The context-of-utterance

Any utterance-token that is produced on some particular occasion is an actual utterance (cf. 1.6). In certain situations, the utterance that is produced (as a token of a particular type) is very highly determined by factors which we may describe, loosely for the moment, as contextual. For example, the utterance of Hello when answering the telephone or of Good morning upon entering a shop at a certain time of day is highly determined by the social role that the utterer is playing and his recognition of what utterance-types are appropriate to this role and by a variety of more particular contextual features. Generally speaking, however, we can say that actual utterances are in contrast with indefinitely many potential utterances which might have been actualized on the occasion in question, but were not.

Every actual utterance is spatiotemporally unique, being spoken or written at a particular place and at a particular time; and, provided that there is some standard system for identifying points in space and time, we can, in principle, specify the actual spatiotemporal situation of any utterance-act (which has as its product an actual utterance-signal: cf. 1.6) by giving its spatiotemporal co-ordinates within the framework of the standard system. We can say, for example, that a particular utterance-token was produced by X at 12 noon on 6 January 1971, in Edinburgh; and we can be more or less precise than this in our specification of the spatiotemporal co-ordinates of the utterance-act.

Parts-of-speech, form-classes and expression-classes

In this and the immediately following sections, we shall be concerned primarily with two questions: (i) Do all languages have the same parts-of-speech (i.e. nouns, verbs, adjectives, etc.)? (ii) To what degree are semantic considerations relevant to the definition of such terms as ‘noun’, ‘verb’ or ‘adjective’? These two questions, as we shall see, are intrinsically connected. Curiously enough, they are only rarely discussed nowadays. And yet they are crucial in any treatment of the relation between grammar and semantics.

Although most of the published grammars and dictionaries of particular languages make use of the traditional terms ‘noun’, ‘verb’, ‘adjective’, etc., the standard definitions of such terms have long been criticized by linguists as being unsatisfactory in several respects. It has been argued that they are circular; that they depend upon a mixture of morphological, syntactic and semantic criteria, which do not necessarily coincide in particular instances; and that they are inapplicable to languages whose grammatical structure differs significantly from that of the classical Indo-European languages. The approach which, in outline only, we present here concedes that there is considerable force in these criticisms, but also gives due recognition to those aspects of the traditional theory which are relevant to the central concerns of this book and, with certain qualifications and clarifications, are of enduring validity.

Lexical entries

Conventional dictionaries are essentially lists of what might be called lexical entries. Each of these entries is introduced by a head-word in its standard orthographic representation; and the lexical entries are alphabetized in terms of their head-word. Alphabetization is of course no more than a technique for listing the entries according to a conveniently applicable, but theoretically irrelevant, principle. The conventional dictionary can, for our purposes, be thought of as an unordered set of lexical entries, each of which is indexed by means of its head-word.

The fact that the head-word is represented orthographically (and may or may not be furnished with a phonetic or phonological transcription) is something that will not concern us here. We should not forget, however, as linguists, that most adult native speakers of English are accustomed to thinking of word-forms as relatively stable written entities whose pronunciation may be somewhat variable. Homonymy, to which we will return presently, is traditionally based upon orthographic type-token identity (cf. 13.4); and this is something that the lexicographer cannot but be concerned with, since the organization of the conventional dictionary depends upon it. Faced with the fact that the noun ‘bank’ (whose written forms are bank, banks, bank's and banks') has several different meanings, he must decide how many lexical entries (all indexed by the head-word bank) he will put into his dictionary.

Levels of analysis

Most linguists distinguish at least three levels of structure in their analysis of sentences: the phonological, the syntactic and the semantic. To these three they may or may not add morphology to serve as a bridge between the syntax and the phonology in particular languages.

Looked at from the point of view of its phonological structure, every sentence may be represented as a sequence of phonemes with a certain prosodic contour superimposed upon it (cf. 3.1). The phonemes of a language are conventionally represented by means of letters enclosed within a pair of oblique strokes. For example, there is in English a phoneme /b/ which occurs in the initial position of the forms bed, bread, boil, etc., and is pronounced as a bilabial, voiced, non-nasal stop; and this phoneme, like all the other phonemes of English, has a characteristic distribution throughout the word-forms of the language. It is part of the phonologist's job to list, for the language that he is describing, all the phonemes that occur in that language and to specify the principles which determine their co-occurrence, or combination, in actual and potential word-forms. He will tell us, for example, that the combination of /b/ with /n/ is impossible in the first two positions of English word-forms; and he may account for this in terms of the more general principle that stop consonants do not precede nasal consonants in English at the beginning of a syllable.

Person-deixis

The term ‘deixis’ (which comes from a Greek word meaning “pointing” or “indicating”) is now used in linguistics to refer to the function of personal and demonstrative pronouns, of tense and of a variety of other grammatical and lexical features which relate utterances to the spatiotemporal co-ordinates of the act of utterance. As employed by the Greek grammarians, the adjective deictic (‘deiktikos’) had the sense of “demonstrative”, the Latin ‘demonstrativus’ being the term chosen by the Roman grammarians to translate ‘deiktikos’ in the works of the Stoics, of Dionysius Thrax and of Apollonius Dyscolus, which laid the foundations of traditional grammar in the Western world. It is worth noting that what we now call demonstrative pronouns were referred to as deictic articles in the earlier Greek tradition and that the Greek word ‘arthron’, from whose Latin translation, ‘articulus’, the technical term article derives, was no more than the ordinary word for a link or joint. It was only in the later tradition that the Greek equivalent of ‘pronoun’ was used; and this fact is of some significance. The point is that in early Greek, no sharp distinction can be drawn, in terms of their forms or syntactic and semantic function, between demonstrative pronouns, the definite article and the relative pronoun: the term ‘article’ was at first applied to them all, and it was chosen, presumably, because they were regarded as connectives of various kinds.

When I began writing this book six years ago, it was my intention to produce a fairly short one-volume introduction to semantics which might serve the needs of students in several disciplines and might be of interest to the general reader. The work that I have in fact produced is far longer, though in certain respects it is less comprehensive, than I originally anticipated; and for that reason it is being published in two volumes.

Volume 1 is, for the most part, more general than Volume 2; and it is relatively self-contained. In the first seven chapters, I have done my best, within the limitations of the space available, to set semantics within the more general framework of semiotics (here defined as the investigation of both human and non-human signalling-systems); and I have tried to extract from what ethologists, psychologists, philosophers, anthropologists and linguists have had to say about meaning and communication something that amounts to a consistent, if rather eclectic, approach to semantics. One if the biggest problems that I have had in writing this section of the book has been terminological. It is frequently the case in the literature of semantics and semiotics that the same terms are employed in quite different senses by different authors or that there are several alternatives for what is essentially the same phenomenon.

Opposition and contrast

The notion of sense* (as distinct from denotation* and reference*) has already been introduced (chapter 7). Our purpose in this chapter is to develop and reformulate what seem to be the basic principles of the theory of semantic fields in terms of sense-relations* (i.e. relations of sense holding within sets of lexemes) without postulating any underlying conceptual or perceptual substance (cf. 8.4). The treatment will be relatively informal and at times somewhat speculative. We begin by discussing the notion of paradigmatic opposition.

From its very beginnings structural semantics (and indeed structural linguistics in general) has emphasized the importance of relations of paradigmatic opposition*. Trier himself opens his major work (1931) with the challenging statement, that every word that is pronounced calls forth its opposite (seinen Gegenteil) in the consciousness of the speaker and hearer; and this statement can be matched with similar assertions by other structural semanticists. Trier, it will be noted, claims, as others have done, that the opposite is in some way present in the mind of the speaker and hearer during an act of utterance. Whether this is true or not is a psychological question, and one that is more relevant to the construction of a theory of language-behaviour than it is to the analysis of a language-system (cf. 1.6). In what follows we make no assumptions about what goes on in the mind of the speaker and hearer during an utterance.

Verbal and non-verbal signalling

The terms ‘verbal communication’ and ‘non-verbal communication’ are quite widely employed to distinguish language from other semiotic* systems: i.e. systems of signalling-behaviour. They are terms which, from the point of view adopted in this book, are doubly unfortunate: (i) ‘non-verbal communication’ is commonly applied to signalling behaviour in man and animals of a kind which, though it may be informative, is not necessarily communicative (cf. 2.1); (ii) ‘verbal communication’, in so far as it refers to communication by means of language, might be taken to imply that language-utterances are made up solely of words, whereas, as we shall see in this section, there is an important, and indeed essential, non-verbal component in spoken language. The use of such expressions as ‘verbal communication’ or ‘verbal behaviour’ to refer to language-behaviour is, at least potentially, misleading.

We may begin our discussion of spoken language by distinguishing between vocal* and non-vocal* signals, according to whether the signals are transmitted in the vocal-auditory channel or not. The vocal-auditory channel* is here defined it will be observed, in terms of its two endpoints and of the manner and mechanisms by means of which the signals are produced at the source and received at the destination, rather than simply in terms of the properties of the channel itself which links the terminals and along which the signal travels. This point in itself is worth noting, since there are alternative definitions to be found in the literature.

What is communication?

To say that language serves as an instrument of communication is to utter a truism. Indeed, it is difficult to imagine any satisfactory definition of the term ‘language’ that did not incorporate some reference to the notion of communication. Furthermore, it is obvious, or has appeared so to many semanticists, that there is an intrinsic connexion between meaning and communication, such that it is impossible to account for the former except in terms of the latter. But what is communication? The words ‘communicate’ and ‘communication’ are used in a fairly wide range of contexts in their everyday, pre-theoretical sense. We talk as readily of the communication of feelings, moods and attitudes as we do of the communication of factual information. There can be no doubt that these different senses of the word (if indeed they are truly distinct) are interconnected; and various definitions have been proposed which have sought to bring them under some very general, but theoretical, concept defined in terms of social interaction or the response of an organism to a stimulus. We will here take the alternative approach of giving to the term ‘communication’ and the cognate terms ‘communicate’ and ‘communicative’ a somewhat narrower interpretation than they may bear in everyday usage. The narrowing consists in the restriction of the term to the intentional transmission of information by means of some established signalling-system*; and, initially at least, we will restrict the term still further – to the intentional transmission of factual, or propositional, information.

Structuralism

In this section we shall be concerned with the more general principles of what is commonly known, in Europe at least, as structural linguistics*. Unfortunately, the term ‘structuralism’ has acquired a somewhat different, and much narrower, sense in the United States, where it now tends to be employed with reference to the theoretical and methodological principles of the so called post-Bloomfieldian school, which was dominant in American linguistics in the period immediately following the Second World War. Many of the principles of post-Bloomfieldian structuralism were not only alien to, but at variance with, the principles of what we may here refer to (for reasons which will be explained below) as Saussurean* (including post-Saussurean) structuralism. We need not go into all the differences between post-Bloomfieldian and Saussurean structuralism. Most of them are irrelevant in the present context. What must be emphasized, however, in view of the polemical associations which attach to the term ‘structuralism’ in the works of Chomsky and other generative grammarians (cf. 10.5), is that there is, in principle, no conflict between generative grammar and Saussurean structuralism, especially when what we are calling Saussurean structuralism is combined, as it has been in certain interpretations (as we shall see below) with functionalism* and universalism*. In particular, it should be noted that Saussurean structuralists, unlike many of the post-Bloomfieldians (for whom ‘structural semantics’ would have been almost a contradiction in terms), never held the view that semantics should be excluded from linguistics proper.