Introductions should be short. They are often last written and last read (if they are read at all). But having an introduction is one of the universal conventions observed by authors - often a kind of self-indulgence, an opportunity to appeal to the considerateness of the reader, to try to anticipate criticism or reluctance to embark on the solid matter of the book itself. Some authors use it, at inordinate length (see Hegel, see Kant), to foreshadow the main content of the work, to claim novelty or certainty for what is proposed (see Wittgenstein's Tractatus or the Principia Mathematica) or to explain the author's personal motives for the choice of subject, the importance of the matter; others, with more justification, use it both to acknowledge intellectual debts, to pay tribute to precursors and to recognise the unavoidable limits of the author's personal knowledge of quite distinct fields of philosophical or scientific study. The more far-reaching the theoretical view presented in a book, the more need there is to concede the speculative and tentative nature of what is proposed, the extent to which, on specialised topics, there is a risk of misunderstanding, or error, or simply of not having kept up to date with a vast and rapidly growing literature. "At single points every philosophical treatise may be pricked (for it cannot be armed at all points like a mathematical one"(1) - Kant's comment on the Critique of Pure Reason (which he was convinced was in total unassailable) applies with even greater force to a work such as the present which attempts to bring together evidence and theories from many so far unrelated disciplines and form them into a coherent whole, drawing on linguistics, the physiology of perception, neurology, psychology and the philosophy of language and perception.

The starting-point for the book is the straightforward question: Where do these words come from? What is the source of the unbroken stream or river of language, which we all experience, both in talking to others and in formulating our own thoughts? The answer proposed (for which the whole volume is one long, cumulative argument) is that words, the fabric of language, are not arbitrary, a conventional cultural product of human ingenuity, but derive directly from, evolutionarily and physiologically, and are integrated with, perception and action, the other main components of total human behaviour. In asserting this, the book directly challenges the foundation assumption of modern linguistics that language is arbitrary and words are arbitrary. But science in the past has only progressed through challenging the unchallengeable. Kuhn, in his Structure of Scientific Revolutions, has illustrated how normal it is for new directions in science to start from questioning assumptions which no one of the scientific community of the time would venture to question. Though it is not claimed that the theory advanced here is of the same scientific importance, Kuhn records that Copernicus was called mad because he. proclaimed that the earth moved and his ideas made few converts for almost a century after his death; that the opponents of Newton said that his theories, with their reliance on innate forces, would return science to the Dark Ages, that Kelvin never accepted electromagnetism, Priestley never accepted the theory of oxygen (rather than that of phlogiston); many naturalists refused to accept Darwin's assertion that animal species had developed one from another or that such marvellously adapted organs as the eye and hand of man were products of a process that moved steadily from primitive beginnings rather than the product of distinct creation. The belief of linguists, and following them of philosophers and psychologists, in the arbitrariness of language is as unquestioning and undemonstrated as ever was the belief of earlier astronomers in the stationary earth or of earlier physicists in an all-pervasive ether. Kuhn comments that "an apparently arbitrary element, compounded of personal and historical accident, is always a formative ingredient of the beliefs espoused by a given scientific community at a given time"(2) and for the emerging community of the relatively new science of linguistics, the belief in the arbitrariness of the word is itself the arbitrary element of the kind Kuhn describes.

A few years ago, it would have been necessary to apologise for even venturing to discuss the origin of language, never mind for suggesting that words are not arbitrary, but such an apology is no longer necessary. The origin of language, so extensively discussed in the 18th century and earlier, is once again a living and respectable subject for scientists with many different types of expertise. The new interest in the origin of language was perhaps most clearly marked by the wide-ranging conference organised in 1975 by the New York Academy of Sciences(3) precisely on this subject, but a great deal of other work is now in progress bearing on the origin and development of language both in the individual human being and in the human race. Though linguists more than a century ago officially rejected any further discussion of the origin of language within the framework of linguistics and though.professional linguists still take this position, this has not prevented neurologists, psychologists, anthropologists and many others from pursuing their researches. As Roger Brown commented in his book Words and Things (writing as a psychologist), the subjects he discussed (the character of primitive language, the relations between language and thought, the nature of meaning) were "a set of real chestnuts, most of them either given up for dead, or demonstrated to be pseudo-questions or proscribed by scholarly societies ... but there is a lot of new evidence on these matters ... and no one today would suggest that the topics should be given up for dead"(4).

A legitimate function of an Introduction, it has already been said, is to acknowledge intellectual debts. Roger Brown would certainly be one of the creditors, not only for his discussion of sound symbolism but also for the mass of interesting evidence on the development of language in children presented in A First Language(5). Others to whom the ideas in this book owe a clear debt include both authors whose ideas anticipate or coincide with those presented and authors whose discussion of language has stimulated ideas which have gone to form the theory, not necessarily because they are in harmony but because opposing ideas were presented fully and seriously enough to require attention. This latter category would include, inevitably, Chomsky (his development of transformational grammar has clearly brought linguistics to a point of crisis), Whorf (though this book totally disagrees with his assertion of the priority of language over perception) and Wittgenstein. Wittgenstein has been an especially powerful influence because, at one time or another, he adopted such diverse views on the nature of language, and supported the opposing views with such ingenuity and vigour The rightness of the early Wittgenstein (on the view presented here) is precisely balanced by the wrongness of the later Wittgenstein. In his Notebooks of 1914-1916 (before he forced his ideas into the straitjacket of the Tractatus) his observations that "Language is a part of our organism and no less complicated than it", that names "themselves are connected; in this way the whole images the situation like a tableau vivant", "The name compresses its whole complex reference into one" and "Words are like the film on deep water"(6) - would have fitted well into the theory presented here, had he not subsequently equally sharply asserted the arbitrary and conventional nature of language and the uselessness of seeking any philosophical enlightenment from it. The same category of authors, those stimulating by disagreement, would have to include Locke(7). Despite his denial of the possibility of innate ideas and his assertion of the arbitrariness of words, Locke's recognition of language as the great instrument and common tie of society and his careful discussion of the relation of words and knowledge have provided material used later in the book. Amongst linguists proper, Saussure and Sapir have been influential, in expressing so clearly ideas with which the present theory disagrees.

The positive debts can be referred to more briefly because evidence of them is apparent at many points in the book. Most important of all have been the ideas of Karl Lashley as a neurologist on the structural relation between speech, vision and action; his view that the rudiments of every human behavioural mechanism will be found far down in the evolutionary scale, that the problems of syntax and of the organisation of language are characteristic of almost all other cerebral activity, that temporal integration is found similarly in language, vision and action, and that spatial and temporal order appear to be almost completely interchangeable in cerebral action (with integration carried out hierarchically at a series of levels), are the direct foundation for the central argument presented in this book. One might repeat here his observation that "the study of comparative grammar is not the most direct approach to the physiology of the cerebral cortex yet speech is the only window through which the physiologist can view cerebral life... language presents in a most striking form the integrative functions that are characteristic of the cerebral cortex"(8). Other important sources have been Lenneberg's pioneering Biological Foundations of Language(9) with its discussion of children's acquisition of language as a maturational process within a critical period, Richard Gregory's stimulating ideas on the 'grammar of vision'(10) and his speculation that language and vision are indeed based on common ground and the basic problems of both must be solved together. Last but far from least, Konrad Lorenz's(11) broad approach to the development and integration of animal and human behaviour as well as his study of the vitally important process of 'imprinting', that is, genetically-programmed neurological development, making it possible for the cerebral structures of the animal (or human being) to be modified, after birth, to match the specific environment, social or physical, to which the individual is in fact exposed.

The chapter headings indicate in a summary way the 'articulation and concatenation of the whole system', the systematic development of the hypothesis from chapter to chapter, starting from the elementary units of speech, vision and action and progressing to the interrelation of sentence, visual scene and complex action. However, whether the theory seems probable, well argued and convincing depends not on the outline but on the detailed presentation and argument both in each chapter individually and in all the chapters taken together as mutually supporting each other. How readily the new view presented will be given a hearing, or accepted, is subject not only to the usual and often scientifically justified suspicion of the unorthodox - Locke remarks that new opinions are always suspected and usually opposed without any other reason but because they are not already common - but also to two special considerations, first that there is a powerful community of professional linguists whose careers and work have been founded on an assumption totally incompatible with the basis of the present hypothesis and secondly that there is no existing community of scientists or philosophers whose interests range as widely as the assumptions and evidence presented in this book require. We live in an age of specialists and sub-specialists. Even philosophers. who once took all life and all knowledge as their field now are often specialists only in philosophy in a narrow sense, in one corner of philosophy. There are no general 'natural philosophers' in the academic community, though there are psychologists, sociolinguists, neurolinguists, physiologists, whose specialisms ultimately can only be comprehensible as part of a total science of human nature. Perhaps the nearest successors to the 'ancient philosopher' who took the whole of nature as his study are to be found, as Monod(12) suggests, among the biologists, or the sociobiologists - or amongst the exponents of artificial intelligence techniques. The most an author presenting a theory as wide-ranging as the present one can look for is that it should be treated on its merits, not dismissed out of hand. A theory of this kind must persuade; it cannot be cast in the form of a logical or mathematical proof. Hume, the great sceptic, at one point remarked that " a true sceptic will be diffident of his philosophical doubt"(13). All that the author of the present work would hope for is that those reading it will be prepared, provisionally, to be diffident of their certainties - and particularly of the certainty so generally prevailing that words and language must be arbitrary and cannot be natural.