Early discourse competence in preschool children: An elicitation study

点击翻译按钮获取中文摘要

Discourse competence is the linguistic knowledge that enables the adequate encoding of information in a given context, reflecting what the speaker assumes the hearer to know/believe. It is commonly assumed that preschool children lack the discourse competence necessary to encode information in an adult-like fashion. Experimental studies have argued for instance that preschool children use definites (such as 'the duck') rather than indefinites ('a duck') in contexts where only indefinites are possible in the adult language (see e.g. Maratsos 1974; Emslie & Stevenson 1981; Hickmann et al1996).This research proposes new, theoretically motivated diagnostics to identify signs of discourse competence. In particular, it aims to ascertain whether preschool children master the discourse notion of topic - a claim put forth in De Cat (2002) which has never been confirmed experimentally. Topics express what the sentence is about. In Spoken French, topics are obligatorily dislocated: they appear at the periphery of the sentence - indicated by the comma in (1) - and are resumed by an element inside it (here: 'il'). Topics have to correspond to information that is salient and recoverable in the context to enable the hearer to identify what is being talked about. In most cases, they must be definite.(1) Le monsieur, il lui donne sa balle. the man he to-her gives her ballTwo elicitation tasks will be carried out to ascertain (i) whether preschool children master the discourse notion of topic and from what age and (ii) whether they encode new vs old information like adults and from what age. A third, non-linguistic task will assess children's Theory of Mind in an attempt to disentangle linguistic from cognitivesources of target-deviance in the encoding of information. Each task will target the same 45 monolingual preschool children acquiring French (3 groups, aged 2.5, 3.5 and 4.5 respectively), plus 15 adult controls.Task 1 will reveal the extent of their knowledge of the notion of topic. The test material will consist of 8 sets of 3 pictures setting up a scenario which will prompt the child to produce two types of sentences, depending on whether the subject is new or old information: sentences requiring a dislocated subject because a topic interpretation is obligatory (as in (1)); and sentences with a subject in its non-dislocated position because a topic interpretation is impossible (as in (2)).(2) Un monsieur lui donne sa balle. a man to-her gives her ballTask 2 will test children's ability to encode new discourse items (or referents) like adults. The general outline of this task is as in Hickmann et al (1996): the child is made to blindfold the interviewer and to explain to her what happens in two sequences of 5 pictures representing an event. A new referent is introduced in each picture (as an additional character). Children's productions will be analysed in two respects: (i) to evaluate whether they know the rules of new referent encoding and (ii) to evaluate the impact of non-linguistic factors on their encoding of new referents.The Theory of Mind task will assess children's ability to distinguish their listener's perspective from their own. The proposed study seeks to demonstrate the need to distinguish linguistic from cognitive underpinnings of discourse competence. Based on the applicant's previous research (including a longitudinal study of spontaneous production and a pilot experiment for Task 1), it is expected that children will be able to encode obligatory topicslike adults and that the contexts yielding errors in the encoding of information contrasts will be those requiring certain milestones to have been reached in Theory of Mind development. This research will contribute to narrowing down significantly the search space for causes of children's defineness errors, which have been widely reported but remain little understood.