CNL Lunch

Graciela Tesan & Rosalind Thornton

Negation in Child Language: What's in Children's Heads?

Thursday January 31st, 12:30pm, 3416 Marie Mount Hall

Two-year-old children’s production shows a rapid development of inflectional morphemes. However, omission of those morphemes is also characteristic of two-year-old speech (Wexler, 1994; Harris and Wexler, 1996). In this study, we focus on the development of the Simple Present (SPr) inflectional morpheme –s and its relative, the dummy verb do. We will probe into the apparent symbiotic nature of their relation by examining negative sentences from elicited production and spontaneous speech.

Our findings are based on longitudinal data from four children. First, we will suggest that the emergence of do-support in child language derives from the morpho-phonological requirements of the NEG head (Schutze, 2001). Our second finding deals with the morphological status of the elements NEG and INFL in child language. We claim that children try out different morphological hypotheses as a response to the morphological needs of these categories without violating UG. We observe grammars that generate medial negation (1) ‘he not goes there’, which realize inflection on the verb in the presence of a NEG marker. This non-adult option suggests that children initially hypothesize (a) the SPr morpheme is a phonologically weak unbound morpheme, and (b) the negative marker not is a specifier.
Once do-support is an option in child grammar, the morphological status of INFL will change to that of a bound morpheme and the NEG marker is reanalyzed as head. The dummy verb do thus becomes an auxiliary verb that can host NEG and INFL heads and the morphological alternative (1) disappears.

Our results suggest that the functional categories that make up the grammar of the target are present early in language development. A big part of the language-learning task consists in determining the right mapping of vocabulary items that instantiate those categories. Figuring out the mapping is not straightforward in English, because the input is “messy”. It is thus expected that access to the relation between form and meaning/sound be not direct for the learner.