What we have here, then, is a hypothesis about the nature of language acquisition. So far, it has not been tested in any convincing way (and it may not be possible to test it, in the usual sense); but it has provoked a great deal of speculation. In particular, it raises the question of how far the innate features could be identified with the primitive meaning-relations of grammatical theory - that is, the linguistic universals talked about at the end of Chapter 4. Are all children born with an ability to discriminate 'subjects' from 'objects', let us say, in some sense? How many such basic relations might one plausibly ascribe to the child? And how specific is its innateness? Clearly, it is not possible to suggest that the child has any features of a particular language innate, for instance a particular feature of English syntax which does not occur in French or German. To suggest this would be tantamount to saying that children of any race would find it easier to learn English than to learn other languages (that is, their brains would predispose them towards English); and all available evidence points to the implausibility of this conclusion. A Zulu child learns Zulu just as rapidly as an English child learns English, it seems. No, the innate features must be sufficiently general, sufficiently 'deep', to be capable of equally readily underlying the structure of any language. And on this point, the identity of interests between linguistic and psycholinguistic theory (at least, in this field) should be clear. There have of course been a number of objections raised to the innateness hypothesis - for example, on the grounds that what is innate is not so much deep structural information, but rather learning principles of a more general kind. Some people would like to see what would happen if the hypothesis were formulated in terms other than those provided by Chomsky's later work. As someone put it once, 'Why should we see the child as if it were born with a copy of Aspects of the Theory of Syntax tucked inside its head!' Unkind, perhaps; for without Aspects, and the work which followed it, many interesting questions might never have been raised. The issue, however, is by no means determined.
In the 1980s, the interest in the innateness hypothesis has been largely replaced by a focus on the relationship between language development and a child's cognitive skills, following on the influential work of Jean Piaget and other psychologists. There has been renewed interest in the strategies which children use in acquiring language, and the significance of such topics as imitation has come to be reconsidered in this light. Above all, there has been a concern to study the factors which characterize children's learning environment - in particular, the nature of the input language they receive from mothers and other caretakers (motherese). The Journal of Child Language, which commenced publication in 1975, is now the best source of information on current trends in the subject. Its contributors span the disciplines of psychology and linguistics, and their work illustrates a wide range of experimental and naturalistic approaches to the subject. Without doubt, the field of language acquisition remains one of the most intriguing areas of linguistics study, at the present time, and one which will certainly remain in the forefront of linguists' attention over the next few years.
There are many other applications of linguistics in fields not so far mentioned, which tend to be grouped together anonymously as 'applied linguistics'. Foreign language teaching and learning is the major application, as suggested in Chapter 1; but there is also native language teaching, translation (either individually, or using machines), the many facets of telecommunications, lexicography . . . The list could go on for some time. Each of these fields selects its basic information and theoretical framework from the overall perspective which linguistics provides, and applies it to the clarification of some general area of human experience. And it is surely the many branches of applied linguistics that will ultimately provide the main link between Chapters 1 and 4, if such a link be needed. But, as always, we must remember that an application is but the tip of a theoretical iceberg: many hours of research and discussion, much of it highly specialized, abstract, and quite unpractical, will have taken place in order to provide the basic knowledge which can be implemented in a specific application. Indeed, in many cases it is only through the illuminating models developed in linguistic theory, and the demonstration of a coherent system underlying apparently disorganized data, that applications and approaches to a problem have been thought of at all.
[...] 'What does it matter', such queries run, 'whether the basic phonological unit is the phoneme or the distinctive feature? or whether the morpheme concept fits all cases? or whether there is a boundary-line between syntax and semantics?' If these questions are still being asked, then the arguments underlying my Chapter 3 about the scientific aims of linguistics have not been appreciated. The kinds of distinction drawn there are essential if we hope to build up a general theory of language; we have to appreciate the kinds of reasoning relevant to this task, even if we do not always agree with the conclusions reached. If we are adopting a rational approach to our study of (or interest in) language, then we cannot just blindly analyse and describe in a random, arbitrary way. Whatever our purpose, whether 'pure' or 'applied', we must know why we are doing what we are doing, if we hope to be clear and consistent and wish to convince others (or even ourselves) of its validity. It does matter about these questions, and many others like them, because the answers constitute our world-view of language. Choosing to work with distinctive features is one choice we make, along with many others, which ultimately builds up a coherent and self-consistent picture of language structure that intuitively satisfies us. We sit back and say, 'Yes, that makes sense.' To a certain extent, then, our final decisions about which concepts to work with are a matter of taste. But the more we understand the relative merits and demerits of the various theories, descriptions and procedures which the subject provides, the more likely we will be to reach a view of language that is reasonable and convincing, as well as personally satisfying.
Kennedy G. An Introduction to Corpus Linguistics.
London and New York: Addison Wesley
Longman Limited, 1998. – pp. 1-12.
In the language sciences a corpus is a body of written text or transcribed speech which can serve as a basis for linguistic analysis and description. Over the last three decades the compilation and analysis of corpora stored in computerized databases has led to a new scholarly enterprise known as corpus linguistics. The purpose of this book is to introduce the various activities which come within the scope of corpus linguistics, and to set current work within its historical context. It brings together some of the findings of corpus-based studies of English, the language which has so far received the most attention from corpus linguists, and shows how quantitative analysis can contribute to linguistic description. It is hoped that, by concentrating in particular on some of the results of corpus analysis, the book will whet the appetites of the growing body of teachers and students with access to corpora to discover more for themselves about how languages work in all their variety. The book is intended primarily for those who are already familiar with general linguistic concepts but who want to know more of what can be done with a corpus and why corpus linguistics may be relevant in research on language. Corpus linguistics is not an end in itself but is one source of evidence for improving descriptions of the structure and use of languages, and for various applications, including the processing of natural language by machine and understanding how to learn or teach a language.