This issue contains a new Review Article and Discussion section. The idea is to ask a leading researcher to contribute an article which can either be a review of a field or an extended book review and to then ask for commentaries on the piece by other researchers. In this issue the section consists of a commentary by Michael Tomasello on Adele Goldberg's Constructions: a construction grammar approach to argument structure together with twelve commentaries and a reply. The next of these sections will be an extended review by Matthew Rispoli of Rethinking innateness: a connectionist perspective on development by Jeffrey Elman, Elizabeth Bates, Mark Johnson, Annette Karmiloff-Smith, Domenico Parisis and Kim Plunkett. I very much hope that readers will find the new section an interesting and useful contribution to theoretical and methodological discussion in the field. Of course the success of this venture depends on contributions from a wide range of theoretical perspectives and on people's willingness to make these contributions even where they sharply disagree with the thesis of the review article.

Derived nominals are abstract nouns derived from verbs and adjectives by nominalization. This study traces the route taken by Hebrew-speaking children in the acquisition of Hebrew derived nominals (HDNs) of two types: action nominals which conform in structure to the small set of obligatory verb patterns (binyanim) (e.g. ktiva ‘writing’) and deverbal nouns, which belong to separate nominal patterns (e.g. maga ‘touch’). One hundred native speakers of Hebrew (children aged 5, 8, 11 and 15, and adults) were tested on comprehension and production of HDNs. The test items were grouped by binyan patterns and by morphological regularity. Results showed that acquisition of HDNs starts at about age 8 and is not complete by age 15, and that task type, binyan pattern and morphological regularity all affect ease of acquisition. We consider the strategies employed in the course of acquisition of HDNs and offer an explanation for this late acquisition which requires a vast amount of prior integrated knowledge.

Children's overextensions (e.g. referring to a pomegranate as apple) raise intriguing questions regarding early word meanings. Specifically, how do object shape, taxonomic relatedness, and prior lexical knowledge influence children's overextensions? The present study sheds new light on this issue by presenting items that disentangle the three factors of shape, taxonomic category, and prior lexical knowledge, and by using a novel comprehension task (the screened-alternative task) in which children can indicate negative exemplars (e.g. which items are notapples). 49 subjects in three age groups participated (Ms=2;0, 2;6, and 4;5). Findings indicate: (1) Error patterns differed by task. In production, errors were overwhelmingly due to selecting items that matched the target word in both shape and taxonomic relatedness. In comprehension, more errors were based on either shape alone or taxonomic relatedness alone, and the nature and frequency of the overextensions interacted with prior lexical knowledge. (2) Error patterns also differed markedly based on the word being tested (apple vs. dog), in both comprehension and production. (3) As predicted, errors were more frequent in production than comprehension, though only for children in the two younger age groups. Altogether, the study indicates that overextensions are not simply production errors, and that both taxonomic relatedness and object shape play a powerful role in early naming errors.

This article contributes to a debate in the linguistic and psychological literature that centres around the representation of morphologically complex words in the grammar and in the lexicon. The issue is whether inflectional morphology is rule-based (i.e. symbolically represented), or whether the assumption of pattern association is more adequate to account for the facts. On the basis of the analysis of acquisitional data the article strongly argues for the latter alternative. In a classic experiment that helped shape the development of acquisition theory Berko (1958) reported substantial support for item-and-process rules in the acquisition of plural morphology in English. A large part of her results were zero responses (repetition of the stimulus). A reinterpretation of these zero responses in light of schema theory and the cue strength hypothesis shows a striking departure from randomness. Berko's subjects tended to repeat stimuli just to the extent that these already resembled a plural schema. A reinterpretation of data reported in Innes (1974) achieved compatible results. This data set is far more extensive than Berko's and is used in the present study to put the schema model to a more stringent test. A reinterpretation of a parallel experiment with German children, using the cue strength analysis of the more complex plural morphology of German yielded parallel results. Finally, natural acquisitional data obtained from seven German speaking children aged between 2;1 and 2;9 are analysed. Again, strong support is found for the schema model. It is suggested that a schema-learning mechanism may underlie the acquisition of morphology, even when the end product of the learning process involves item-and-process rules, as in the case of English plural formation. In a schema-learning model, the child builds schematic representations for possible singular and plural lexical items as whole gestalts, and attempts to map concrete forms onto these schemata in deciding whether the forms have singular or plural value.

This study extends prior investigations of children's abilities to conceptualize distinctive phonological features in development. The purpose was to establish the relationship between a child's productive and conceptual knowledge, and to examine its potential influence on phonological change. Six children evidencing phonological disorders (aged 3;4 to 5;7) participated in evaluations of the productive sound system and the metalinguistic ability to classify phonological information, with particular emphasis on fricatives. Then, children were enrolled in clinical treatment aimed at accurate production of fricatives. Following treatment, production and classification abilities were again tested. Results indicated that if a child's productive knowledge of distinctive featural contrasts changed following treatment, there was also a corresponding change in conceptual knowledge. The specific featural dimensions that were used to classify phonological information were either in parallel to, or in advance of, those features also used in production by the child. Conceptual knowledge was thus equal to or better than productive knowledge for these children.

The effect of regional accent on children's processing of speech is a theoretically and practically important aspect of phonological development that has been little researched. 48 children from London, aged four and seven years old, were tested on their ability to repeat and define single words presented in their own accent and in a Glaswegian accent. Results showed that word comprehension was significantly reduced in the Glaswegian condition and that four-year-olds performed less successfully than seven-year-olds. Both groups made similar numbers of lexical misidentifications, but the younger children were more likely to fail to access any word at all. On the repetition task, the younger children showed a different pattern of errors to the older children, their productions being apparently more influenced by the phonetics of the Glaswegian stimuli. It is suggested that such phonetic responses are related to the younger children's failure to map the unfamiliar accent onto their own phonological representations. It is proposed that the lexical misidentifications, common to both age groups, are more likely to be induced by lack of context. The paper concludes with discussion of implications of these findings for our understanding of how children develop the ability to process unfamiliar regional accents.

This study examines sources of individual variation in child vocabulary competence in the context of a multivariate developmental ecological model. Maternal sociodemographic characteristics, personological characteristics, and vocabulary, as well as child gender, social competence, and vocabulary competence were evaluated simultaneously in 126 children aged 1;8 and their mothers. Measures of child vocabulary competence included two measures each of spontaneous speech, experimenter assessments, and maternal reports. Maternal measures, from proximal to distal, included vocabulary, verbal intelligence, personality, attitudes toward parenting, knowledge of parenting, and SES. Structural equation modelling supported several direct unique predictive relations: child gender (girls higher) and social competence as well as maternal attitudes toward parenting predicted child vocabulary competence, and mothers' vocabulary predicted child vocabulary comprehension and two measures of mother-reported child vocabulary expression. In addition, children's vocabulary competence was influenced indirectly by mothers' vocabulary, social personality, and knowledge of child development. Maternal vocabulary itself was positively influenced by SES, maternal verbal intelligence, and mothers' knowledge about parenting. Individual variation in child vocabulary competence might best be understood as arising within a nexus of contextual factors both proximal and distal to the child.

We report a study of the acquisition of colour terms by Russian children which had two main aims: first, to test Berlin & Kay's (1969) theory of colour universals using acquisition order as a measure of basicness; and secondly, to see if the two blue terms of Russian are genuinely basic. Two hundred children aged from three to six-years-old were tested on three colour-tasks – colour term listing, colour term production and colour term comprehension. To a reasonable approximation, the order of colour term acquisition was in accord with Berlin & Kay's theory, but the data are also consistent with the weaker claim that primary terms tend to be learned before derived terms. On balance the data were consistent with Russian exceptionally, having an extra term for the blue region. But, the two blue terms – goluboj ‘light blue’ and sinij ‘dark blue’ – were confused more often than other pairs of terms even by the five- to six-year-old sample.

According to the storage hypothesis (Kail & Leonard, 1986), word-finding deficits in young children are not the direct result of deficient retrieval strategies; they are a manifestation of a general delay in language development that affects lexical storage. In the current study, we explored one aspect of lexical storage, the hierarchical organization of the semantic system, in 13 preschoolers with word-finding deficits (WF) and 13 preschoolers with normal language abilities (ND), ranging in age from 3;3 to 6;7. The children named a series of objects at multiple levels of the noun hierarchy in response to contrast questions (e.g. for rose they were asked, ‘Is this an animal?’ to elicit plant [superordinate]; ‘Is this a tree?’ to elicit flower [basic]; ‘Is this a dandelion?’ to elicit rose [subordinate]). Both groups readily named at multiple levels, providing evidence of hierarchical organization of the lexicon. However, there were several differences between WF and ND groups that suggested that WF children did not have enough stored information to discriminate between similar semantic neighbours. We conclude (1) that hierarchical organization of the semantic lexicon is a robust developmental phenomenon, apparent in both ND and WF preschoolers and (2) that the word-finding deficits of preschoolers appear to reflect insufficient depth and breadth of storage elaboration rather than deficits in hierarchical semantic organization.

Review essay on: Goldberg, A. Constructions: A construction grammar approach to argument structure. University of Chicago Press (1995). Pp. xi+265.

The cornerstone of traditional descriptive grammars is the construction: a recurrent pattern of linguistic elements that serves some well-defined communicative function. Prototypical constructions are sentence-level patterns such as, in English: the imperative, the ditransitive, the passive, the resultative, the yes–no question, and the cleft (each of which may have some subtypes). Also included in some theorists' definition of construction are components of sentences such as the prepositional phrase, the noun phrase, or the genitive noun phrase. Traditional constructions may have some specific words or morphemes associated with them (e.g. by in the full passive, 's in the genitive), but these are generally closed-class morphemes. Almost by definition, traditional constructions are relatively abstract patterns that apply across whole classes of open-class morphemes.

One of the defining features of modern-day generative grammar is the absence of constructions. Chomsky (1981) hypothesized that grammatical structure comprises two primary levels: the level of principles and parameters, which is much more abstract than constructions and includes everything from the subjacency constraint to the empty category principle, and the level of the lexicon, which includes all of the concrete morphemes and words of a particular language. In this view, constructions represent a ‘middle level’ of analysis that is, in effect, an epiphenomenon resulting from the interaction of the two primary levels. One outcome of this theoretical move has been that generative linguists concerned with construction-level phenomena have had to fill the generative lexicon with ever richer types of linguistic information, especially for verbs (e.g. Bresnan, 1982; Jackendoff, 1990; Levin, 1995; Pinker, 1989).

Tomasello's stimulating review of Goldberg's (1995) book, Constructions: a construction grammar approach to argument structure, raises several themes worthy of discussion. Tomasello suggests numerous reasons why functional and cognitive linguistic approaches in general, and Goldberg's approach in particular, are central to current work in child language. In my commentary I will summarize Tomasello's central claims about what he sees as significant about Goldberg's book, and I will raise the question of whether construction grammar – rather than any other cognitive or functional approach – is worth pursuing. While generally sympathetic to Goldberg's approach, I will discuss two issues that I think are worthy of further consideration in future research.

Tomasello highlights three very important reasons why he believes Goldberg's approach makes a significant contribution to child language research. First, he states that construction grammar provides a way of understanding language development as a whole, and not just particular aspects of language development, such as core grammar. Second, Tomasello characterizes Goldberg's approach as noteworthy because it provides a way of relating language development to other domains of human cognition. A third advantage, Tomasello claims, is that construction grammar allows for the view of language development as protracted (e.g. not instantaneous), something Tomasello suggests fits well with his own research findings (see Tomasello, 1992).

The commentators on this review raised many interesting and important points, and I am grateful to each of them for their time and their thoughtfulness. Each commentary deserves a thorough discussion of its own, but unfortunately, due to extreme limitations of space, I will have to focus here on what I see as the major points raised by one or more of them.