Doctoral Dissertation Research: Licensing Conditions for Long Distance Binding in Child Language

University Of Hawaii, Honolulu

Investigators

Abstract

This doctoral dissertation project investigates an apparent delay in children's learning of "long-distance binding" (LDB), a phenomenon whereby an element such as a pronoun is grammatically associated to its antecedent, from which it may be separated by several words within or across sentences. Delays in the acquisition of certain aspects of language are thought to reveal something about how children learn language, and something about the nature of human language itself. This project shows that, contrary to previous literature assuming delayed learning of LDB in the language of interest, children are in fact not delayed with respect to LDB, and the reason previous studies found evidence for a delay is that children's understanding of the verbs that were used (such as 'think', 'believe', etc.) differs from that of adults. This difference in understanding leads to the illusion that the learning of LDB is delayed. This project uses experimental judgment tasks to investigate children's understanding of LDB while controlling for the problematic verbs, shedding light on children's language learning process. The project also includes an undergraduate researcher, contributing to the training of the next generation of language scientists. LDB is investigated in a target language where a reflexive word meaning 'self' can refer to nouns that occur earlier in the sentence. The experiment tests whether children can understand that in a sentence equivalent to English 'John thinks that Mary brushed [self]', 'self' can refer to Mary (the local antecedent) but also John (the long distance antecedent). Children have been found to strongly prefer the local antecedent, while adults allow both antecedents. The experiment finds that in order for the long distance antecedent to be licensed it must be logophoric, which means it is the source of belief or reported information. In a sentence like 'John thinks that Mary brushed [self]', 'John' is considered logophoric, and thus is an eligible antecedent to 'self'. However, children fail to assign the first noun as logophoric because of their difficulty in acquiring the correct properties of the verb 'think'. This experiment replicates previous findings that children struggle with LDB in such sentences. It then includes a manipulation that highlights the logophoric meaning of 'think', thereby forcing children to consider the first noun phrase as logophoric. Pilot studies find that in the first experiment (no logophoric manipulation), children fail to allow the LDB interpretation, while in the second experiment (highlighting logophoricity), children do allow the LDB interpretation. This therefore shows that the locus of difficulty is not the computational complexity involved in LDB, but rather is related to children’s acquisition of verb like 'think' and similar verbs denoting mental states. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.