The development of negation in early child language

Abstract

The development of early logic in the use of language is investigated in this thesis by studying the acquisition of negation in the language of children. In particular, the use of negation to express judgements about the truth-value of statements made by others is viewed as a linguistic manifestation of early logical abilities. The central hypotheses of the thesis are 1) the logical operation of negation to deny statements (truth-functional negation) becomes evident near the onset of syntax in the language of the two year-old child; 2) the developmental sequence for the emergence of semantic functions which negation serves progresses from the expression of internal states (rejection), to comments on external states (e.g., object disappearance), to comments on correspondences between external states and language use (truth-functional negation); and 3) the ability to judge negative statements as true is a later development than either judging misnamings to be false or judging another's denials to be incorrect, i.e., there is an interaction between PRAGMATIC factors (where the judged utterance is appropriate, given prior discourse) and SEMANTIC factors (where the judged utterance is correct, given specific truth-conditions), with an early overemphasis of pragmatic factors. Hypotheses (1) and (2) were supported by the studies of very young children reported here, while evidence for (3) was only suggestive. These findings support several claims about early linguistic competence: First, that the child has abstract, reflective knowledge of at least one major feature of language, that it is a system relating to the world with logical structure, and second, that the child is nonegocentric in certain aspects of language use in that he or she evidences knowledge of truth conditions on the use of language which involve not only his or her own language but that of others.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 159 citations exist upstream.