§ 1. Referential Approach There are broadly speaking two schools to Meaning of thought in present-day linguistics representing the main lines of contemporary thinking on the problem: the referential approach, which seeks to formulate the essence of meaning by establishing the interdependence between words and the things or concepts they denote, and the functional approach, which studies the functions of a word in speech and is less concerned with what meaning is than with how it works.
1 See ‘Introduction’, § 1.
2 Sometimes the term semantics is used too, but in Soviet linguistics preference is given to semasiоlоgу as the word semantics is often used to designate one of the schools of modern idealistic philosophy and is also found as a synonym of meaning.
3 D. Bolinger. Getting the Words In. Lexicography in English, N. Y., 1973.
4 See, e. g., the discussion of various concepts of meaning in modern linguistics in: Л. С. Бархударов. Язык и перевод. М., 1975, с, 50 — 70.
All major works on semantic theory have so far been based on referential concepts of meaning. The essential feature of this approach is that it distinguishes between the three components closely connected with meaning: the sound-form of the linguistic sign, the concept underlying this sound-form, and the actual referent, i.e. that part or that aspect of reality to which the linguistic sign refers. The best known referential model of meaning is the so-called “basic triangle” which, with some variations, underlies the semantic systems of all the adherents of this school of thought. In a simplified form the triangle may be represented as shown below:
As can be seen from the diagram the sound-form of the linguistic sign, e.g. [dAv], is connected with our concept of the bird which it denotes and through it with the referent, i.e. the actual bird.1 The common feature of any referential approach is the implication that meaning is in some form or other connected with the referent.
Let us now examine the place of meaning in this model. It is easily observed that the sound-form of the word is not identical with its meaning, e.g. [dAv] is the sound-form used to denote a peal-grey bird. There is no inherent connection, however, between this particular sound-cluster and the meaning of the word dove. The connection is conventional and arbitrary. This can be easily proved by comparing the sound-forms of different languages conveying one and the same meaning, e.g. English [dAv], Russian [golub'], German [taube] and so on. It can also be proved by comparing almost identical sound-forms that possess different meaning in different languages. The sound-cluster [kot], e.g. in the English language means ‘a small, usually swinging bed for a child’, but in the Russian language essentially the same sound-cluster possesses the meaning ‘male cat’. -
1 As terminological confusion has caused much misunderstanding and often makes it difficult to grasp the semantic concept of different linguists we find it necessary to mention the most widespread terms used in modern linguistics to denote the three components described above:
sound-form — concept — referent
symbol — thought or reference — referent
sign — meaning — thing meant
sign — designatum — denotatum
For more convincing evidence of the conventional and arbitrary nature of the connection between sound-form and meaning all we have to do is to point to the homonyms. The word seal [si:l], e.g., means ‘a piece of wax, lead’, etc. stamped with a design; its homonym seal [si:l] possessing the same sound-form denotes ‘a sea animal’.
Besides, if meaning were inherently connected with the sound-form of a linguistic unit, it would follow that a change in sound-form would necessitate a change of meaning. We know, however, that even considerable changes in the sound-form of a word in the course of its historical development do not necessarily affect its meaning. The sound-form of the OE. word lufian [luvian] has undergone great changes, and has been transformed into love [lАv], yet the meaning ‘hold dear, bear love’, etc. has remained essentially unchanged.
When we examine a word we see that its meaning though closely connected with the underlying concept or concepts is not identical with them. To begin with, concept is a category of human cognition. Concept is the thought of the object that singles out its essential features. Our concepts abstract and reflect the most common and typical features of the different objects and phenomena of the world. Being the result of abstraction and generalisation all “concepts are thus intrinsically almost the same for the whole of humanity in one and the same period of its historical development. The meanings of words however are different in different languages. That is to say, words expressing identical concepts may have different meanings and different semantic structures in different languages. The concept of ‘a building for human habitation’ is expressed in English by the word house, in Russian by the word дом, but the meaning of the English word is not identical with that of the Russian as house does not possess the meaning of ‘fixed residence of family or household’ which is one of the meanings of the Russian word дом; it is expressed by another English polysemantic word, namely home which possesses a number of other meanings not to be found in the Russian word дом.
The difference between meaning and concept can also be observed by comparing synonymous words and word-groups expressing essentially the same concepts but possessing linguistic meaning which is felt as different in each of the units under consideration, e.g. big, large; to, die, to pass away, to kick the bucket, to join the majority; child, baby, babe, infant.
The precise definition of the content of a concept comes within the sphere of logic but it can be easily observed that the word-meaning is not identical with it. For instance, the content of the concept six can be expressed by ‘three plus three’, ‘five plus one’, or ‘ten minus four’, etc. Obviously, the meaning of the word six cannot be identified with the meaning of these word-groups.
To distinguish meaning from the referent, i.e. from the thing denoted by the linguistic sign is of the utmost importance, and at first sight does not seem to present difficulties. To begin with, meaning is linguistic whereas the denoted object or the referent is beyond the scope of language. We can denote one and the same object by more than one word of a different meaning. For instance, in a speech situation an apple can be denoted
by the words apple, fruit, something, this, etc. as all of these words may have the same referent. Meaning cannot be equated with the actual properties of the referent, e.g. the meaning of the word water cannot be regarded as identical with its chemical formula H2O as water means essentially the same to all English speakers including those who have no idea of its chemical composition. Last but not least there are words that have distinct meaning but do not refer to any existing thing, e.g. angel or phoenix. Such words have meaning which is understood by the speaker-hearer, but the objects they denote do not exist.
Thus, meaning is not to be identified with any of the three points of the triangle.
§ 2. Meaning in the Referential Approach
It should be pointed out that among the adherents of the referential approach there are some who hold that the meaning of a linguistic sign is the concept underlying it, and consequently they substitute meaning for concept in the basic triangle. Others identify meaning with the referent. They argue that unless we have a scientifically accurate knowledge of the referent we cannot give a scientifically accurate definition of the meaning of a word. According to them the English word salt, e.g., means ’sodium chloride (NaCl)’. But how are we to define precisely the meanings of such words as love or hate, etc.? We must admit that the actual extent of human knowledge makes it impossible to define word-meanings accurately.1 It logically follows that any study of meanings in linguistics along these lines must be given up as impossible.
Here we have sought to show that meaning is closely connected but not identical with sound-form, concept or referent. Yet even those who accept this view disagree as to the nature of meaning. Some linguists regard meaning as the interrelation of the three points of the triangle within the framework of the given language, i.e. as the interrelation of the sound-form, concept and referent, but not as an objectively existing part of the linguistic sign. Others and among them some outstanding Soviet linguists, proceed from the basic assumption of the objectivity of language and meaning and understand the linguistic sign as a two-facet unit. They view meaning as “a certain reflection in our mind of objects, phenomena or relations that makes part of the linguistic sign — its so-called inner facet, whereas the sound-form functions as its outer facet.” 2 The outer facet of the linguistic sign is indispensable to meaning and intercommunication. Meaning is to be found in all linguistic units and together with their sound-form constitutes the linguistic signs studied by linguistic science.
The criticism of the referential theories of meaning may be briefly summarised as follows:
1. Meaning, as understood in the referential approach, comprises the interrelation of linguistic signs with categories and phenomena outside the scope of language. As neither referents (i.e. actual things, phenomena,
1 See, e. g., L. Bloomfield. Language. N. Y., 1933, p. 139.
2 А. И. Смирницкий. Значение слова. — Вопр. языкознания, 1955, № 2. See also С. И. Ожегов. Лексикология, лексикография, культура речи. М., 1974, с. 197.
etc.) nor concepts belong to language, the analysis of meaning is confined either to the study of the interrelation of the linguistic sign and referent or that of the linguistic sign and concept, all of which, properly speaking, is not the object of linguistic study.
2. The great stumbling block in referential theories of meaning has always been that they operate with subjective and intangible mental processes. The results of semantic investigation therefore depend to a certain extent on “the feel of the language” and cannot be verified by another investigator analysing the same linguistic data. It follows that semasiology has to rely too much on linguistic intuition and unlike other fields of linguistic inquiry (e.g. phonetics, history of language) does not possess objective methods of investigation. Consequently it is argued, linguists should either give up the study of meaning and the attempts to define meaning altogether, or confine their efforts to the investigation of the function of linguistic signs in speech.
§ 3. Functional Approach to Meaning
In recent years a new and entirely different approach to meaning known as the functional approach has begun to take shape in linguistics and especially in structural linguistics. The functional approach maintains that the meaning of a linguistic unit may be studied only through its relation to other linguistic-units and not through its relation to either concept or referent. In a very simplified form this view may be illustrated by the following: we know, for instance, that the meaning of the two words move and movement is different because they function in speech differently. Comparing the contexts in which we find these words we cannot fail to observe that they occupy different positions in relation to other words. (To) move, e.g., can be followed by a noun (move the chair), preceded by a pronoun (we move), etc. The position occupied by the word movement is different: it may be followed by a preposition (movement of smth), preceded by an adjective (slow movement), and so on. As the distribution l of the two words is different, we are entitled to the conclusion that not only do they belong to different classes of words, but that their meanings are different too.
The same is true of the different meanings of one and the same word. Analysing the function of a word in linguistic contexts and comparing these contexts, we conclude that; meanings are different (or the same) and this fact can be proved by an objective investigation of linguistic data. For example we can observe the difference of the meanings of the word take if we examine its functions in different linguistic contexts, take the tram (the taxi, the cab,, etc.) as opposed to to take to somebody.
It follows that in the functional approach (1) semantic investigation is confined to the analysis of the difference or sameness of meaning; (2) meaning is understood essentially as the function of the use of linguistic units. As a matter of fact, this line of semantic investigation is the primary concern, implied or expressed, of all structural linguists.
1 By the term distribution we understand the position of a linguistic unit in relation to other linguistic units.
§ 4. Relation between the Two Approaches
When comparing the two approaches described above in terms of methods of linguistic analysis we see that the functional approach should not be considered an alternative, but rather a valuable complement to the referential theory. It is only natural that linguistic investigation must start by collecting an adequate number of samples of contexts.1 On examination the meaning or meanings of linguistic units will emerge from the contexts themselves. Once this phase had been completed it seems but logical to pass on to the referential phase and try to formulate the meaning thus identified. There is absolutely no need to set the two approaches against each other; each handles its own side of the problem and neither is complete without the other.
It is more or less universally recognised that word-meaning is not homogeneous but is made up of various components the combination and the interrelation of which determine to a great extent the inner facet of the word. These components are usually described as types of meaning. The two main types of meaning that are readily observed are the grammatical and the lexical meanings to be found in words and word-forms.
§ 5. Grammatical Meaning
We notice, e.g., that word-forms, such as girls, winters, joys, tables, etc. though denoting widely different objects of reality have something in common. This common element is the grammatical meaning of plurality which can be found in all of them.
Thus grammatical meaning may be defined ,as the component of meaning recurrent in identical sets of individual forms of different words, as, e.g., the tense meaning in the word-forms of verbs (asked, thought, walked, etc.) or the case meaning in the word-forms of various nouns (girl’s, boy’s, night’s, etc.).
In a broad sense it may be argued that linguists who make a distinction between lexical and grammatical meaning are, in fact, making a distinction between the functional (linguistic) meaning which operates at various levels as the interrelation of various linguistic units and referential (conceptual) meaning as the interrelation of linguistic units and referents (or concepts).
In modern linguistic science it is commonly held that some elements of grammatical meaning can be identified by the position of the linguistic unit in relation to other linguistic units, i.e. by its distribution. Word-forms speaks, reads, writes have one and the same grammatical meaning as they can all be found in identical distribution, e.g. only after the pronouns he, she, it and before adverbs like well, badly, to-day, etc.
1 It is of interest to note that the functional approach is sometimes described as contextual, as it is based on the analysis of various contexts. See, e. g., ^ Semantics. Oxford, 1962, pp. 64-67.
It follows that a certain component of the meaning of a word is described when you identify it as a part of speech, since different parts of speech are distributionally different (cf. my work and I work).1
§ 6. Lexical Meaning
Comparing word-forms of one and the same word we observe that besides grammatical meaning, there is another component of meaning to be found in them. Unlike the grammatical meaning this component is identical in all the forms of the word. Thus, e.g. the word-forms go, goes, went, going, gone possess different grammatical meanings of tense, person and so on, but in each of these forms we find one and the same semantic component denoting the process of movement. This is the lexical meaning of the word which may be described as the component of meaning proper to the word as a linguistic unit, i.e. recurrent in all the forms of this word.
The difference between the lexical and the grammatical components of meaning is not to be sought in the difference of the concepts underlying the two types of meaning, but rather in the way they are conveyed. The concept of plurality, e.g., may be expressed by the lexical meaning of the world plurality; it may also be expressed in the forms of various words irrespective of their lexical meaning, e.g. boys, girls, joys, etc. The concept of relation may be expressed by the lexical meaning of the word relation and also by any of the prepositions, e.g. in, on, behind, etc. (cf. the book is in/on, behind the table). “
It follows that by lexical meaning we designate the meaning proper to the given linguistic unit in all its forms and distributions, while by grammatical meaning we designate the meaning proper to sets of word-forms common to all words of a certain class. Both the lexical and the grammatical meaning make up the word-meaning as neither can exist without the other. That can be also observed in the semantic analysis of correlated words in different languages. E.g. the Russian word сведения is not semantically identical with the English equivalent information because unlike the Russian сведения the English word does not possess the grammatical meaning of plurality which is part of the semantic structure of the Russian word.
§ 7. Parf-of-Speech Meaning
It is usual to classify lexical items into major word-classes (nouns, verbs, adjectives and adverbs) and minor word-classes (articles, prepositions, conjunctions, etc.).
All members of a major word-class share a distinguishing semantic component which though very abstract may be viewed as the lexical component of part-of-speech meaning. For example, the meaning of ‘thingness’ or substantiality may be found in all the nouns e.g. table, love, sugar, though they possess different grammatical meanings of number, case, etc. It should be noted, however, that the grammatical aspect of the part-of-speech meanings is conveyed as a rule by a set of forms. If we describe the word as a noun we mean to say that it is bound to possess
1 For a more detailed discussion of the interrelation of the lexical and grammatical meaning in words see § 7 and also А. И. Смирницкий. Лексикология английского языка. М., 1956, с. 21 — 26.
a set of forms expressing the grammatical meaning of number (cf. ^ case (cf. boy, boy’s) and so on. A verb is understood to possess sets of forms expressing, e.g., tense meaning (worked — works), mood meaning (work! — (I) work), etc.
The part-of-speech meaning of the words that possess only one form, e.g. prepositions, some adverbs, etc., is observed only in their distribution (cf. to come in (here, there) and in (on, under) the table).
One of the levels at which grammatical meaning operates is that of minor word classes like articles, pronouns, etc.
Members of these word classes are generally listed in dictionaries just as other vocabulary items, that belong to major word-classes of lexical items proper (e.g. nouns, verbs, etc.).
One criterion for distinguishing these grammatical items from lexical items is in terms of closed and open sets. Grammatical items form closed sets of units usually of small membership (e.g. the set of modern English pronouns, articles, etc.). New items are practically never added.
Lexical items proper belong to open sets which have indeterminately large membership; new lexical items which are constantly coined to fulfil the needs of the speech community are added to these open sets.
The interrelation of the lexical and the grammatical meaning and the role played by each varies in different word-classes and even in different groups of words within one and the same class. In some parts of speech the prevailing component is the grammatical type of meaning. The lexical meaning of prepositions for example is, as a rule, relatively vague (independent of smb, one of the students, the roof of the house). The lexical meaning of some prepositions, however, may be comparatively distinct (cf. in/on, under the table). In verbs the lexical meaning usually comes to the fore although in some of them, the verb to be, e.g., the grammatical meaning of a linking element prevails (cf. he works as a teacher and he is a teacher).
§ 8. Denotational and Connotational Meaning
Proceeding with the semantic analysis we observe that lexical meaning is not homogenous either and may be analysed as including denotational and connotational components.
As was mentioned above one of the functions of words is to denote things, concepts and so on. Users of a language cannot have any knowledge or thought of the objects or phenomena of the real world around them unless this knowledge is ultimately embodied in words which have essentially the same meaning for all speakers of that language. This is the denotational meaning, i.e. that component of the lexical meaning which makes communication possible. There is no doubt that a physicist knows more about the atom than a singer does, or that an arctic explorer possesses a much deeper knowledge of what arctic ice is like than a man who has never been in the North. Nevertheless they use the words atom, Arctic, etc. and understand each other.
The second component of the lexical meaning is the connotational component, i.e. the emotive charge and the stylistic value of the word.