Selection, or deliberate choice of language, and the ways the chosen

5^t^jire treated are the main distinctive features of individual style.

The treatment of the selected elements brings up the problem of the norm. The notion of the norm mainly refers to the literary language and always presupposes a recognized or received s t a n d a r d. At the same time it likewise presupposes vacillations of the received standard.

In order to get a workable definition of the norm for the purposes set in this book and, particularly, in connection with the issue of indi­vidual style, it will be necessary to go a little bit deeper into the concept.

We shall begin with the following statement made by Academician L. V. Scherba:

"Very often when speaking of norms people forget about stylistic norms (emphasis added) which are no less, if not more, important than all others."1

This pronouncement clearly indicates that there is no universally accepted norm of the standard literary language, that there are different norms and that there exist special kinds of norm which are called sty­listic norms. Indeed, it has long been acknowledged that the norms of the spoken and the written varieties of language differ in more than one respect (see p. 35). Likewise it is perfectly apparent that the norms of emotive prose and those of official language are heterogeneous. Even within what is called the belles-lettres style of language (see p. 33—34) we can observe different norms between, for instance, poetry and drama.

In this connection I. Vachek of the Prague School of Linguistics states that "it is necessary to reject the possibility of the existence of an abstract, universal norm which subordinates written and oral norms in any of the natural languages."2

The same view is expressed by M. A. K. Halliday who states:

"There is no single universally relevant norm, no one set of expec­tancies to which all instances may be referred."3

This point of view is not, however, to be taken literally. The fact that there are different norms for various types and styles of language does not exclude the possibility and even the necessity of arriving at some abstract notion of norm as an invariant, which should em­brace all variants with their most typical properties. Each style of language will have its own invariant and variants (see p. 33—34), yet all styles will have their own invariant, that of the written variety of language. Both oral ^colloquial) and written (literary) varieties can also be integrated into an invariant of the standard (received) language.

The norm is regarded by some-linguists as "a regulator which controls a set of variants, the borders of variations and also admissible and inadmissible variants." (E. A. Makayev)

Here are some other definitions.

"The norm is an assemblage (a set) of stable (i.e. regularly used) means objectively existing in the language and systematically used."

"A certain conventionally singled out assemblage of realizations of language means recognized by the language community as a model.'" (Gukhman & Semenyuk)

"The norm is a linguistic abstraction, an idea thought up by linguists and existing only in their minds." (A. E. Darbyshire)

"There is, of course, no such thing as the norm to be found in actual usage. It is a concept which must be expressed by means of a formula, and it is a concept about that which is left of uses of language when all stylistic qualities have been taken away from them." (A. E. Darbyshire)

The last of the definitions elaborates the idea of the norm as something stripped of its stylistic qualities. This is not accidental. Many linguists hold the view that anything which can be labeled stylistic is already a deviation from the established norm (see a number of the definitions of 'style' given on page 11). They forget that regular deviations from the norm gradually establish themselves as variants of the norm; the more so because, as has been stated, 'deviations' of a genuinely stylistic character are not deviations1 but typified and foregrounded natural phenomena of language usage, though sometimes carried to the extreme.

So, finally, we can arrive at the conclusion that the norm presup­poses the oneness of the multifarious. There is a conscious attitude to what is well-formed against what is ill-formed. Well-formness may be represented in a great number of concrete sentences allowing a considerable range of acceptability.

The norm, therefore, should be regarded as the invariant of the pho­nemic, morphological, lexical and syntactical patterns circulating in language-in-action at a given period of time. Variants of these patterns may sometimes diverge from the invariant but they never exceed the limits set by the invariant lest it should become unrecognizable or mis­leading. The development of any literary language shows that the va­riants will always center around the axis of the invariant forms. The variants, as the term itself suggests, will* never detach themselves from the invariant to such a degree as to claim entire independence. Yet, nevertheless, there is a tendency to estimate the value of individual style by the degree it violates the norms of the language.

As we have already cited, G. Saintsbury considers that the real secret of style reveals itself in the breach or neglect of the rules that govern the structure of clauses, sentences, and paragraphs (see p. 15). This conception is aptly illustrated theoretically in the Theory of Deviance mentioned above (p. 15) and practically- in the works of certain modern poets like E. E. Cummings and others who try to break away entirely from the established and recognized invariants and variants of the given norm. They introduce various patterns

which are almost undecodable and consequently require special devices for grasping the messages.l -

Quite a different point of view is expressed by E. Sapir, who states:

"...the greatest — or shall we say the most satisfying — literary artists, the Shakespeare and Heinz, are those who have known subconsciously how to fit or trim the deeper intuition to the provincial accents of their daily speech. In them there is no effect of strain. Their personal intuition appears as a completed synthesis of the absolute art of intuition and the innate, specialized art-of the linguistic medium."2

This idea is common to many stylists who hold that real and genuine individuality of style will reveal itself not in the breach of the rules, in other words, not in deviating from the accepted norms, but in the pecu­liar treatment of them. However, it must be repeated that some devia­tions, if they are motivated, may occur here and there in the text. More­over, let us repeat once more that through constant repetitions such de­viations may become legitimate variants of the norm and establish them­selves as members of the language system.

The problem of variants of the norm, or deviations from the norm of the literary language, has received widespread attention among lin­guists and is central to some of the major current controversies. It is the inadequacy of the concept 'norm' that causes the controversy. At every period in the development of a literary language there must be a tangible norm which first of all marks the difference between literary and non-literary language. Then there must be a clear-cut distinction between the invariant of the norm (as an abstraction) and its variants (in concrete texts). As will be seen later almost every functional style of language is marked by a specific use of language means, thus estab­lishing its own norms which, however, are subordinated to the norm-invariant and which do not violate the general notion of the literary norm.

One of the most characteristic and essential properties of the norm is its flexibility. A too rigorous adherence to the norm brands the writ­er's language as pedantic, no matter whether it is a question of speech or writing. But on the other hand, neglect of the norm will always be regarded with suspicion as being an attempt to violate the established signals of the language code which safeguard and accelerate the process of communication. At the same time, a free handling of the norms may be regarded as a permissible application of the flexibility of the norm.

It must be acknowledged that to draw a line of demarcation between facts that illustrate the flexibility of the norm and those which show its violation is not so easy. The extremes are apparent, but border cases are blurred. Thus "footsteps on the sand of war" (E. E. Cummings) or "below a time" (see other examples on p. 162—163) are clearly violat­ions of the accepted norms of word-building or word-combinations.

But "silent thunder", "the ors and ifs" and the like may from one point of view be regarded as a practical application of the principle of flexibility of the norm and from another—as a violation of the semantic and morphological norms of the English language. Variants interacting with the rigorous rules of usage may reveal the potentialities of the lan­guage for enrichment to a degree which no artificial coinage will ever be able to reach. This can be explained by the fact that semantic changes and particularly syntactical ones are rather slow in process and they reject any sudden imposition of innovations on the code already in action. There is, a constant process of gradual change taking place in the forms of language and their meaning at any given period in the development of the language. It is therefore most important to master the received standard of the given period in the language in order to comprehend the correspondence of this or that form to the recognized norm of the period.

Some people think that one has to possess what is called a "feeling for the language" in order to be able to understand the norm of the lan­guage and its possible variants. But this feeling is deeply rooted in the unconscious knowledge of the laws according to which a language func­tions, and even in its history, which explains much concerning the di­rection it has progressed. When the feeling of the norm, which grows with the knowledge of the laws of the language, is instilled in the mind, one begins to appreciate the beauty of justifiable fluctuations.

Paradoxical though it may seem, the norm can be grasped, nay, es­tablished, only when there are deviations from it. It is therefore best perceived in combination with something that breaks it. In this connec­tion the following remarks made by L". V. Scherba are worth quoting:

"... in order to achieve a free command of a literary language, even one's own, one must read widely, giving preference to those writers who deviate but slightly from the norm."

"Needless to say, all deviations are to some extent normalized: not every existing deviation from the norm is good; at any rate, not in all circumstances. The feeling for what is permissible and what is not, and mainly—a feeling for the inner sense of these deviations (and senseless ones, as has been pointed out, are naturally bad), is developed through an extensive study of our great Russian literature in all its variety, but of course in its best examples."1 •" "

"I say justifiable or "motivated" because bad writers frequently make use of deviations from the norm which are not motivated or justified by the subject matter—that is why they are considered bad writers."2

While dealing with various C9nceptions of the term 'style', we must also mention a commonly accepted connotation of style as establishment of language. This understanding of style is upheld in some of the scientific papers on literary criticism. Language and style as embellishment are regarded as separate bodies. According to this idea language can easily dispense with style, because style here is likened to the trimming on a dress. Moreover, style as embellishment of language ^viewed as something that hinders understanding. It is, as it were,

Спорные вопросы русской грамматики.— “Русский язык в школе”. 1*' 1у39, № 1, с. 10. Ibid.

alien to language and therefore needs to be excluded from the observa­tions of language scholars. That is why almost all contemporary books on grammar and general linguistics avoid problems of style or, at most, touch upon them in passing. The notion of style as embellishment pre­supposes the use of bare language forms deprived of any stylistic de­vices, of any expressive means deliberately employed. In this connect ion Middleton Murry writes:

"The notion that style is applied ornament had its origin, no doubt, in the tradition of the school of rhetoric in Europe, and in its place in their teaching. The conception was not so monstrous as it is today. For the old professors of rhetoric were exclusively engaged in instructing their pupils how to expound an argument or arrange a pleading. Their classification "of rhetorical devices was undoubtedly formal and extrav­agant... The conception of style as applied ornament ... is the most popular of all delusions about style."1

The notion of style as embellishment of language is completely er­roneous. No matter how style is treated, it is the product of a writer's deliberate intention to frame his ideas in such a manner as will add some­thing important, something indispensable in order to secure an ade­quate realization of his ideas. To call style embellishment is the same thing as to strip it of its very essence, that is, to render unnecessary those elements which secure the manifold application of the language units.

No doubt there are utterances which contain all kinds of unmotivated stylistic means. Moreover, there are writers whose style abounds in such utterances. But they are either those who, admiring the form, use it at the expense of the matter, or those who, by experimenting with the po­tentialities of language means, try to find new ways of rendering their ideas. In both cases the reader is faced with difficulties in decoding the message and this greatly hinders understanding.

A very popular notion of style among teachers of language is that style is t e с h n i q u e of expression. In this sense style is generally difined as the ability to write clearly, correctly and in a man­ner calculated to interest the reader. Though the last requirement is not among the indispensable, it is still found in many practical manuals of style, most of which can be lumped together under the title "Compo­sition and Style". This is a purely utilitarian point of view of the issue in question. If this were true, style could be taught. Style in this sense of expression studies4he normalized forms of the language. The teaching process aims at lucidity of expression. It sets up a number of rules as to how to speak and write well and generally discards all kinds of devia­tions as being violations of the norm. The norm in these works is treated as something self-sustained and, to a very great extent, inflexible.

The utilitarian approach to the problem is also felt in the following statement by E. J. Dunsany, an Irish dramatist and writer of short stories:

"When you can with difficulty write anything clearly, simply, and

emphatically, then, provided that the difficulty is not apparent to the reader, that is style. When you can do it easily, that is genius."

V. 'G. Belinsky also distinguished two aspects of style, making a hard and fast distinction between the technical and the creative power of any utterance.

'To. language merits belong correctness, clearness and fluency," he states, "qualities which can be achieved by any talentless writer by means of labour and routine."

"But style (слог) — is talent itself, the very thought."1

Almost the same point of view is held both by A. N. Gvozdev and F. L. Lucas. Gvozdev states that "Stylistics has a practical value, teach­ing students to master the language, working out a conscious approach to language"2 and Lucas declares that the aims of a course in style are: a) to teach to write and speak well, b) to improve the style of the writer, and c) to show him means of improving his ability to express his ideas.3

It is important to note that what we call the practical approach to the problem of style should by no means be regarded as something er­roneous. The practical side of the problem can hardly be over-estimated. But should it be called style? The ability to write clearly and emphatic­ally can and should be taught. This is the domain of grammar, which today rules out the laws and means of composition. The notion of style cannot be reduced to the merely practical aspect because in such a case a theoretical background for practical „aims cannot be worked out. Moreover, stylistics as a branch of linguistics demands investigation into the nature of such language means as add aesthetic value to the utterance.

Just as the interrelation between lexicology and lexicography is accepted to be that of theory and practice, so theoretical and practical stylistics should be regarded as two interdependent branches of linguistic science. Each of these branches may develop its own approach and methods of investigation of linguistic data.

The term 'style' is widely used in literature to signify literary genre. Thus, we speak of classical style or the style of classicism, realistic style, the style of romanticism and so on. The use of the word 'style' has sometimes been carried to unreasonable lengths, thus blurring the terminological aspect of the word. It is applied to various kinds of literary works: the fable, novel, ballad, story, etc. The term is also used to denote the way the plot is dealt with, the arrangement of the parts of literary composition to form the whole,4he place and the role of the author in describing and depicting events.

It is suggested in this work that the term * style' be used to refer to purely linguistic facts, thus avoiding the possible ambiguity in its application. After all the origin of the word 'style' is a justification for the suggestion. However, we are fully aware of the fact that such a pro-

position will be regarded as an encroachment on the rights of literature to have its own terms in spite of the fact that they are the same as terms in linguistics.

Now let us pass to the discussion of an issue the importance of which has to be kept clearly in mind throughout the study of stylistics, that is the dichotomy of language and s p e e с h or, to phrase the issue differently, language- as -a-s у stem and language-in-action. It deserves at least a cursory discussion here not only because the issue has received a good deal of attention in recent publi­cations on linguistic matters, but also because, as will be seen later, many stylistic devices stand out against the background of the distinc­tive features of these two above-mentioned notions. The simplicity of the issue is to some extent deceptive. On the surface it seems that language-in-action takes the signs of language-as-a-system and arranges them to convey the intended message. But the fact is that the signs of the latter undergo such transformations in the former that sometimes they assume a new quality imposing new signification on the signs of the language code. There is compelling evidence in favour of the theory which demands that the two notions should be regarded in their unity, allowing, however, that each of them be subjected to isolated obser­vation.

Language-as-a-system may figuratively be depicted as an exploiter of language-in-action. All rules and patterns of language which are col­lected and classified in works on grammar, phonetics, lexicology and stylistics first appear in language-in-action, whence they are genera­lized and framed as rules and patterns of language-as-a-system.

It is important here to call attention to the process of formation of scientific notions. Whenever we notice a phenomenon that can be singled out from a mass of language facts we give it a name, thus abstracting the properties of the phenomenon. The phenomena then being collected and classified are hallowed into the ranks of the units of language-as-a-system. It must be pointed out that most observations of the nature and functioning of language units have been made on material presented by the written variety of language. It is due to the fixation of speech in writing that scholars of language began to disintegrate the continuous flow of speech and subject the functioning of its components to analysis.

So it is with stylistic devices. Being born in speech they have grad­ually become recognized as certain patterned structures: phonetic, morphological, lexical, phraseological and syntactical, and duly taken away from their mother, Speech, and made independent members of the family, Language.

The same concerns the issue of functional styles of language. Once they have been recognized as independent, more or less closed subsys­tems of the standard literary language, they should be regarded not as styles of speech but as styles of language, inasmuch as they can be pat­terned as to the kinds of interrelation between the component parts in each of the styles. Moreover, these functional styles have been subjected to various classifications, which fact shows that the phenomena now belong to the domain of language-as-a-system.

However, it must constantly be born in mind that the units which belong to this domain are abstract in their nature. Functional styles re merely models deprived of material substance, schemes which can be materialized in language forms. When materialized in language forms they 'become practical realizations of abstract schemes and signify the variants of the corresponding invariants of the models.

This relatively new science, stylistics, will be profitable to those who have a sound linguistic background. The expressive means of Eng­lish and the stylistic devices used in the literary language can only be understood (and made use of) when a thorough knowledge of the language-as-a-system, i.e. of the phonetic, grammatical and lexical data of the given language, has been attained.

It goes without saying that the more observant the student is, the easier it will be for him to appreciate the peculiar usage of the language media.

Justification for bringing this problem up is that some language scholars frighten students out of studying stylistics on the ground that this subject may effectively be studied only on the basis of a perfect command of the language. Such scholars, aware of the variables and un­knowns, usually try in their teaching to sidestep anything that may threaten well-established theories concerning the laws of language. Alert­ness to 'the facts of language-in-action should be inherent, but it can be developed to a degree necessary for an aesthetic evaluation of the works of men-of-letters. And for this purpose it is first of all necessary to get a clear idea of what constitutes the notions ' expressive means' and 'sty­listic devices'.

EXPRESSIVE MEANS (EM) AND STYLISTIC DEVICES (SD)

In linguistics there are different terms to den _by which utterances are foreground, i.e. made more conspicuous, more "effective and therefore imparting some additional information. They are called expressive means, stylistic means, stylistic markers, stylistic devices, tropes, "figures of speech and other names. All these terms are used indiscriminately and are set against those means which we shall conventionally call neutral. Most linguists distinguish ordinary (also: substantial, referential) semantic and stylistic diffe­rences in meaning. 58), others besides these contain specif. lc. meanings which may be called sty I i s t i c. Such meanings go alongside primary meanings and, as it were, are superim­posed on them.

Stylistic meanings are so to say deautomatized. As is known, the process of automatization, i.e. a speedy and subconscious use of lan­guage data, is one of the indispensable ways of making communication easy and quickly decodable.

But when a stylistic meaning is involved, the process of deautomatization checks the reader's perception of the language. His attentionis arrested by a peculiar use of language media and he begins, to the best of his ability, to decipher it. He becomes aware of the form in which the utterance is cast and as the result of this process a twofold use of the language medium—ordinary* and stylistic—becomes apparent to him. As will be shown later this application of language means in some cases presents no difficulty. It is so marked that even a layman can see it, as when a metaphor or a simile is used. But in some texts grammatically redundant forms or hardly noticeable forms, essential for the expression of stylistic meanings which carry the particular addi­tional information desired, may present a difficulty.

What this information is and how it is conveyed to the mind of. the reader can be explored only when a concrete communication is subjected to observation, which will be done later in the analyses of various stylistic devices arid in the functioning of expressive means.

In this connection the following passage from "Investigating English Style" by D. Crystal and D. Davy is of interest: "Features which are stylistically significant display different kinds and degrees of distinctiveness in a text: of two features, one may occur only twice in a text, the other may occur thirty times,— or a feature might be uniquely iden­tifying in the language, only ever occurring in one variety, as opposed to a feature which is distributed throughout many or all varieties in dif­ferent frequencies."1

What then is a stylistic device? Why is it so important to distinguish it from the expressive and neutral means of the language? To answer . these questions it is first of all necessary to elucidate the notion 'ex­pressiveness'.

The category of expressiveness has long been the subject of heated discussions among linguists. In etimological sense expressive­ness may be understood as a kind of intensification of an utterance or of a part of it depending on the position in the utterance of the means that manifest this category and what these means are.

But somehow lately the notion of expressiveness has been confused with another notion, viz. emotiveness. Emotiveness, and corresponidingly the emotive elements of language, are what reveal the emotions of writer or speaker. But these elements are not direct manifestations ^f"the*^molT6ns—they are just the echoes of real emotions, echoes which have undergone some intellectual recasting. They are designed to awaken co-experience in the mind of the reader.

Expressiveness a broader notion than emotiveness and is by no means to be reduced to the latter. Emotiveness is an integral part of expressiveness and, as a matter of fact, occupies a predominant position in the category of expressiveness. But there are media in language which aim simply at logical emphasis of certain parts of the utterance. They do not evoke any intellectual representation of feeling but merely serve the purpose of verbal actualization of the utterance. Thus, for example, when we say "It was in July 1975 that the cosmos experiment of a joint American-Soviet flight took place" we make the utterance logically em-

hatic by a syntactical device which will be described in due course. The same thing is to be observed in these sentences:

(1) Mr. Smith was an extremely unpleasant person.

(2) Never will he go to that place again.

(3) In rushed the soldiers!

(4) It took us a very, very long time to get there.

In sentence (1) expressiveness is achieved by lexical means—the word 'extremely'. In (2) and (3) by syntactical means—different types of inversion. In (4) the emphasis is materialized by the repetition of the word 'very7 which is in itself a word used to intensify the utterance.

But in the sentences:

(1) Isn't she cute!

(2) Fool that he was!

(3)" This goddam window won't open!

(4) We buddy-buddied together.

(5) This quickie tour didn't satisfy our curiosity, we can register positive emotiveness, in as much as there are elements that evoke certain representations of the feeling of the speaker. In sen­tence (1) and (2) there are syntactical means which evoke this effect. In (3) and (4) there are lexical means—'goddam', 'buddy-buddied' (=were on very friendly relations); in (5)—a morphological device (the suffix—te).

It must be noted that to draw a hard and fast distinction between logical and emotional emphasis is not always possible. The fact is that the logical and the emotional frequently overlap. A too strong logical emphasis may colour the utterance with emotional elements, thus causing a kind of expressiveness which is both logical and emotive. However, the extremes are clearly set one against the other, ...,,^ /^

Now it should be possible to define the notiono^xpressivemeans^TheC^/^и expressive means of a lafigtraf are those pKbnetfc^'mofplioTogical, word- ' Building, lexical, phraseological and syntactical forms which exist In language-as-a-system for the purpose of logical and/or emotional in­tensification of the utterance. These intensifying forms, wrought by social usage and recognized by their semantic function, have been singled * out in grammars, courses in phonetics and dictionaries (including phra­seological ones) as having special functions in making the utterances emphatic. Some of them are normalized, and good dictionaries label them as "intensifies". In most cases they have corresponding neutral synonymous forms. Compare, for example, the following pairs:

(1). He shall do it! = I shall make him do it.

(2) Isn't she cute! = She is very nice, isn't she?

Expressiveness may also be achieved by compositional devices in utterances comprising a number of sentences—in syntactical wholes and in paragraphs. This will be shown in the chapter on syntactical sty­listic devices.

The most powerful expressive means of, anxJMguag6 are phonetic. The human voice can indicate subtle nuances of meaning that no other means can attain. Pitch, melody, stress, pausation, drawling out certain syllables, whispering, a sing-song manner and other ways of using the voice are much more effective than any other means in intensifying an utterance emotionally or logically. In the language course of phonetics the patterns of emphatic intonation have been worked out, but many devices have so far been little investigated.

Paradoxical though-it may seem, many of these means, the effect of which rests on a peculiar use of the voice, are banned from the linguistic domain. But there has appeared a new science—"paralinguistic"—of which all these devices are the inventory. The writer of this book holds the opinion that all the vocal peculiarities enumerated .should be recog­nized as legitimate members of the phonetic structure of language and that therefore the term * paralinguistics' should be done away with.

Professor Seymour Chatman introduces the term 'phonostylistics' and defines it as a subject the purpose of which is "the study of the ways in which an author elects to constrain the phonology of the language beyond the normal requirements of the phonetic system."1 As can be inferred from this quotation, phonetic expressive means and particu­larly phonetic stylistic devices (seep. 123) are not deviations from "the normal requirements of the phonetic system" but a way of actualizing the typical in the given text. Vocal phenomena such as drawling, whisper­ing, etc. should be regarded as parts of the phonemic system on the same level as pitch, stress and tune.

In this part of the book where general ideas are presented in an in­troductory aspect only, there is no need to go deeper into the issue of what constitutes the notion expressive means of the phonetic system. The reader is referred to part III "Phonetic Expressive Means and Sty­listic Devices" (p. 123).

Passing over to some preliminary remarks on the morpholog­ical expressive means of the English language, we must - point to what is now a rather impoverished set of media to which the quality of expressiveness can be attributed. However, there are some which alongside their ordinary grammatical function display a kind of emphasis and thereby are promoted to EMs. These are, for example, The Historical Present; the use of shall in the second and third person; the use of some demonstrative pronouns with an emphatic meaning as those, them ("Those gold candles fixed in heaven's air"—Shakespeare); 'some cases of nominalization, particularly when conversion of verbal stems is alien to the meaning of the verbs or the nominalization of phrases and sentences and a ituniber of other morphological forms, which acquire expressiveness in the context, though this capacity is not yet registered as one of the latent properties of such forms.

Among the w о r d - b и i I d in g me a n s we find a great many forms which serve to make the utterance more expressive by intensifying some of their semantic and/or grammatical properties. The diminutive suffixes,^.(-fe), -let, e.g. 'dearie', 'sonny', 'auntie', “streamfef, add some emotional colouring to the words. We may also refer to what are called neologisms and nonce-words formed with non-productive suffixes

with Greek roots, as “mistressmansWp', 'cleanorama' (see p. 92). Certain affixes have gained such a power of expressiveness that they begin functioning as separate words, absorbing all of the generalizing meaning they attach to different roots, as, for example, 'isms and olo-

At the lexical I e v Јj.„there are a great many words which due to theiiHiTlTrf^1ф^ constitute a special layer (see chart on p 71). There are words with emotive meaning only (mteijections), words-..jzdlich,.hays... both..,'referential and emotive meaning (epithets), words which still retain a twofold meaning: denotative and connotative (love, hate, sympathy), words_hЈloiigingJ^^

words, or to poetic or archaic layers. The expressive power of these words cannot be doubted, especially when they are compared with the neutral vocabulary.

All kinds of set phrases (phraseological units) generally possess the property of expressiveness. Set phrases, catch words, proverbs, sayings comprise a considerable number of language units which serve to make speech emphatic, mainly from the emotional point of view. Their use in every-day speech is remarkable for the subjective emotional colouring they produce.

It must be noted here that due to the generally emotional character of colloquial language, all kinds of set expressions are natural in every­day speech. They are, as it were, part and parcel of this form of human intercourse. But when they appear in written texts their expressiveness comes to the fore because written texts, as has already been pointed out, are logically directed unless, of course, there is a deliberate attempt to introduce an expressive element in the utterance. The set expression is a time-honoured device to enliven speech, but this device, it must be repeated, is more sparingly used in written'texts. In everyday speech one can often hear such phrases as: "Well, it will only add fuel to the fire" and the like, which in fact is synonymous to the neutral: "It will only make the situation worse."

Finally, at the syntactical level there are many construc­tions which, when set against synonymous neutral ones, will reveal a certain degree of logical or emotional emphasis.

In order to be able to distinguish between expressive means and stylistic devices, to which we now pass, it is necessary to bear in mind that expressive means are concrete facts of language. They are studied in the respective language manuals, though it must be once again re­gretfully stated that some grammarians iron out all elements carrying expressiveness from their works, as they consider this quality irrelevant to the theory of language.

Stylistics studies the expressive means of language, but from a spe­cial angle. It takes into account the modifications of meanings which various expressive means undergo when they are used in different func­tional styles. Expressive means have a kind of radiating effect. They noticeably colour the whole of the utterance no matter whether they are logical or emotional.

What then is a stylistic device? It is a conscious and intentional intensification of some typical structural and/or semantic property of a language unit (neutral or expressive) prompted to a generalized status and thus Becoming a generative model, It follows then that an SD is an abstract pattern, a mould into which any content can be poured. As is known, the typical is not only that which is in frequent use, but that also which reveals the essence of a phenomenon with the greatest and most evident force. • -

SDs function in texts as marked units. They always..carry some kind of additionjff^^^ That is why the meffiod~ of free variation employed in descriptive linguistics1 cannot be used in stylistics because any substitution may cause damage to the semantic and aesthetic aspect of the utterance.

A. W. De Groot points out the significance of SDs in the following passage:

"Each of the aesthetically relevant features of the text serves to create a feature of the gestalt2 of the poem. In this sense the relevant linguistic features may be said to function or operate as gestalt factors."3

The idea of the function of SDs is expressed most fully by V. M. tir-munsky in the following passage:

"The justification and the sense of each device lies in the wholeness of the artistic impression which the work of art as a self-contained thing produces on us. Each separate aesthetic fact, each poetical device (em­phasis added) finds its place in the system, the sounds and sense of the words, the syntactical structures, the scheme of the plot, the composi­tional purport — all in equal degree express this wholeness and find justification."4

The motivated use of SDs in a genuine work of emotive literature is hot easily discernible, though they are used in some kind of relation to the facts, events, or ideas dealt with in the artistic message. Most SDs display an application of two meanings: the ordinary one, in.other words, the meaning (lexical or structural) which has already been estab­lished in the language-as-a-system, and a special meaning which is superimposed on the unit by the text, i.e. a meaning which appears in the language-in-action.

Sometimes, however, the twofold application of a lexical unit is accomplished not by the interplay of two meanings but by two words (generally synonyms) one of which is perceived against the background of the -other. This will be shown in subsequent chapters.

The conscious transformation of a language fact into a stylistic de­vice has been observed by certain linguists whose interests in linguistic theory have gone beyond the boundaries of grammar. Thus A. A. Poteb-nya writes:

1 By 'free variation' is meant the substitution of one form by another without any change of meaning.

2 'Gestalt' is a term in psychology which denotes a phenomenon as a whole, a kind of oneness, as something indivisible into component parts. The term has been borrowed by linguistics to denote the inseparability of the whole of a poetic work.

"As far back as in ancient Greece and Rome and with few exceptions n to the present time, the definition of a figurative use of a word has been based on the contrast between ordinary speech, used in its own, natural, primary meaning, and transferred speech."1

The contrast which the author of the passage quoted points to, can not always be clearly observed. In some SDs it can be grasped immediately in others it requires a keen eye and sufficient training to detect it. It must be emphasized that the contrast reveals itself most clearly when our mind perceives twofold meanings simultaneously. The meanings run parallel: one of them taking precedence over the other.

Thus in "The night has swallowed him up" the word 'swallow' has W°a) referential and b) contextual (to make disappear, to make vanish). The meaning (b) takes precedence-over the referential (a).

The same can be observed in the sentence: "Is there not blood enough upon your penal code that more must be poured forth to ascend to Heaven and testify against you?" (Byron)

The interrogative form, i.e. the structural meaning of a question, runs parallel with the imposed affirmative thought, i.e. the structural meaning of a statement, and it is difficult to decide which of the two structural meanings—the established or the superimposed—takes the upper hand.

In the following chapters where detailed analysis of the different SDs will be carried out, we shall try, where possible, to consider which of the two meanings realized simultaneously outweighs the other.

The birth of SDs is a natural process in the development of language media. Language units which are used with more or less definite aims of communication in various passages of writing and in various func­tional styles begin gradually to develop new features, a wider range, of functions, thus causing polyfunctionality. Hence they can be presented as invariants with concrete variables.

The interrelation between expressive means and stylistic devices can be worded in terms of the theory of information. Expressive means have a greater degree of predictability than.stylistic devices. The latter may appear in an environment which may seem alien and therefore be only slightly or not at all predictable. Expressive means, on the con­trary, follow the natural course of thought, intensifying it by means commonly used in language. It follows that SDs carry a^g^a^amoyjt of information and therefore require a certain effort to decode their meaning and purport. SDs must be regarded as a special code which has to be well known to the reader in order to be deciphered easily.

The notion of language as a special code is now very much practiced in the analyses of the functions of language units. E. Stankievicz sees

a kind of code-switching when SDs are employed. He also acknowledges

j| the twofold application of the language code when "... the neutral,

К basic code serves as the background against which the elements of an-

p other system acquire expressive prominence within the context of the basic

system."1 SDs are used sparingly in emotive prose, lest they should over­burden the text with implications thus hindering the process of decoding. They are abundantly used in poetry and especially so in some trends of poetical tradition, consequently retarding mental absorption of the content.2

Not every stylistic use of a language fact will come under the term SD, although some usages call forth a stylistic meaning. There are practically unlimited possibilities of presenting any language fact in what is vaguely called its stylistic use. For a language fact to be promo­ted to the level of an SD there is one indispensable requirement, which has already been mentioned above, viz. that it should so be used to call forth a twofold perception of lexical or/and structural meanings. Even a nonce use can and very often does create the necessary conditions for the appearance of an SD. But these are only the prerequisites for the appearance of an SD. Only when a newly minted language unit which materializes the twofold application of meanings occurs repeatedly in different environments, can it spring into life as an SD and subse­quently be registered in the system of SDs of the given language.

Therefore it is necessary to distinguish between a stylistic use of a language unit, which acquires what we call a stylistic meaning, and a stylistic device, which is the realization of an already well-known ab­stract scheme designed to achieve a particular artistic effect. Thus many facts of English grammar are said to be used with stylistic meaning, for example, the morphological expressive means mentioned on p. 28. But most of them have not yet been raised to the level of SDs because they remain unsystematized and so far perceived as nonce uses. They are, as it were, still wandering in the vicinity of the realm of SDs without being admitted into it. This can indirectly be proved by the fact that fhey have no special name in the English language system of SDs. An exception, perhaps, is the Historical Present which meets the requirements of an SD.

So faf the system of stylistic devices has not been fully recognized as legitimate members of the general system of language. This is mainly due to the above-mentioned conception of grammatical theory as dealing exclusively with a perfectly organized and extremely rigid scheme of language rules, precise and accurate in its application.