Developing knowledge of the vehicle, animal, and fruit categories was traced in six children from 1; 0 to 1; 8. Data from mothers' language diaries and from bi-monthly sessions with the children were pooled to analyse the growth, content, and internal structure of the three categories over time. The children developed some grasp of most of the focal concepts in each category, but they made fewer differentiations than adults do. Overextension of a single concept term to encompass a cluster of related referents was common. The frequent discrepancies between comprehension and production of concept terms highlighted the importance of examining both modes. The data showed marked individual differences in style of category acquisition.

Various kinds of children's lexical errors, mostly based on similarity in sound, are presented and classified. At the earliest stage some children are found to pursue a HOMONYM STRATEGY, actively seeking to combine adult word-patterns to limit their output repertoire. The associations between words underlying these productive homonyms, together with perception-based errors, blends, and other word-substitutions, are compared with malapropisms from a slightly older group of children as well as with data from adults. Analysis of these data makes it possible to chart developmental changes in the phonological links which form one substructure of the lexicon.

The paper examines the acquisition of selected aspects of the inflectional system of Modern Hebrew, a language rich in bound morphology. By age three, children acquire the major inflectionally marked categories of the system, in the sense that they make semantically relevant distinctions of tense, person, number, and gender. Certain morphologically complex forms are simplified by neutralization or reformulation or by analytic paraphrases of bound constructions. Various anomalous forms are handled by regularization of lexical exceptions or by conflating forms belonging to different lexical patterns, while forms which are opaque due to neutralization of historically distinct root consonants or to inaccessibility of rules governing their alternations are processed by reference to certain ‘paradigms’ taken as basic.

An allative bias has been noted in children's comprehension of deictic verbs and directional prepositions. This study investigates the allative bias and its susceptibility to contextual modification. Three-year-olds were asked to act according to requests to move one object to or from another, and to identify where an object had come from or gone to. A reliable bias was found whereby ‘to’ was easier than ‘from’. Attempts to make the utterance more naturally related to the task, by using movement and fronted objects, modulated the bias to a small extent, but did not abolish it. An analysis of directional expressions in terms of the components SOURCE, PATH and GOAL is presented in relation to the task-demands.

Child partners were compared on their relative tendency to use the interrogative form in spontaneous dyadic verbal interactions. A given child's tendency to use the interrogative form more frequently than the other member of the dyad remained constant across partners. This relative stability in formulating utterances in interrogative form cannot be attributed to the child's producing more talk in general than the partner. The observed effect is explained as a type of conversational synchrony in which participants achieve a complementary balance rather than congruence. It was concluded that the interrogative form is a linguistic means for child peers to establish cooperation as well as controlling kinds of relationships during an ongoing interaction.

Three experiments investigated the ability of children aged 2; 2–4; 10 to use identifying expressions to introduce new referents and the definite article to refer to already introduced referents. The results suggest that young children initially master the nominative use of a, but that from the age of 3 they have mastered the non-egocentric use of the definite article. On first mention their referring expressions are predominantly indefinite, and on second mention definite. The results are discussed in relation to previous studies, and the questions of suitable adult controls and ‘appropriateness’ of article usage are also considered.

Based on Clark & Lucy's (1975) study of adult comprehension of different types of indirect requests, this study addresses the questions: ‘How well are children aged 4 to 7 able to understand these same indirect requests?’ and ‘How does the relation between request type and ease of comprehension compare from children to adults?’ Results showed: (1) these children are able to comprehend a wide variety of indirect requests; (2) there is a general developmental pattern of acquisition; (3) interrogative forms are more difficult than declarative forms; (4) conveyed negative requests are more difficult than corresponding conveyed positive requests; (5) children are heavily influenced by the surface polarity of the conveyed request; and (6) the relationship between request type and ease of comprehension is strikingly similar for both children and adults.

First, third, and fifth graders judged a speaker's belief about entailment and propositional truth in sentences containing five subordinating conjunction-verb combinations: if, because, although, unless combined with indicative verbs, and if with a subjunctive verb. Children were more accurate when a speaker believed the truth of his utterance (because and although) than when he voiced disbelief (if + subjunctive) or doubt (unless and if + indicative). They were also more accurate in judging entailment when the speaker thought the events in the two clauses were positively related (if and because) than when the speaker expressed a negative relation (unless and although). The pattern of errors in judgement conformed to a pragmatic analysis of the conjunctions. First graders based their judgements on semantic cues and did not differentiate propositional truth and entailment concepts. Third and fifth graders understood the difference between entailment and propositional truth and used syntactic cues to differentiate positive from non-positive statements. However, they thought that a subordinating conjunction only determines belief about its own clause, and they did not understand the appropriate cues expressing uncertainty vs. disbelief.

The present study concerned the acquisition of double-object, bi-transitive sentences which can be expressed in either BASIC (S-V-DO-to-IO) or TRANSFORM (S-V-IO-DO) constructions. Three-, four-, and five-year-old subjects were tested on comprehension and production of these two alternative forms, investigating the distribution of Basic and Transform sentences in acquisition, and also the effect of prototypicality of the described event on performance. The results showed that the Basic structure was earlier in acquisition, with the Transform appearing later and, at first, only in descriptions of the more prototypical events. Overall, the more prototypical an event, the more competent the children were in comprehension and production and the better able they were to handle the more difficult Transform sentences. Decreased prototypicality resulted in lower performance and greater dependence on the Basic as opposed to the Transform bi-transitives. The findings are discussed within the framework of a general theory of language performance being developed by the first author.

Forty-eight normally developing children, twelve each at ages three, four five, and six years, were asked to define and describe ten concrete nouns expressively with no attendant visual stimuli. Responses were coded according to the semantic relations expressed. Nineteen different semantic relationships were isolated in these definitions, and significant differences were found between the age levels in the various configurations of semantic relational categories represented in the first five relations elicited from each child by each noun. Compatible with recent hypotheses concerning the development of internal semantic memory structures, these differences also proved to be highly predictive of a subject's chronological age. The results suggest that semantic relational analysis may be a productive method of documenting the characteristics of lexical semantic systems of very young children.