Document Type

Conference Proceeding

Publication Date


Journal Title or Book Title

Congrès Mondial de Linguistique Française (CMLF 2010)


Publisher's PDF

Publisher's Statement

© Owned by the authors, published by EDP Sciences, 2010




The aim of this study is to contribute to the current debate on the nature of morphosyntactic representations in young children which opposes two hypotheses. On the one hand, the so-called "lexicalist" or "constructivist" approaches suggest that the young child is sensitive to the structures he frequently hears and that the first constructions he represents are based on combinations of specific linguistic elements. understood combinations of inflected words and marks. On the other hand, generative theories give a lesser role to the linguistic environment and stresses the importance of the capacities of young children to form abstract morphosyntactic representations which do not systematically reflect the frequency of combinations in the linguistic environment and which apply to lexical categories (eg Noun, verb etc) and are not limited to colloquial words. The analysis of the spontaneous production of young children is the traditional method that has been most often used to test these hypotheses. The study presented below tests the two hypotheses at the heart of the current debate by using two other complementary experimental approaches - the Head Turn Paradigm and the Intermodal Paradigm of the Preferential Gaze - which make it possible to study the preference and the understanding of the chord. subject-verb in native French speakers between 14 and 30 months. The data obtained on 88 children between 14 and 30 months were analyzed. The verbs and constructions used in the experiments are also the subject of detailed quantitative analyzes of the language to which young children are exposed. These analyzes covered a total number of 54,000 statements. These three sources of data on the preference and understanding of agreement in young children and the properties of the linguistic environment allow us to contribute to the current debate on the nature of early morphosyntactic representations. The results of the three studies reveal a) that 18-month-old French speakers prefer grammatical constructions that involve a Nominal Syntagma and an irregular verb in the third person singular and plural, b) that at 30 months children understand constructions that contain a third-person clitic subject and a verb that begins with a vowel, in the singular and in the plural and c) that these two results do not directly reflect the statements to which the children are exposed in which the forms and constructions tested are not frequent and which reveal a significant asymmetry between the singular and the plural. These results do not seem compatible with the so-called “lexicalist” or “constructivist” hypotheses according to which the performance of children of these ages should reflect the combinations and forms frequent in the linguistic environment. This study allowed us to better understand the morposyntactic capacities of the young child. The preference data are compatible with those published on English and German, but our study makes an additional contribution because the forms used in our experience are not regular.