Head movement and allomorphy in children's negative questions

Abstract

English-speaking preschoolers occasionally produce negative questions with a "doubled" auxiliary (e.g. Why did you didn't know?). These 2AuxQs apparently involve a failure to raise [_NEG n't] to C (cf. Why didn't you know?). I analyze 2AuxQs as the product of two independent errors: a planning error (raising T-to-C without raising Neg-to-T first) and an allomorphy error (overgeneralization of n't). The planning error results from lack of practice: serial head-movement is relatively uncommon in English, and true Neg-to-T-to-C may be rarer than appearances suggest. In e.g. Why don't we play, ok?, -n't is not interpreted within TP - and strikingly, 2AuxQs are unattested here.