This paper concerns the development of semi-modal want/want to, and modals may, can and could in two Trinidadian children, whose Verb Phrase development was studied in naturalistic settings, between the ages of 2;3 and 4;1, and 2;4 and 4;9. The similarities and differences in development between the two are important for underlining a number of key factors in the acquisition process. Accepting the salience of requests/demands in early child language, the study draws attention to the child's propensity for using the means most readily available to him/her in the input, for expression of this function. Additionally, there is evidence of the child's ability to make stylistic discriminations at a very early age and of the propensity for making semantic distinctions among forms. The study indicates that work on the development of grammatical features is incomplete without consideration of sociolinguistic aspects of usage as integral to grammatical acquisition from the outset.

Attempts were made to elicit poems from 133 children between the ages of 2;3 and 6;11. Seventy-eight of the children produced 606 poems between them. Forty-five per cent of the poems contained the syntactic device of modified repetition: a substitution exercise where a grammatical frame is repeated and the substitution occurs in part of the grammatical frame. This was so despite the fact that modified repetition was not present in the examples that were used to elicit poems from the children. The frequency and types of modified repetition used by the children did not vary much with age. The question of whether children's use of modified repetition in their rhythmical poems has the function of helping them to practise grammatical forms, or whether it is simply one reflection of a general human tendency towards the use of pattern in language, is discussed.

Shared book reading is considered to be an influential factor in language and literacy development and has been associated with more complex and demanding forms of tutorial language use. The present study seeks to examine mothers' discourse in joint picture book reading in terms of the level of representational demand which it places on the child, that is the extent to which it requires the child to engage in abstract thought. This was compared with the language used in another interactive, but more product oriented task, modelling a clown using play-doh. Twenty-four mother child dyads (in which the mean age of the children was 4;2) were videotaped while engaged on these tasks and the conversation was analysed for representational demand (levels of abstraction), interactive features and communicative function. The mother's conversation was found to be at a higher level of abstraction in reading than in play-doh modelling. The accuracy of child responses to high level requests for information was superior for reading, while responses to low level requests were similar in both situations. The interactive categories were more utterance-based and less attention-based than in the modelling task.

Children between the ages of three and seven occasionally make errors with locative verbs like pour and fill, such as * I filled water into the glass and * I poured the glass with water (Bowerman, 1982). To account for this pattern of errors, and for how they are eventually unlearned, we propose that children use a universal linking rule called OBJECT AFFECTEDNESS: the direct object corresponds to the argument that is specified as ‘affected’ in some particular way in the semantic representation of a verb. However, children must learn which verbs specify which of their arguments as being affected; specifically, whether it is the argument whose referent is undergoing a change of location, such as the content argument of pour, or the argument whose referent is undergoing a change of state, such as the container argument of fill. This predicts that syntactic errors should be associated with specific kinds of misinterpretations of verb meaning. Two experiments were performed on the ability of children and adults to understand and produce locative verbs. The results confirm that children tend to make syntactic errors with sentences containing fill and empty, encoding the content argument as direct object (e.g. fill the water). As predicted, children also misinterpreted the meanings of fill and empty as requiring not only that the container be brought into a full or empty state, but also that the content move in some specific manner (by pouring, or by dumping). Furthermore, children who misinterpreted the verbs' meanings were more likely to make syntactic errors with them. These findings support the hypothesis that verb meaning and syntax are linked in precise ways in the lexicons of language learners.

This study examines the use of the co-temporal connectives when, while and as in the elicited narratives of 71 children between 4;10 and 11;11. A group of 26 adults provided comparison data. Subjects were asked to ‘tell a story’ about each of three sets of story pictures. Analysis of the adult data confirmed a preference for preposed over postposed when clauses and supported the notion of a continuum of simultaneity for when, while and as. The oldest children (like the adults) were more likely to prepose when clauses than were younger children, a finding which suggests that with increasing awareness of the information needs of the listener, children begin to use preposed adverbial clauses as information ‘guideposts’. Surprisingly, however, children in general were far more likely than adults to use durative predicates with when clauses and did not use co-temporal as – substituting when or while in those contexts in which as appeared in the adult protocols. These two findings are shown to be connected, since aspect in the adult data is differentially distributed with respect to the three connectives. It is hypothesized that for narrative purposes children acquire when before while and while before as, in the order of increasing specificity and constraint, with the determining factor being a convergence of syntax, semantics and information flow.

Young children's skilfulness in conversational interaction is of great importance in understanding pragmatic aspects of language development. The purpose of this investigation was to describe the portion of preschool verbal interaction which could not be considered successful dialogue. Ten dyads of four-year-olds were videotaped during free play. Segments of talk were identified as dialogue or non-dialogue. Non-dialogue was coded when one child did not appear to direct his/her utterances to the other or when an attempted topic failed to capture the interest of the other child. Thirteen types of non-dialogue were identified. Results revealed that all dyads exhibited non-dialogue, but differences were noted in regard to amount and type. Dyads labelled as less co-operative displayed greater amounts of self-directed talk (monologue).

The early production of nine cohesive devices during narration about personal experience was examined in an 18-month longitudinal study of 10 children between approximately age 2;0 and 3;6. The specification of noun phrases and types of noun errors were also explored. The number of cohesive ties increased with both age and MLU, due to increased pronominal reference and conjunctions (while clausal and verbal ellipsis decreased). Specific cohesive devices also were acquired at different MLU levels. Most noun phrases were non-problematic and errors declined with increasing age and MLU; specifically, noun omissions declined. However, when new nouns were introduced, approximately one out of five were ambiguous throughout the study. Thus, children's narratives become more comprehensible with age as cohesive links increase and noun errors decrease, but at 3;6 children are still having difficulty properly managing the introduction of novel nouns.

Young speakers in dialogue must establish mutual knowledge. Traditionally, researchers have focused on how children used indefinite and definite articles to signal novel and shared information. In this study of 170 children aged seven to thirteen, the form of introduction chosen, whether question or statement, is more significant than the type of article used. No developmental effects on article use emerge, with statement + indefinite always the least common choice. However, young speakers use question introductions significantly less than older children. There is also a developmental interaction between speakers' and listeners' behaviour. Question introduction elicit effective listener responses at all ages, but only twelve-year-olds respond reliably to statements. Unfortunately, young speakers prefer just those forms of introduction to which young listeners are least likely to provide informative feedback. Thus two separate developmental effects appear to combine to hamper young subjects' chances of achieving successful communication in dialogues.

Defining is a metalinguistic skill. This study charts the development of the ability to define. A total of 120 participants (40 at each of the ages of 10, 14 and 18 years) defined eight concrete and eight abstract nouns. Each definition was coded to identify the types of responses present in it. Results indicate that there are significant differences in the way concrete and abstract nouns are defined. All three age-groups defined concrete nouns mainly by class and characteristic responses. But it was not until age 18 that abstract nouns were defined by class and characteristic responses. The development of abstract noun definitions follows the development of concrete noun definitions. The ability to define is dependent upon knowledge of the meaning of the word but it is also dependent upon implicit or explicit knowledge of the definitional form. The ability to define abstract nouns is a late developing metalinguistic skill.

Standard outcome measures used in a referential communication (route finding) task showed that 7-year-olds were inferior to 10-year-olds and adults in terms of adequacy of messages provided (as speakers) and selection of referent (as listeners). Because the task involved negotiation of route directions by telephone, a richer dialogue resulted than is usual in referential communication studies. This made possible an analysis of the strategies used and revealed differences untapped by the standard measures. The types of strategies used (COMPONENTS, NUMBERING, DIRECTIONAL, MINIMAL) varied as a function of age, with the adults using much directional information and the children very little. Much of the younger children's description was either minimal or used inadequately specified mature strategies. Adults frequently used dual strategies (e.g. directional + numbering). Older children showed little evidence of these individually but achieved them, in practice, through dialogue with their partners. Age differences in quantity of communication (number of turns) were found to be linked to strategy use. The advantages of the telephone as an experimental medium of communication are discussed.

This study reports on four school-age boys' use of the specialized register of sportscasting as performed both spontaneously and when elicited by the researcher, and compares the boys' production of formal register-marking features to adult usage. Like the commentary of professional sportscasters, the boys' commentating is characterized by heavy use of simple-present action verbs, use of constructions lacking elements that are ordinarily grammatically required (the subject pronoun he, an auxiliary verb, or a main verb), and (in the elicited version) frequent occurrence of passives. The boys invest the most ‘ungram-matical’ utterance types (those lacking finite verbs) with a function that is not clearly present in adult sportscasting discourse, that of pointing forward in the discourse.

The way seven- and nine-year-old French children and adults interpret ambiguous object descriptions was studied in four experiments. All four experiments followed the same basic procedure and consisted of two phases. During the training phase, plastic blocks varying in shape, colour and size were associated with fruit names. Some objects were always given just one label, the generic term, whereas others could be given either the generic or the specific label. For example, large circles were referred to as ‘pommes’ (‘apples’), and large BLUE circles as ‘reinettes’ (a kind of apple). Still others were given no name at all. During the test phase of the experiment, subjects were presented with referentially ambiguous messages and asked to point to a given object (Experiments 1, 2 and 3) or to draw an object (Experiment 4). The messages consisted of either a generic term preceded by an article (e.g. ‘(la pomme’ – ‘the apple’) or a verbal description based on a colour adjective (e.g. ‘le bleu’ – ‘the blue one’). When presented with the generic term, older children and adults persistently chose the object for which no specific term had been given during the training phase. When presented with an adjective description, they chose the referent for which no fruit name had been given during the training phase. Younger children's choices depended on whether the article preceding the description was definite or indefinite. These results are consistent with the Gricean approach to referential communication proposed by Jackson & Jacobs (1982) and with an extended version of the principle of contrast proposed by Clark (1987, 1988).

This study investigated the changing functions of evaluative devices in children's narratives. The evaluative devices included (a) references to ‘frames of mind’, particularly to emotions, (b) character speech, (c) ‘hedges’, (d) negative qualifiers, and (e) causal connectors. Narratives were elicited from a 24-picture story book. The subjects were three groups of native English-speaking Americans (12 per group): five- and nine-year-old children and college undergraduate students. A quantitative comparison revealed that (i) adults used evaluative devices three times as often as five-year-olds, and two-and-a-half times as often as the nine-year-old children; (ii) adults used significantly more references to ‘frames of mind’ and ‘hedges’ than the children; and (iii) whereas five-year-olds used each evaluative type equally often, nine-year-olds and adults used references to ‘frames of mind’ significantly more than the other four evaluation types. A second analysis, focusing specifically on the discourse functions of references to ‘frames of mind’ revealed that, early on, this particular device is used to express a local evaluative perspective on particular events, while with increasing age it is used to signal the hierarchial organization of the story events. These findings are discussed with regard to two non-linguistic developmental achievements, the formation of event schemas and the formation of a theory of mind.

Twenty Italian six-year-olds and 20 eight-year-olds were asked to interpret eight ambiguous and eight clear definite descriptions. All ambiguous descriptions could refer to three drawings, one of which had been described by the subjects immediately before the comprehension task. In half of the trials with ambiguous messages the children's interlocutor was present while the children were describing the drawings; in the other half he was absent. In both conditions subjects showed a preference for the referents they had already described, indicating that they applied egocentrically a comprehension strategy based on the Maxim of Antecedent (Jackson & Jacobs, 1982). Children's failures to differentiate their responses in the two conditions are considered to be due to difficulties in taking account of the given-new distinction for relevant information.

Garman (1974), reporting on twenty Tamil children aged three to five, postulated a linguistic strategy and two prelinguistic strategies to explain the results of a question-picture choice task involving sentences with embedded and subordinate clauses. In a reanalysis of his data, we identify four processing strategies and show that some of Garman's findings are better explained not as the outcome of prelinguistic strategies but as an artefact of the experimental design. In fact the data provide evidence of a grammatical sensitivity which is consonant with a sensitivity – demonstrated in recent language-acquisition studies – to the branching direction of the language being acquired.

This study investigates young children's understanding of homophones in two different linguistic contexts. Research has shown that children interpret homophones according to their ‘primary’ sense even though this is inappropriate to the linguistic context. The present study investigates the influence of different contexts on their interpretations of homophones. A picture test, similar to the British Picture Vocabulary Scale (BPVS), was devised, depicting ten homophones in both their primary and secondary senses. Forty-eight children who were in matched pairs following the BPVS pre-test and whose ages ranged from three to six were asked to select the picture which showed the homophone, first when the homophone was in no linguistic context, and secondly following one of two conditions of the same story which differed according to the richness of context provided. Results showed that those children assigned to story B, the expanded version with full linguistic details and explanations of the words it contained, were more able to identify correctly the secondary meanings of the homophones intended than those children given story A, the unexpanded version. Analyses of variance showed both a significant age trend and a story-type effect. These results are interpreted as showing that the linguistic context is an important factor in young children's understanding of word meaning.