LINGUIST List 13.3414

Mon Dec 23 2002

Diss: Lang Acquisition: Anderson "The Acquisition..."

Editor for this issue: Karolina Owczarzak <karolinalinguistlist.org>

Directory

1. dla20, Lang Acquisition: Anderson "The Acquisition of Tough-Movement in English"

New Dissertation Abstract

Institution: University of Cambridge
Program: Research Centre for English and Applied Linguistics
Dissertation Status: In Progress

Author: Deborah L. Anderson 

Dissertation Title: 
The Acquisition of Tough-Movement in English


Linguistic Field: Language Acquisition

Subject Language: English

Dissertation Director 1: Ianthi M. Tsimpli


Dissertation Abstract: 

In the child language acquisition literature, it is traditionally
claimed that the acquisition of tough-movement or easy-to-please
constructions (hereafter, TCs) is problematic for child speakers of
English and that, consequently, adult-like grammatical knowledge of
these structures is late acquired. While Carol Chomsky (1969) was the
first to advance such a claim on the basis of her experimental
findings, it has been argued that the results of a number of
subsequent studies, involving both child speakers of English and
French, provide further support for the relatively delayed acquisition
of TCs in these languages. Nevertheless the precise reasons why
children find these structures so problematic have yet to be
identified.

Cromer (1970) was the first to propose that there are three clearly
definable stages in the acquisition of TCs, a view which is now widely
accepted. In the Primitive-rule Use stage, children are presumed to
lack the computational ability to interpret TCs, invoking instead a
default parsing strategy which assigns co-reference to the matrix
subject NP and embedded infinitive PRO; i.e. in the sentence A giraffe
is hard to kiss, a Primitive-rule User will interpret the sentence to
mean that the giraffe finds it difficult to kiss an unspecified
patient. In the Intermediate stage, children are observed to perform
in an inconsistent or random manner on experimental tests of TC
comprehension and therefore it has been suggested that they may
possess adult-like knowledge of the structural properties of the TC at
this stage but nevertheless remain uncertain of the precise
circumstances in which this knowledge is to be applied. Finally,
subjects in the third and final stage of acquisition are labeled
Passers and are proposed to have acquired full adult-like knowledge of
the TC.

In this thesis, we present an alternative account of the acquisition
of tough-movement which challenges the traditional representation of
the Intermediate stage of acquisition as one characterized by
intermittent or unreliable application of adult-like grammatical
knowledge of the TC. Instead, we present empirical evidence which
suggests that children in the Intermediate stage have the
computational ability to assign two distinct syntactic representations
to TCs, only one of which is acceptable in present-day English.
Additionally, we present evidence that the Intermediate stage of
acquisition is typically quite prolonged, which we suggest gives rise
to the widely-held impression that children struggle for some time to
master the exceptional control properties of tough predicates and the
details of their syntactic distribution. Instead we propose that
children's performance in this particular stage of acquisition is
better explained in terms of different rather than deficient
grammatical knowledge of TCs, with the interpretive options licensed
by the child's grammar initially representing a superset of those
licensed by the adult grammar.

Lastly, we develop a theoretical account of the Intermediate stage
which links the availability of an alternative, non-adult-like,
reading of the TC in child grammars with a syntactic option for tough
predicates that was licensed in previous stages of English. We
explore the possibility that children in this stage may license a
case-assignment alternation for the Experiencer argument of tough
predicates, which is not attested in the input, but which is
nevertheless consistent with options afforded by UG. The relative
delay that children experience in eliminating this interpretive
option, and acquiring a fully adult-like representation of TCs, is
proposed to stem from markedness considerations, specifically, that
the relatively restricted interpretive option for the TC attested in
present-day English represents the more marked situation
cross-linguistically.