Measures of pragmatic development for children in the second year of life were developed, based on a brief (30 minute) language sample and on structured elicitation conditions. When age is partialled out, number of pragmatic functions expressed is not statistically related to MLU, due to the fact that the number of functions grows steadily during the one-word and very early two-word phases. Performance on the structured imperative tasks also reflects development during this period, though performance on the declarative tasks does not. We conclude that the range of pragmatic functions in the second year is measurable and contributes information not provided by a measure of syntactic development such as MLU. Such information should prove useful for research on language and cognition in normal and language-impaired children.

A psycholinguistic experiment elicits highly reliable judgements from young English-speaking children aged 2; 6–3; 0 about illocutionary force of utterances presented in controlled contexts. Puppet play simulated extralinguistic features judged capable of constituting felicity conditions upon the illocutionary acts Request and Offer. The experimental data bear upon two questions: (1) What set of features, linguistic and pragmatic, constitute cues for the discrimination of illocutionary force by young children? (2) What is the lower developmental bound of the emergence of discrimination of illocutionary force for Requests and Offers? While grasp of illocutionary force for Offers was well-established by 2; 6, discrimination skills for Requests probably continues to develop between 2; 6 and 3; 0. Relevant pragmatic features are explicated, and a model of the comprehension of illocutionary force proposed in the light of the results.

An investigation was made of the episodic structure of maternal speech to young children during a free-play session. The findings indicate that maternal speech was organized so as to provide a high degree of redundancy with regard to the referential component of speech. The beginning of episodes, in which the same object was repeatedly referred to, was clearly marked; furthermore speech inside episodes with a common referent was highly repetitive. All this suggests that maternal speech is organized in a way that increases the predictability in the child's language environment.

This paper reports on a longitudinal study of the acquisition of the voicing contrast in American English word-initial stop consonants, as measured by voice onset time. Four monolingual children were recorded at two-week intervals, beginning when the children were about 1; 6. Data provide evidence for three general stages: (1) the child has no contrast; (2) the child has a contrast but one that falls within the adult perceptual boundaries of one (usually voiced) phoneme and thus is presumably not perceptible to adults; and (3) the child has a contrast that resembles the adult contrast. The rate and nature of the developmental process are discussed in relation to two competing models for phonological acquisition and two hypotheses regarding the skills being learned.

The role of a strategy of reduplication in phonological acquisition and behaviour was examined in terms of: (1) the relationship between adoption of this strategy and children's failure to produce nonreduplicated multisyllabic forms and final consonants when attempting corresponding adult models; and (2) the role of the process of reduplication in production constraints. Comparisons of the phonologies of six children (1; 3–1; 9) classified as reduplicators and six children (1; 5–2; 0) classified as non-reduplicators revealed that the adoption/nonadoption of this strategy was closely related to the extent of production of nonreduplicated multisyllabic forms. Analyses of attempted words produced as reduplications revealed that this process served primarily to constrain multisyllabic productions. To a lesser degree, reduplication appeared to serve in conspiracies to constrain final consonant production.

Several researchers (e.g. Bever 1970, Strohner & Nelson 1974) have suggested that young children commonly employ a word-order strategy of the form: ‘the first mentioned participant is agent’ when interpreting active and passive sentences. Pre-school children (aged 2; 6–5; 0) were presented with reversible active and passive sentences in four comprehension test settings (two acting-out tasks, a selection task and a verification task). Their responses were analysed not in terms of the percentage correct responses for each group of subjects, as has been the usual practice, but in terms of each individual child's pattern of responses. The results revealed age-related changes in the type of response patterns found. The response pattern associated with the word-order strategy was not a frequent occurrence. The most common patterns were those accounted for by extralinguistic cues such as the relative proximity, mobility or position of the referents. The significance of early interpretation rules of this sort for any characterization of young children's language comprehension processes is discussed.

To evaluate Brown's assertion that frequency of input is not a significant variable in language acquisition, some of Brown's data were reanalysed. The data pertaining to Adam, Eve, Sarah and their parents were predominantly obtained from Brown's (1973) book and were supplemented from the transcripts. Parental input frequency and the children's age at mastery were highly related for each of the three triads. Then Sarah's input and rate of language acquisition were compared with that of the other two children. The correlation between Sarah's relative input deprivation and her relative linguistic delay was 0·66. Finally, an analysis of Eve's acquisition of specific prepositional phrases involving the preposition in were made. Again it was found that frequency of input was highly related to frequency of production. Detailed analyses of parent-child interactions provided evidence also for short-term effects of input frequency.

Thirty children with a mean length of utterance ranging from 1·00 to 4 and an age range of 1; 7 to 5; 5 were tested by their own mothers for comprehension of two-noun possessive phrases such as mommy's shoe. Three types of possessive relationships (alienable, intrinsic and reciprocal) in addition to anomalous possessive phrases were used to uncover children's knowledge of the semantics and syntax of English possession. Results indicated, first, that even young children may have detailed notions of which objects are likely to serve as possessors and which as possessions, and second, that word order may not be used to comprehepossessive phrases until considerable linguistic development has occurred.

Judgements of grammatical acceptability were made for correct and reversed ‘Y if X’ and ‘If X, Y’ sentences. Children aged 4; 10–8; 7 judged sentences ‘sensible’ or ‘silly’ and modified the sentences to make them ‘silly’ or ‘sensible’. Children failed to differentiate correct and reversed-order sentences prior to age 7 or 8. There was no evidence of an order-of-mention response strategy, but ‘If X, Y’ and ‘If Y, X’ sentences were easier to judge than ‘Y if X’ and ‘X if Y’, suggesting that children prefer sentences in which stated order parallels temporal order. The strategies used to modify the sentences changed with age from content changes to order reversals. The modifications suggested that contingency becomes a component of the meaning of if before the child assigns unidirectional event order as a component of if.

This paper presents a hypothesis-testing theory of syntax acquisition. The first section presents our model. We claim that: (1) children learn a transformational grammar, including a set of phrase structure and transformational rules; (2) linguistic universals and Occam's razor constrain the initial hypothesis space available to the device; (3) hypotheses tested by the device consist of candidate phrase structure and transformational rules; (4) linguistic evidence confirms or disconfirms hypotheses. Specific examples of incorrect phrase structure and transformational hypotheses are presented.

The second section briefly surveys other approaches to language acquisition – both syntactic and non-syntactic – and compares them to our model. In the third section, we address several methodological issues: (1) the relevance of linguistic theory to the model; (2) how the model is tested; (3) the domain of the theory.