- Principle B, VP ellipsis, and interpretation in child grammar by Rosalind Thornton, Kenneth Wexler
This book contains an investigation of a specific aspect of child language acquisition, children’s interpretation of pronouns. Working within the generative tradition of the principles-and-parameters framework, the authors adopt the general path to children’s nonadult linguistic behavior provided by (an adapted version of) the strong continuity hypothesis: Adult and child grammars can only differ in as far as the grammars of different languages differ from each other—coupled with the assumption that children’s real-world knowledge is not fully developed. Thornton and Wexler take the properties of the parser to be identical for adults and children, that is, full access to a sentence-processing mechanism is available to children, and it is uniform across the species.
Specifically, this study concerns children’s misinterpretation of pronouns that appears to be subject to Principle B of the binding theory and offers an account in terms of (lack of) pragmatic knowledge that follows in the footsteps of previous research on Principles A and C; as such, it shows that Principle B does not constitute a learnability puzzle, and children ‘know’ all grammatical properties of binding theory. It further extends the basic idea to VP ellipsis and the referential parallelism constraint. The upshot is that while children have all the hardware to parse and understand Principle B, they lack the full knowledge of the world available to adults—in other words, misinterpretations result from pragmatic differences rather than syntactic or biological ones.
Beyond an analysis and explanation for the phenomena studied, this book offers a detailed presentation of the carefully designed experimental investigation the authors conducted. As such, it also serves as another demonstration of the suitability of the Truth Value Judgment task to elicit child data as well as to investigate the constraints of particular principles (such as binding theory) and how children’s grammars may differ from the target grammar.
In Ch. 1, ‘Introduction’ (1–22), Thornton and Wexler lay out the theoretical framework adopted, the principles-and-parameters model of grammar, with particular reference to issues [End Page 322] relating to children’s acquisition of language. They follow the approach to children’s nonadult behavior under which children differ from adults in pragmatic (‘real-world’) knowledge; all principles of universal grammar (UG) are intact and operational (Chien & Wexler 1990). They also briefly introduce the role of positive and negative evidence in child language research (with emphasis on relevant to binding theory), summarize basic patterns of interpretation of pronouns by children, and point to further issues regarding pronouns and antecedents, theory- and acquisition related. The end of this chapter presents a useful overview of the structure, set-up, and content of this book.
To set the stage for T&W’s own proposals, Ch. 2 discusses the literature, experiments, and results so far in ‘The acquisition of Principle B’ (23–88), focusing primarily on the findings by Chien and Wexler 1990. The basic observation is that some children allow a reflexive interpretation of the pronoun in sentences like Mama Bear washed her but not if the subject is quantificational (such as Every bear washed her), in apparent violation of Principle B. Chien and Wexler hypothesized already that children lack certain pragmatic knowledge responsible for this interpretation. T&W lay out the particulars of this hypothesis and present five alternative accounts, as varied in nature as finding an explanation in terms of processing, inadequate experimental design, perceptual differences, difficulties in understanding other speakers, intentions, and purely syntactically. Beyond a mere presentation of these previous studies, T&W point out problems or issues to answer each one of these approaches. The picture that emerges is that none of these accounts can satisfactorily explain the state of affairs.
‘A new account’ (89–130) is presented in Ch. 3, taking into consideration VP ellipsis. The guiding assumption is that children’s misinterpretations of certain pronouns are not violations of Principle B but that the reason lies outside the computational...