Volume 20, Issue 3

Children generate novel utterances from the outset of multiword speech. In this study, we apply a usage-based method called ‘traceback’ to the multiword utterances of four two-year-olds to see how closely related these utterances are to their previous utterances. Data was collected from the age of 2;0 until 6 weeks later on a relatively dense sampling schedule. We attempted to match each novel multiword utterance in a two-hour corpus to lexical strings and schemas that the child had said before. Matches were found for between 78–92 percent of all multiword utterances. Between 62–91 percent of the slots in schemas created by these tracebacks were for referring expressions and were filled with nouns or noun phrases. For one child, recording continued throughout his third year and we compared his data at MLUs matched with the other three children to investigate developmental changes. We found that, with increasing MLU, and developmentally, children were less repetitive within sessions, the tracebacks required a wider range of semantic slots and the material placed in these slots increased in complexity.

This paper provides a corpus-linguistic, usage-based approach to the acquisition of be-going-to-V and be-gonna-V . Based on longitudinal data from two American children, it is argued that the constructions develop on the basis of several low-level chunks of varying degrees of morphosyntactic complexity. I propose an empirical way of grouping these chunks according to their structural and developmental properties, which allows us to trace how constructional networks emerge, expand and change in early childhood. In addition, this method reveals insights into the way the historically transmitted layering of the constructions is accessed in language acquisition. In particular, I uncover and account for apparent ‘grammaticalization effects’ in child speech, and discuss the relationship between acquisition and change in the cognitive-functional paradigm.

In numerous comprehension studies, across different languages, children have performed worse on object relatives (e.g., the dog that the cat chased ) than on subject relatives (e.g., the dog that chased the cat ). One possible reason for this is that the test sentences did not exactly match the kinds of object relatives that children typically experience. Adults and children usually hear and produce object relatives with inanimate heads and pronominal subjects (e.g., the car that we bought last year ) (cf. Kidd et al., Language and Cognitive Processes 22: 860–897, 2007). We tested young 3-year old German- and English-speaking children with a referential selection task. Children from both language groups performed best in the condition where the experimenter described inanimate referents with object relatives that contained pronominal subjects (e.g., Can you give me the sweater that he bought? ). Importantly, when the object relatives met the constraints identified in spoken discourse, children understood them as well as subject relatives, or even better. These results speak against a purely structural explanation for children's difficulty with object relatives as observed in previous studies, but rather support the usage-based account, according to which discourse function and experience with language shape the representation of linguistic structures.

A number of researchers have claimed that questions and other constructions with long distance dependencies (LDDs) are acquired relatively early, by age 4 or even earlier, in spite of their complexity. Analysis of LDD questions in the input available to children suggests that they are extremely stereotypical, raising the possibility that children learn lexically specific templates such as WH do you think S-GAP? rather than general rules of the kind postulated in traditional linguistic accounts of this construction. We describe three elicited imitation experiments with children aged from 4;6 to 6;9 and adult controls. Participants were asked to repeat prototypical questions (i.e., questions which match the hypothesised template), unprototypical questions (which depart from it in several respects) and declarative counterparts of both types of interrogative sentences. The children performed significantly better on the prototypical variants of both constructions, even when both variants contained exactly the same lexical material, while adults showed prototypicality e¤ects for LDD questions only. These results suggest that a general declarative complementation construction emerges quite late in development (after age 6), and that even adults rely on lexically specific templates for LDD questions.

This study tests accounts of co-reference errors whereby children allow “Mama Bear” and “her” to co-refer in sentences like “Mama Bear is washing her” (Chien and Wexler, Language Acquisition 1: 225–295, 1990). 63 children aged 4;6, 5;6 and 6;6 participated in a truth-value judgment task augmented with a sentence production component. There were three major finding: 1) contrary to predictions of most generativist accounts, children accepted co-reference even in cases of bound anaphora e.g., “Every girl is washing her” 2) contrary to Thornton and Wexler (Principle B, VP Ellipsis and Interpretation in Child Grammar, The MIT Press, 1999), errors did not appear to occur because children understood referring expressions to be denoting the same person in different guises 3) contrary to usage-based accounts, errors were less likely in sentences that contained lower as opposed to higher frequency verbs. Error rates also differed significantly according to pronoun type (“him”, “her”, “them”). These challenging results are discussed in terms of possible processing explanations.

A general framework is sketched for thinking about problems of usage and acquisition from a cognitive linguistic perspective. It is a dynamic, usage-based approach emphasizing the temporal dimension of language structure as an aspect of cognitive processing. A variety of topics are discussed involving the abstraction of linguistic units, their mental representation, and their activation in usage events.