Eve Clark's comprehensive book makes an important addition to those works already available on first language acquisition. A commendable feature is the inclusion of many non-English examples, although most are from English. In the field of language acquisition there has been an emphasis on cross-language comparisons over the past few years, as shown by the number of journal articles and chapters published, but there has not been a comprehensive single-authored book which has attempted to incorporate the recent cross-language findings. Clark's work goes a long way to filling this gap.

Throughout the volume the emphasis is on the social setting of acquisition, combined with the cognitive foundations on which children build. Language is viewed as the product of social interaction. In adopting a process account of language acquisition, Clark takes account of the dynamic nature of conversation rather than viewing structures in isolation.

The volume is divided into four parts. The first examines conversations between adult and child, segmentation of the speech stream, and the child's early words. The second part focuses on structure – that is, syntax and morphology – as well as the coining of new words by young children. The social dimension is the focus of part 3, while part 4 examines the biological specialization for language. A number of recurring themes link the chapters across the four sections, including the distinction between comprehension and production, the conservative nature of young children's productions, the richness of the input, and the child's contributions to the acquisition process. The chapters cover much of the acquisition research undertaken to date, although Clark only briefly mentions the formal grammar/nativist perspective. Some examples are taken from Clark's diary study of her son's spontaneous speech, and others from comprehension and production data reported by others.

The general introduction (chap. 1) poses numerous questions that face acquisition researchers as well as providing a brief introduction to some of the terminology used in the field. In chap. 2, Clark argues that the four general conditions for communication – attending to the same focus (joint attention), establishing common ground, using appropriate speech acts, and listening in order to make appropriate responses – are relevant even in the earliest exchanges between adult and child. These four conditions are revisited throughout the book; for example, chap. 12 elaborates on the conditions pertaining to speech acts and appropriate responses.

The view that emerges throughout the work is that learning a language is part of learning to communicate. The nature of early interactions between babies and adults, Clark argues, represents the turn-taking of adult conversation. Thus, even a young child is shown how to be a partner in interaction. With examples from English, Clark discusses how infants learn to get an adult's attention, and proposes that by the age of 2 years infants can focus on communicating their intentions, are able to clarify what they want when there is misunderstanding on the part of the adult, and consider the knowledge state of the adult when making a request. In sections covering the structure and functions of “Child Directed Speech,” Clark discusses the fact that not all cultures modify their speech to infants in the same way. However, given the emphasis on the social context of acquisition, it would have been appropriate to focus a little more on the fact that in many cultures infants grow up in multiparty contexts; thus, research findings that focus on joint attention between mother and child will be more relevant to some cultures than to others. Clark suggests that we focus on the similarities rather than the differences, but it is only by documenting and interpreting the socialization patterns of children in different cultures that we identify how infants and toddlers do attend to the language around them. Just as language typology influences acquisition, so do socialization patterns.

The main focus of chap. 3 is how infants analyze the speech stream. The research on babies' early perceptions and the reorganization of perceptual biases is well covered. Infants need to recognize recurring patterns so they can attach meaning, and the view presented in this chapter is that infants discover the sound system in trying to work out the communicative significance of different utterances. The clear mismatch between a child's production and how the form is represented in memory is discussed; this topic is expanded in chap. 5. The gradual emergence of words from early communicative exchanges and gestures is discussed in chap. 4. Clark argues that adults direct infants' attention while talking about objects and events in the here-and-now; infants bring conceptual categories to the task of word learning. Drawing on research from various contexts, Clark raises the issue of whether it is possible to classify the infant's early words as nouns or verbs based on the adult system. Categorization, she decides, is better identified by the child's use of words.

The development of sounds is the topic of chap. 5. What is critical for the process of acquisition is the asymmetry between comprehension and production, according to Clark; this point is taken up again in the concluding chapter of the book. She suggests that the retention in memory of representations starting at around 9–10 months allows infants to recognize words, starting with familiar chunks in the speech stream. Gradually infants are able to fill in more about meaning and form. Infants' representations also serve as targets for their productions, something against which to check their outputs. In the final chapter of part 1, Clark evaluates the constraints approach to acquisition. She raises questions about where constraints would come from, where they would start, how long they would last, and why they would be abandoned. The alternative, which Clark prefers, is that children build on conceptual categories in combination with pragmatic information.

Part 2 elaborates on the continuity of language acquisition. In chap. 7, Clark argues that sequences of single words are planned together and so mark the emergence of structural relations. She also discusses, among other things, evidence for categories and evidence for productivity. In chap. 8, which looks at the modulation of word meaning, language typology is discussed in relation to acquisition. A section of the chapter elaborates on rule vs. schema approaches to past tense acquisition; Clark points out that while it is difficult to distinguish between the two approaches at the end point (i.e., when past tense verb forms have been acquired), the issue is whether children rely on rules or schemata in acquiring the forms. The dual encoding hypothesis is also discussed, but, using a number of examples to illustrate the problems it raises, Clark argues that while it is convincing for English it is not for other languages.

Chap. 9 includes discussion on a number of topics, including linking rules, preferred argument structure, questions, negatives, locative and causative alternations, and passives. Chap. 10 focuses on clause combining. Clark proposes that complement clauses of verbs such as think and know actually function as evidentials for young children; in support of this view is that their distribution is restricted to first and second person subjects. In chap. 11, Clark draws on her previously published work reporting on compounding and derivation, and also discusses other aspects of word formation. The topic of asymmetry between production and comprehension is revisited.

Evidence and arguments to support the view that learning a language is part of learning to communicate are drawn together in part 3. The four conditions needed for communication which were introduced in chap. 2 are again taken up, in particular the two conditions “using appropriate speech acts” and “listening in order to make appropriate responses.” Social roles, register, resolving conflict, and related issues are some of the topics covered, as is “stage direction.” Clark illustrates (from English) that by 3; 6 to 4 years, children can direct each other and role-play; that is, they distinguish play from the real world. Chap. 15 looks at the early language development of children raised with input from two languages. Differences in input across social classes and the effect on the child's vocabulary development is another topic discussed.

Part 4 is quite short, and the topics are not as well developed as those in other chapters. Chap. 15 contains an overview of the specialization for language, with discussion of sensitive periods and methods of studying brain activity (PET, ERP and fMRI). As is typical in mentions of critical periods, there is a brief discussion of feral and isolated children. The final chapter takes up a very crucial aspect of any account of language acquisition: the mechanisms that cause change in the system. The child's internal representations – for understanding others and as targets for production – change over time. From the early detection of sounds and patterns in the input, the child develops a language system and knowledge of appropriate use, gradually drawing closer to the adult target. Children start small and build up their knowledge; what influences change seems to be the child's developing ability to take up what is offered in the input. Clark introduces models of learning here, and issues such as the poverty of the stimulus, but there is limited discussion of these topics.

No one volume could do justice to the field of language acquisition. In this book of over 500 pages, Clark argues convincingly that language acquisition starts very early in a social context, and that social context cannot be separated from language acquisition. The issues raised in the early chapters are elaborated through the volume, with a few of the examples repeated. It is a difficult task to cover the material without some overlap since there are numerous strands to be woven together: different aspects of the learning environment, the child's development, the language features that emerge at different stages, and possible explanations. In fact, having some overlap across the chapters helps them stand alone. For example, as readings for an advanced undergraduate group without any background in acquisition research or theory, I selected just four chapters from different sections of the book. The students found the chapters easy to follow. Most technical terms were explained appropriately. However, we did note that semantic bootstrapping was mentioned on p. 189 but not explained until p. 201 (and the index erroneously lists page 199 for 189).

In conclusion, the book provides remarkable coverage of the acquisition process from a social/pragmatic perspective. It is a valuable resource for students and researchers of language acquisition.