A Cross-linguistic Approach to the 'Erroneous' Genitive Subjects: Underspecification of Tense in Child Grammar Revisited
Naoko Sawada and Keiko Murasugi
209-226 (complete paper
or proceedings contents
At around the age of two, when children produce non-finite verbs in matrix clauses (Root Infinitives (RIs) / Root Infinitive Analogues (RIAs)), non-Nominative subjects and copula-drop phenomenon are observed. Schütze and Wexler (1996) (S&W), among others, argue that the 'erroneous' Genitive subjects are produced when features in Tense (T) are not fully specified. This paper, based on the corpus analysis of Japanese-speaking children (0;6-3;8) and English-speaking children (1;6-3;5), argues that the 'erroneous' Genitive subjects are optionally produced also by Japanese-speaking children around the age of two, because (i) (the default) Root Clause is CP (Rizzi 1994) and (ii) the features [±N/±Gen] in T are not fully specified then. It thereby provides a theoretical account, following the lead of S&W, for the phenomenon observed across languages. It is finally argued that children retreat from the 'errors' when they discover (i) that sentential modifiers in NPs (with genitive subjects) are TP, but not CP, by setting the Relative Clause parameter (Murasugi 1991), and (ii) the difference between CP progressives and DP gerunds in English. Thus, the hypothesized connection among RIs/RIAs, Case errors, copula-drop and T-specification is corroborated.
Selected Proceedings of the 4th Conference on Generative Approaches to Language Acquisition North America (GALANA 2010)
edited by Mihaela Pirvulescu, María Cristina Cuervo, Ana T. Pérez-Leroux, Jeffrey Steele, and Nelleke Strik
Table of contents