Cognitive operations underlying semantic approximations produced by young children: A Lexical-Constructional approach to language acquisition and language teaching

Publié le 20 mai 2016 Mis à jour le 29 novembre 2016
le 23 juin 2016
14h - 16h
Salle D155 - Maison de la recherche

Lorena Pérez Hernández, Universidad de La Rioja - Logroño Séminaire CLLE ERSS

Despite their prominent role in language learning (Agustín Llach, 2005; Gaskell & Ellis, 2009; Hauser et al., 2002; Jiménez Catalán, 1992; Pinker & Jackendoff, 2005), semantic approximations in first language acquisition have received much less attention than in relation to second language learning (Agustin Llach, 2011; Singleton, 1999; Zimmerman 1986, 1987). Taken as sheer by-products of the process of learning a first language, most of the attention they have drawn has been devoted to their classification (Jaegger, 2005), the identification of those factors (i.e. conventionality, pre-emption, entrenchment and semantic fit, among others) that are involved in their retreat (Ambridge, 2013; Ambridge et al. 2013; Boyd & Goldberg, 2011; Goldberg, 2011), the tactics and strategies that may help children overcome them (Clark & Bernicot, 2008; Chouinard & Clark, 2003; Ramscar & Yarlett, 2007; Strapp & Federico, 2000), the exploration of the effects of their correction or the lack of it (Nooteboom, 1980; Verhallen & Schoonen, 1993), and their consequences on school performance and cognitive delay (Ghassabian et al. 2014).

To date, however, little attention has been paid to the factors that motivate those semantic approximations and the strategies that underlie their production. A recent attempt to do so can be found in Duvignau (2002, 2003) and Duvignau et al. (2007). These authors claim that semantic approximations in child language stem from analogic and pragmatic strategies, and help young speakers communicate their ideas and feelings by making up for their still immature lexical pools. In this, Duvignau's account complements previous research on the cognitive bases of semantic approximations in child language development (Bowerman, 1982; Clark, 1995; Dromi, 1987). In contrast to those semantic approximations emerging from the interferences of the speakers' native languages in second language acquisition, most of those included in our corpus of monolingual children are found not to interfere with communication, but rather to facilitate it. This particular status sets them apart from other types of random semantic approximations. In much the same vein, Iozzi and Barbieri (2009) carried out an experimental study from which it was concluded that in referential communication, preschoolers appear to produce 'nonconventional' messages involving analogies that allow them to succeed in communicating their thoughts in the absence of a conventional name for the intended referent.

The present study delves deeper into the conceptual nature of those semantic approximations produced by children under the general hypothesis that a vast number of them conform to the workings of a well-structured limited set of cognitive strategies, including conceptual metaphor, conceptual metonymy, and conceptual complexes. In this context, the present study represents a much needed corpus-based investigation of conceptual operations and complexes in child language. Current advances in cognitive modeling (Ruiz de Mendoza & Galera, 2014) are applied to the analysis of our corpus of semantic approximations in the language of a group of monolingual French-speaking children with a view on assessing the productivity of the aforementioned metaphorical and metonymic strategies, and conceptual complexes (i.e. metaphtonymies, double metonymies) in relation to this specific type of linguistic production. Additionally, the analysis of the data also investigates the level of conceptual complexity of the cognitive operations underlying the production of semantic approximations in child language. Finally, age effects are also considered.

The results of the investigation spell out the role played by these cognitive operations in the language of young children. Additionally, our findings also call for a revision and extension of the inventory of functions traditionally assigned to cognitive metaphor and metonymy.