Vocalizations of two female infants, recorded over a 5-week period after the first emergence of cooing, were studied. The vocalizations were classified as cry, discomfort, vegetative and cooing (comfort) sounds. Fifty vocalizations of each sound type, distributed across the 5-week period of study, were selected from the output of each infant (400 vocalizations in all). The features of voicing, breath direction and consonant-like versus vowel-like were documented for each sound type. It was found that the features of the more primitive sound types regrouped themselves in comfort sounds. For the first time the feature of consonant-like was found to co-occur with voicing and egressive (outgoing) breath direction. The implications for theories of pre-speech development are discussed.

The nature of early games and how they might assist the infant in language acquisition were explored in a longitudinal study of two mother–infant dyads, using video-recordings of their free play. Analysis of appearance and disappearance games, in particular, revealed: (1) a restricted format, with a limited number of semantic elements, and a highly constrained set of semantic relations; (2) a clear repetitive structure, which allowed both for anticipation of the order of events and variation of the individual elements; (3) positions for appropriate vocalizations which could in turn be used to mark variations; and (4) the development of reversible role relationships between mother and child.

Child phonologists have made varying assumptions concerning the nature of phonological characteristics of children's imitative utterances and their relationship to the characteristics of spontaneous utterances. This study examined this relationship in children in early Stage I. A task was devised which permitted an inspection of children's unsolicited imitations of nonsense words reflecting syllabic shapes and consonants in and out of the child's phonological system. The results are discussed with respect to production constraints, phonological idioms, and selection and avoidance rules operative in imitative speech during this period of development. The findings indicated that while these imitative utterances were subject to the same production constraints they were not subject to the same selection and avoidance rules operative in spontaneous speech.

Following Ferguson & Farwell (1975), this paper looks at phonological development in lexical terms and extends their method to consideration of syllables within words. Ferguson & Farwell dealt with word-initial consonants only; we attempt to determine whether the sets of phone classes in all positions are the same, whether differences that occur from position to position are primarily within the phone class or in the number and distribution of phone classes at each level in the tree, and what effect position in the syllable has on phone classes.

Although deictic terms are among the earliest words children acquire, the proximal/non-proximal contrast (the true deictic contrast) between here and there, and between this and that, takes several years to master. As research on spontaneous production shows, children may start, for example, by using here with a deictic meaning, there with a non-deictic meaning, and a gesture to indicate a deictic contrast. On the basis of two experiments on comprehension, we argue that children go through at least three stages in acquiring the deictic contrasts. They start with NO CONTRAST, work out a PARTIAL CONTRAST used only in certain contexts, and finally master a FULL CONTRAST equivalent to the adult's. However, children follow different routes through these stages, depending on their initial choice of (a) the point of reference for the contrast – themselves or the speaker – and (b) the spatial relation to that point of reference – proximity or distance.

Fifty children between the ages of 3; 0 and 5; 6 were asked six types of wh- questions following videotaped sequences. It is argued that differential difficulty of various forms of who questions supports a parallel model of information retrieval and processing during discourse. The differential difficulty of why and when questions are, however, attributable to a necessary progression in the ability to encode the relevant concepts linguistically. Responses to how questions are difficult because they involve a number of unrelated skills.

The process of topicalization occasionally produces pleonasms with nominal-pronominal reduplication (NPR). In these structures the topic nominal is often accompanied by a comment clause containing a pronoun which is co-referential with the topic. This article discusses NPR in the language of children ages seven and nine in four different situations, one of which was child-child in free play, three of which were child-adult in retelling stories, describing pictures, and answering questions. The younger children produced more NPR. Influenced by their attitudes toward their interlocutors, the content of the situations and themselves as communicators, all of the children produced little NPR talking to their peers and increasing amounts talking to adults. Examples are given and analysed.

This paper rebuts a recent article in this journal (Petretic & Tweney 1977), which claimed to overthrow prior findings of ours. We show that subject populations, methods, and scoring procedures differed materially in the two studies, logically disallowing comparative interpretation of them. We demonstrate that the original study concerned syntactic organization; and that interpretation of it, or of the recent study, as bearing on comprehension is largely unjustified. Finally, we comment on our prior work in the context of new developments and findings about child language which postdate its publication in 1969.