Do data from children with specific language impairment support the agreement/tense omission model?

Abstract

One of the most influential recent accounts of pronoun case-marking errors in children's speech is C. T. Schütze and K. Wexler's (1996) agreement/tense omission model (ATOM). This model predicts that non-nominative subjects with agreeing verbs will be so rare that they can be reasonably disregarded as noise in the data. The present study tested this prediction on data from 4 children with specific language impairment (SLI) by comparing the frequency with which each child produced non-nominative subjects with agreeing verbs and the frequency with which they would be expected to produce such errors by chance given the number of nominative and non-nominative subjects and the number of agreeing and nonagreeing verb forms in the data. The results show (a) that although 3 of the 4 children used non-nominative subjects in their speech, only 2 of them (Nathan and Dan) produced non-nominative subjects with agreeing verbs significantly less often than one would expect by chance; (b) that to the extent that there was an asymmetry in Nathan's and Dan's use of nominative and non-nominative subjects with agreeing verbs, this asymmetry could be explained in terms of developmental changes in their ability to mark case and agreement correctly and the use of potentially unanalyzed contractions; and (c) that Nathan and Dan produced non-nominative subjects with noncontracted agreeing forms about as often as one would expect by chance. These findings are discussed in terms of their theoretical implications for the ATOM and their methodological implications for future work on patterns of pronoun case marking error in the speech of typically developing children and children with SLI.