This paper investigates developmental changes, as well as inter-linguistic and inter-individual variations, in the expansion and composition of French children's early lexicons. Two studies were conducted using children's naturalistic productions: a longitudinal study of one child between 1;2 and 2;6, and a cross-sectional study of two groups (12 children each) aged 1;8 and 2;6. Analyses indicate that lexical productivity (measured in types, tokens, and new words) strongly increased with age, whereas lexical diversity showed almost no developmental progression. Nouns and para-lexical elements (including interjections, fillers or formulas) were predominant until 1;8 and decreased over time, while predicates and grammatical words increased. As compared to English, French development was characterized by less frequent nouns, initially more frequent predicates, and a remarkable expansion of grammatical words. Inter-individual variability in lexical productivity, in lexical diversity, and in the proportions of different categories was more marked at 1;8 than at 2;6. Lexical profiles found at 1;8 suggest the existence of more diversified organizational patterns than those captured in the referential-expressive distinction.

This paper addresses the question of whether the rate and patterns of pronoun case error are influenced by the composition of an individual pronoun's paradigm. Twenty-nine children, aged 2;6 to 4;0 were audiotaped for two 1-hour sessions, interacting with their primary caregivers engaged in free play, book reading and family album viewing. It was found that her was overextended for she at a significantly higher rate than him for he and them for they, providing evidence of a ‘double-cell’ effect that increases the incidence of error in the feminine pronoun. It was also found that the overextension of nominative forms (e.g. he for him), were antagonistic to the more frequently occurring type of overextension, objective for nominative (e.g. him for he). These findings indicate that the composition of a pronoun's paradigm influenced the patterns of pronoun case error observed.

The simultaneous acquisition of two languages in early childhood presents an interesting test case for language acquisition theories. Children in bilingual environments receive input which could potentially lead to output systems different to those of monolingual children. The speech of three bilingual German-English children was recorded monthly between the ages of 2;0 and 5;0. The analysis of word order in the verb phrase shows that initial structural separation was followed by an extended period of non-target structures in German before the children eventually worked out which structures overlap and which structures differentiate the two languages. The bilingual data point towards language being acquired incrementally, on the basis of cue strength and cue cost. It is suggested that the partially overlapping structures in the input from German and English create structural saliencies for the child before they are functionally accessible. Functional identification eventually leads to structural separation.

Although children acquire Spanish subjunctive morphology early in the process of language acquisition, they only master mood selection in a staged process that lasts for several years. This paper examines the possibility that the acquisition of subjunctive mood selection in particular syntactic contexts is constrained by cognitive development in the area of representational theory of mind. Acquisition of the epistemic aspects of the semantics of subjunctive are shown to be associated with the understanding of false beliefs, a landmark development in children's cognition. Twenty-two Spanish speaking children between the ages of 3;5 and 6;11 participated in an elicited production study designed to test whether children's ability to produce subjunctive relative clauses was related to their ability to pass a false belief task. Results indicate a strong correlation between children's ability to use the subjunctive mood in relative clauses and their capacity for understanding false beliefs.

Twenty-four children between 2;5 and 3;1 were taught two nonce verbs. Each verb was used multiple times by an adult experimenter to refer to a highly transitive action involving a mostly animate agent (including the child herself) and a patient of varying animacy. One of the verbs was modelled in the Two-Participants condition in which the experimenter said: ‘Look. Big Bird is dopping the boat’. The other verb was modelled in the No-Participant condition in which the experimenter named the Two-Participants but did not use them as arguments of the novel verb: ‘Look what Big Bird is doing to the boat. It's called keefing’. It was found that whereas many children produced transitive sentences with the Two-Participants verb, only children close to 3;0 produced transitive sentences with the No-Participant verb. This age is somewhat younger than previous studies in which young children were asked to produce transitive sentences with two lexical nouns for the two animate participants. Also, re-analyses of previously published studies in which children learned novel verbs in sentence frames without arguments found that the few transitive sentences produced by children under 2;6 involved either I or me as subject. One hypothesis is thus that as young children in the third year of life begin to construct a more abstract and verb-general transitive construction, this construction initially contains only certain types of participants expressed in only certain kinds of linguistic forms.

This study investigates the use of relative clauses in French children's narrative monologues. Narrative texts were collected from French-speaking monolinguals in four age groups (five, seven, ten years and adults). Twenty subjects from each group were asked to tell a story based on a picture book consisting of twenty-four images without text (Frog, Where are you?). Relative constructions were coded following the categories defined by Dasinger & Toupin (1994) into two main functional classes: general discourse and narrative functions. The results show that the use of relative clauses in general discourse functions precedes their use in more specific narrative functions. An analysis of textual connectivity (Berman & Slobin, 1994) in one episode reveals that children and adults differ in their choice of preferred structures. The results also show that children use fewer transitive predicates in relative clauses than do adults. Transitive verbs are essential for advancing the narrative plot (Hopper & Thompson, 1980). While subject relative clauses are acquired early and used frequently, the development of their multifunctional use in diverse narrative functions extends well beyond childhood.

The two studies presented here explore the relationship between children's pragmatic skills and their growth in grammar. In study 1, thirty normally developing children were videotaped interacting with their parents at 1;2 and again at 2;7. Using correlational and regression techniques, we found that pragmatic accomplishments of MUTUAL ATTENTION, as well as mother's conversational style, explained 45% of the variance in grammar at 2;7. The second study investigated pragmatic–grammatical relationships with data from 6 high-functioning children with autism. To control for individual variation in skill level at the start of the study, within-individual growth rates for grammar were estimated as our outcome. The results substantiated those of study 1, in that pragmatic accomplishments within mutual attention predicted the per month growth rate in grammar. We interpret these findings as consistent with the position that the infant's social-pragmatic skills contribute to the acquisition of grammar.

In a longitudinal investigation of 40 child–mother dyads, we examined prediction from three indexes of children's own language: (1) vocal imitations, (2) first spontaneous words in production, and (3) receptive language starting at 0;9, and their mothers' verbal responsiveness at 0;9 and 1;1, to the developmental onset of three significant language milestones of the second year: (1) 50 words in productive language, (2) combinatorial speech, and (3) the use of language to express a memory. In these analyses, we utilized EVENTS HISTORY ANALYSIS, a statistical technique well suited to questions concerning when in development certain events begin and the extent to which predictors influence the timing of those events. The timing of children's first words in production, the timing of their achievement of 50 words in receptive language, and maternal responsiveness at 1;1 each contributed uniquely to variation in the timing of the three language milestones. When child and mother factors were considered together, the onset of the three language milestones differed by as much as 0;5 months for children in the lower and upper 10th percentiles of the predictor variables. The present findings contribute to generating and testing specific models about child and mother factors thought to explain variation in key aspects of children's second-year language development.

There is growing evidence that corrective input for grammatical errors is widely available to children (Farrar, 1992; Morgan, Bonamo & Travis, 1995). However, controversy still exists concerning the extent to which children can identify and exploit available negative input. In particular, very little is yet known about the longer-term effects of negative input. Performing a time series analysis on observational data, Morgan et al. (1995) conclude that corrective recasts are not related to future improvements in grammaticality. It is argued here, though, that the data sets analysed in this study are inherently ill-suited to the demands of time series analyses. The present study adopts an experimental approach in order to compare the effects of negative evidence versus positive input on the acquisition of irregular past tense verb forms. Twenty-six children (mean age 3;10) participated in a within-subjects design over a period of five weeks. It was found that improvements in the grammaticality of child speech were considerably greater in cases where negative evidence had been provided. Moreover, children's intuitions concerning the status of irregular and overregularized forms more closely approximated adult intuitions when corrective input was available.