Language Development and Disorders (LDD): research projects
Ordering effects in the acquisition of complex sentences: A multivariate approach
In many languages, complex sentences are flexible in their ordering of main and subordinate clause. However, studies have shown that this flexibility is determined by factors from different domains such as discourse-pragmatics, clause processing and semantics (cf. Diessel 2001;2005;2008). In so doing, researchers have argued that there is a processing advantage of putting given before new information and main clauses before subordinate clauses. The majority of studies in language acquisition have focused on ordering effects on the phrase level (e.g. Baker & Greenfield, 1988; Narasimhan & Dimroth, 2008) whereas there is little work on the complex sentence level. The aim of the thesis was to assess whether children’s ordering of complex sentences is affected by factors from discourse-pragmatics, processing and semantics. Taking a developmental perspective, we investigated ordering effects in complex sentences from the perspective of spontaneous speech followed by an experimental investigation from the perspective of comprehension and production.
In Study 1, we investigated a small sample of corpus data from an English-and a German speaking child. The aim of the study was to investigate whether the children’s ordering of English when- and German wenn-clauses is affected by their information status (i.e. given or new information), their position in the complex sentence (i.e. initial or final), their length (i.e. short or long) and their semantics (i.e. successive or simultaneous). The results showed that neither the children nor their caretakers showed any preference of clause order as a result of the clauses’ information status, length and semantics.
As corpus studies do not allow for any control over variables and contexts, Studies 2 and 3 aimed to investigate ordering effects in complex sentences experimentally. In Study 2, we examined whether clause order and order of information have an impact on children’s and adults act out of English when-clauses. The results showed that children and adults were more likely to change the order of information from new-given to given-new in their act-outs when they were exposed to a new-given structure in the complex sentences. In addition, the adults tended to change the order of act-out to subordinate-main more often when exposed to a main-subordinate structure, suggesting that they are also sensitive to clause-processing factors in their comprehension of complex sentences.
Study 2 showed that children and adults are sensitive to information-structural properties in their comprehension of complex sentences. In Study 3, we tested whether these factors have a similar effect on children’s and adults production of complex sentences. In this study, children were asked to report complex sentences to their caretakers. The results showed that the children did not show any preference for given-before-new information in their production of English when-clauses. However, the adults preferred to report the complex sentences in a given-new structure.
Taken together, the studies in this thesis suggest that factors from the domain of discourse-pragmatics have an impact on children’s ordering of complex sentences in comprehension whereas these factors have less impact in production. In addition, the results show that clause-processing factors are stronger in adults than in children suggesting that these factors might only become important at a later stage of syntactic development (i.e. after the age of 5).The studies in this thesis represent a first attempt to an experimental approach to the ordering of complex sentences in children.
Duration of the project
PhD studentship 2009-2012
Max Planck Institute for Evolutionary Anthropology