LingLunch: Thierry Nazzi (INCC, CNRS – Université Paris Cité)

A crosslinguistic perspective on the development of phonological biases in lexical acquisition.

The sound inventory of all languages contains consonants and vowels. Nespor et al. (2003) proposed a division of labor according to which consonants are given more weight in lexical processes, and vowels in prosodic/syntactic processes. The proposal of a consonant bias (C-bias) in lexical processes led to many studies. While a C-bias was mostly found in adulthood (see Nazzi & Cutler, 2019, for a review), the crosslinguistic picture from developmental studies suggests a more nuanced picture. The present talk will present an overview of the studies conducted in infancy and toddlerhood. First, it will summarize two decades of studies conducted in French, that assessed when the C-bias is first observed in development, and then started testing whether the C-bias is an innate bias, or an acquired bias, and if so whether it is learned in link with phonological or lexical acquisition. Second, it will give an overview of the studies conducted in other languages, in order to assess whether or not the trajectory found in French generalizes to other languages, and if not, the linguistic properties likely to affect this trajectory. From this, we will propose a new model of the acquisition of the phonolexicon, the Sound-of--words model, taking into account the current state of the art on this issue, and linking this acquisition to linguistic properties of the native language(s), input properties of parental/environmental speech, and later vocabulary development.