IRCS Technical Reports Series

Document Type

Thesis or dissertation

Date of this Version

January 2000


University of Pennsylvania Institute for Research in Cognitive Science Technical Report No. IRCS-00-05.


The goal of this dissertation is to account for the fact that young children acquiring English (around age 2 years) often produce utterances like (1), in which they omit a form of the copula, be.

(1) I in the kitchen. (cf. I am in the kitchen)

Children’s production of forms like (1) is interesting for two main reasons: firstly, utterances like these do not occur in the input (adult English); secondly, children’s omission of the copula adheres to a systematic pattern (their omission is neither across the board, nor haphazard). In particular, children do not omit the copula in utterances like (2).

(2) He’s a dog. (@He a dog)

The difference between the constructions in (1) and (2) can be characterized in terms of a difference in the sorts of properties denoted by the respective predicates: a location such as in the kitchen is a temporary property of the subject; a predicate such as a dog denotes a permanent property of the subject. I argue that these predicates differ from each other both semantically and syntactically: "temporary" (stage-level) predicates contain additional functional structure (an AspP) that "permanent" (individual-level) predicates lack. Crosslinguistic support for this proposal is provided.

As for why children acquiring English ever produce forms like that in (1), I link this to the fact that non-finite main clauses are permitted in child English. I define finiteness in terms of a binding relation between an abstract Temporal Operator (TOP) and a functional head in the structure. A main clause is finite only if Infl is bound by TOP in CP. Certain grammars (among them child English) have the option that TOP may bind Asp, if Asp is projected in the particular clause. However, this binding relation does not result in the clause being finite. Since Asp is projected in clauses with stage-level predicates, but not in clauses with individual-level predicates, it follows that stage-level predicates may occur in non-finite clauses while individual-level predicates occur with a finite clause. Coupled with the hypothesis that an overt copula is finite (it is inflected over 99% of the time) and an omitted copula indicates non-finiteness (independent support is provided), the pattern of copula omission and production in child English is accounted for.



Date Posted: 10 August 2006