Dyslexia and the Acquisition of Syntax: Passive and Contol.

This thesis discusses the acquisition of syntax in dyslexic children, focusing on the acquisition of passives and control structures. In chapter 1 we explain what a processing deficit hypothesis is. According to a processing deficit hypothesis, the children is competent and knows the grammar, but he/she is unable to process some constructions because of processing limitations. The processing burden, gathered through different cognitive tasks (processing phonological content, resolving syntactic dependencies, . . . ), is too heavy and causes the failure of the processing operations. In chapter 1 it is shown, in a survey of the literature, how dyslexic children suffer from processing limitations, especially in tasks involving the parsing of complex syntactic structures. Many studies have reported that the results obtained by dyslexic children often match the results obtained by younger children. In chapter 2 it is shown how 9-year-old dyslexics and 5-year-old children suffer from the so called Maratsos Effect: they both fail in tasks involving the comprehension of non-actional passives. We further argue that a processing deficit hypothesis, contrary to a structural lag hypothesis, is able to explain the presence of the Maratsos Effect both in normally developing children and in dyslexics. A structural lag hypothesis would have to explain why and how dyslexic children suffer from Maratsos Effect. The experimental results reported in chapter 2 also invite us to abandon the old theory of passive (Jaeggli 1986) in favour of new theories that can offer insights on the semantic aspects of passives and passivization. In chapter 4, we observe how dyslexic and their peers have both partially acquired control. The performances of the two groups are indistinguishable: they both make exactly the same mistakes. These results also show how dyslexic children generally perform as their peers in tasks involving syntactic parsing. Only few syntactic constructions seem to cause problems to dyslexic children and these constructions are characterized by being particularly cumbersome for processing resources. Chapter 4 argues, following Cohen Sherman and Lust (1993), that the acquisition of complement control is carried on through the integration of syntactic cues and lexical information (provided by positive evidence) rather than through the progressive abandoning of default cognitive strategies. In chapter 3, we provide a survey of the main theories of control. It is not a novelty that there is no comprehensive theory of control phenomena: theories focused on the licensing and interpretation of the simplest complement control phenomena, leaving aside adjunct control, variable control and split control. The chapter shows how every theory is forced to make (sometimes strong) stipulations in order to account for control phenomena. The fact that it seems impossible to explain control phenomena, appar- ently, without resorting to strong stipulations suggests that control is really a “module”, an ensemble of syntactic–semantic phenomena that cannot be fully reduced to independently existing syntactic operations.