Presenter’s Replies

  • Members may log in to read judges’ queries and presenters’ replies.

Members’ Comments

  • Members may log in to read comments.

Icon for: Lisa Hsin

LISA HSIN

Johns Hopkins University

Abstract

LISTEN
This content requires Adobe Flash Player. Download Flash Player or Download Audio

Accelerated Acquisition in Early Bilingualism

Bilingualism is a compelling object of investigation for theories of language acquisition, as both marked advantages and apparent delays color the cognitive development of bilingual children. In the domain of syntactic development, cross-linguistic interference is commonly found in these children’s early speech, even though researchers agree the children start with two separate grammars. Usually, a bilingual child acquires the target uses of a given utterance type in one language later than a monolingual child would, while in some cases bilingual children systematically produce word strings that monolinguals rarely, if ever, produce.

Previous studies have proposed that morphosyntactic cross-linguistic interference occurs at the syntax-pragmatics interface when there is an overlap of linear order for some construction in both languages (e.g. wh-questions or object-drop). But in this study, we present novel data of spontaneous child speech showing that bilingualism can also accelerate first language acquisition of syntax: English-Spanish bilingual children produce virtually perfect wh-questions in both languages from the earliest stages of development, while English-speaking monolinguals make frequent errors of auxiliary omission as late as age 3.

We propose the Transfer/Conflict Hypothesis to account for this difference, from which we argue that the conditions on interference and facilitation are separate: interference does indeed arise from conflicting pragmatic constraints, but facilitation arises out of hierarchical syntactic structural transfer. We discuss other transfer phenomena and their relevance to this new hypothesis, in addition to planned computational models, and behavioral experiments which will test our hypothesis using controlled comprehension and production tasks.