Previous scholars have claimed that the child's babbling (meaningless speech-like vocalizations) includes a random assortment of the speech sounds found in the languages of the world. Babbled sounds have been claimed to bear no relationship to the sounds of the child's later meaningful speech. The present research disputes the traditional position on babbling by showing that the phonetic content of babbled utterances exhibits many of the same preferences for certain kinds of phonetic elements and sequences that have been found in the production of meaningful speech by children in later stages of language development.

Two transitional phases in the child's early language development are described; the first occurs between prelinguistic vocalization and one-word speech and the second between one-word and patterned speech. Cognitive, linguistic and affective inputs to the acquisition of reference and syntax are discussed in the light of the transitional phenomena that were found. We claim that each major linguistic stage is preceded by a transitional phase which serves as a bridging device for the next major acquisition; that sound and meaning develop partly independently in language development; and that the child's earliest patterned speech is not organized in terms of knowledge of grammatical categories, but in terms of more fundamental coordinations of conceptual meanings with phonetic outputs. A theoretical framework is proposed which provides a more systematic treatment of transitional phenomena than has previously been provided. The framework allows for interpretations of transitional phenomena and of their relations to the milestone periods of early language development.

Recent attempts to classify the meanings of two-word utterances expressed by young children have assumed that children always intend ONE of the meanings adults might express. This paper challenges that assumption for two reasons. Firstly, it is by no means self-evidently true. Indeed, since children and adults conceptualize the world differently, they may also differ in what they choose to say about the world. Secondly, it has supported research strategies which are at best unjustified and at worst ambiguous. In fact, research based on the assumption that children always intend a meaning adults might express has provided interesting insights into the interpretations adults place upon children's utterances but said next to nothing about the meaning of these utterances. This paper concludes by suggesting an alternative approach to the latter problem.

In this longitudinal investigation of the emerging grammar of seven children, differences in linguistic acquisition were observed. The syntactic analyses applied to the corpora included examination of emerging complexity, observance of word order constraints and subject-predicate specification among others. These analyses revealed two distinct styles of syntactic acquisition. These linguistic styles appeared to be sex- and speed-related with specific ties to particular utterance types and grammatical-relational specification. The observed styles of syntactic acquisition were differentiated by these differences so that the differences themselves constituted the style characteristics.

This study seeks to explore the defining characteristics of first and third grade children (mean ages 7;0 and 9;1 respectively) in conceptualizing seventeen American kin terms. The data indicate that even when children were able to identify a relationship, they did not all base their identification on the same attributes. Familiarity and experience affected the identification of some terms but did not appear to influence the ability to handle relational aspects. Among Grade 1 children, girls were more proficient than boys. This difference disappeared by Grade 3 where both boys and girls were equally competent and significantly more proficient than the younger children. Finally, in both grades, nuclear family terms were more familiar than extended family terms.

Four children (aged 3;9 to 5;5) were recorded talking to different listeners (adult, peer, baby, and baby doll) and role-playing a ‘baby just learning how to talk’. As a measure of the children's responsiveness to situational cues, each sample was analysed for formal and functional characteristics. The children's speech was different on many of the measures when talking to a baby or doll as compared with the speech to a peer or adult. Speech to a baby doll was similar to that addressed to a real baby, indicating that feedback is not necessary for speech modifications to occur. In the role-playing situation, children changed and simplified their speech when talking as a baby.

The written compositions of 11-year-olds in the National Child Development Study were analysed using the T-unit length (Hunt 1965) as a measure of syntactic maturity, and composition length as a measure of productivity. Only a weak association was found between mean T-unit length (MTUL) and composition length. There were no statistically significant regional differences in either composition length or MTUL. There were statistically significant sex and social class differences in composition length but not in MTUL. Composition length was associated with Reading Comprehension, General Ability test scores, and teachers' ratings of certain scholastic aptitudes. MTUL was associated only slightly or not at all with these measures. The results are discussed in relation to cognitive and linguistic development.

The development of the meaning of shoe in one pre-lingual child, plus additional examples drawn from the literature, support a notion that word meanings start out very narrow and only become overextended later, though sometimes before the word is spoken. This appears to contradict the course of development of meaning hypothesized by Clark (1973). It is argued that the early development of word meaning is simply a special case of a much more general learning process.