The structure of stylistics

Stylistics and other linguistic disciplines. Let us return once more to the beginning of this textbook. At the very start it was proclaimed that such well-known disciplines of linguistics as phonetics, morphology, lexicology, and syntax deal with more or less clear-cut objects: a student would never mistake lexicology for phonetics or otherwise. This comes from the fact that the enumerated subjects are, if one may say so, level disciplines, i.e. disciplines treating one linguistic level each.

Generally speaking, the word level became very popular in twentieth century science (not necessarily linguistics: cf. molecular level) and even in political phraseology: Prime Minister level, on (at) the highest level, etc.

Being very widely employed, the word level has lost all limitations as to its applicability and is now used as a synonym to the words and expressions point of view (or viewpoint), aspect of research, sphere, plane,

domain and so forth. In linguistics, the word level is used (or perhaps misused) in collocations like language level (уровень языка), speech level, observation level (уровень наблюдения), construct level (уровень кон­структов), prosodic level (просодический), phraseological level, the level of the principal parts of the sentence, and even stylistic level ²³ (the latter was once proposed by Galperin).

It goes without saying that if we agreed that the word level is a synonym of viewpoint, aspect of research, etc., the above cited use of it would be quite legitimate, and surely one might then also speak of stylistic level.

But the term level as applied to language is more appropriate when used in the sense implied by the French linguist E. Benveniste, who used it to characterize the hierarchical structure of language itself, not the arbitrary aspects of research. Our compatriot Yu.S. Maslov employs the term tier ('ярус') instead.²⁴

The smallest (shortest) unit of language is the phoneme. The sequence of phonemes making units of higher ranks represents the phonemic level. One or (in most cases) several phonemes combined(in succession) constitute a unit of a higher level, the-second level: that of morphemes, or the morphemic level. One or (usually) more than one morpheme make a word, a 'lexeme': hence, the lexical level. One or (usually) more than one word make an utterance, or, in traditional terminology, a sentence. Hence, the sentence level. Word combinations are best treated as not forming an independent level for two reasons: 1) functionally, they do not differ from words, because they name without communicating; 2) one word does not make a word combination, whereas one word can make an utterance: Out!, Why?, Winter, Nevermore.

We could go on singling out paragraph level and even text level paying homage to the now fashionable text linguistics ²⁵ but for the fact that not every text is divided into paragraphs (especially if it is short), although every paragraph or every text is divisible into sentences (or, sometimes, coincides with one: a paragraph or a text consisting of one single sentence).

Be that as it may, the general principle is: each level consists of units of the neighbouring lower level with nothing besides: a sentence consists only of words; a word is divided into morphemes or sometimes coincides with one; a morpheme contains nothing but phonemes or is represented by one of them, as in makes ([s]), read-er ([э]), pens ([z]).

Summing up, we must say that the first meaning of the word level suggests the idea of horizontal layers of some structure. And indeed, when we come to i nspect language, we discover (as did our predecessors long ago) that language presents a hierarchy of levels, from the lowest up to the highest.

And, as we can easily conclude, each level is described by what we named above a 'level discipline': phonetics, morphology, lexicology, and syntax. To these, the modern text linguistics may be added.

Of course, stylistics does not fit in here. For, as the reader probably understands, stylistics is not a level discipline (just as history of language or comparative typology of English and Russian are not), because stylistics pertains to all the levels, to every level (the same is true, by the way, about history and typology).

Moreover, stylistics must be subdivided into separate, quite indepen­dent branches, treating one level each. Hence we have:

stylistic phonetics stylistic morphology stylistic lexicology stylistic syntax

We shall now look for the difference between general phonetics, morphology, lexicology, syntax, on the one hand, and their stylistic counterparts, on the other.

The reader will remember that the ultimate aim, as well as the general method, of stylistics is description of specific spheres of sublanguages. Therefore, whatever level we take, stylistics describes not what is in common use, but what is specific in this or that respect, what differentiates one sublanguage from others.

General (i.e. non-stylistic) phonetics, both prescriptive and theoreti­cal, investigates the whole articulatory-audial system of language. Stylis­tic phonetics pays attention only to style-forming phonetic features of sublanguages: it describes variants of pronunciation occurring in differ­ent types of speech (cf. recitation or oration with colloquial speech). Special attention is also paid to prosodic features of prose and poetry.

Non-stylistic (general) morphology treats morphemes and grammati­cal meanings expressed by them in language in general, without regard to their stylistic value. Stylistic morphology, on the contrary, is inter­ested in grammatical forms and grammatical meanings that are peculiar to particular sublanguages, explicitly or implicitly comparing them with the neutral ones common to all the sublanguages.

The relationship of what is taught in lecture courses as well as handbooks on lexicology and what is called here " stylistic lexicology" is somewhat more complicated. Actually, it is the chapters in lexicology books that deal with stylistic classification (stylistic differentiation) of the vocabulary that form a part of stylistics (stylistic lexicology), although there is more to stylistic lexicology than just that information. Chapters on word-building are not directly concerned with stylistic problems, unless they indicate where (in what sublanguages) this or that mode of word-formation is current. The etymological analysis of the vocabulary (the problem of borrowings in particular) is stylistically relevant only when the analyst treats cases of " living etymology", i.e. words whose foreign origin is obvious and, therefore, performs a stylistic function. The

circumstance that both lexicology and stylistics have recourse to terms like metaphor or metonymy is explained by the fact that neither of them belongs to either lexicology or stylistics: though used in both, they are terms of 'semasiology' (science of meanings) and 'onomasiology' (science of nomination). The subject matter of these branches is dealt with below.

And, finally, general (non-stylistic) syntax treats word combinations and sentences, analysing their structures and stating what is permissible and what is inadmissible in constructing correct utterances in the given language. The field of action of stylistic syntax is the same, but its ap­proach and its aims are, as the reader is supposed to guess by now, quite different. The stylistic study of syntax (called here stylistic syntax) shows what particular constructions are met with (or should be employed) in various types of speech, what syntactical structures are style-forming (specific) in the sublanguage in question, Besides, stylistic syntax very often operates on longer units, from the paragraph upwards.

It should be remarked here that most handbooks on phonetics or grammar (morphology and syntax), not to speak of lexicology, abound in stylistic information. Whenever the sphere of currency of a unit (or of a phenomenon) is explicitly mentioned, it is pure stylistics that the author deals with. If a phonetician informs the reader about the emphatic intonation as compared with non-emphatic, he acts as a stylist. If a grammarian admonishes the reader not to use the Nominative Absolute (John having returned, we began to work) in colloquial speech, it is stylistic syntax the grammarian is operating with.

Semasiology, onomasiology, and stylistics. Along with their formal characteristics, linguistic units (with the exception of phonemes) have meanings. Morphemes, words, word combinations, and sentences arouse in our minds associations with classes of things, processes, qualities, relations. The content associated by most people with the form of a linguistic unit is its meaning. Meanings are investigated and described by a branch of linguistics called semantics, or semasiology (the latter term is preferable, since the word semantics happens to be used instead of the word meaning).

Certain scholars (those who think the word level to be synonymous to aspect or sphere — see above) say or imply that meanings of linguistic units make an independent level — that of semantics, or semantic level.

Needless to say, this point of view is not valid if we have agreed above that levels are formed by material units (phonemes, morphemes, words, etc.). Meanings are notions, ideas, mental images, and in consequence they cannot be placed on a par with sound complexes, letter combinations, or other material manifestations of whatever kind.

But the most important reason why meanings are not 'level-forming' (in the above-accepted sense of the word level) is that meanings, as

suggested in the beginning of this discourse, associate with morphemes, words, phrases, sentences. That is to say, meanings are not attached to only one level, but, practically speaking, correlate with all of them (save the phonemes, which have no extralingual meanings, only serving to form units of the next, i.e., morphemic, level and differentiating one morpheme or one unimorphemic word from another: pin — tin — fin — bin — kin, etc.). Characteristically, similar or identical meanings maybe conveyed by units of different levels. Compare: -less = without = devoid (of) = which does not possess.

For the reasons discussed, semasiology is, one might say, an 'all-level' discipline. It is practically of no importance then, for stylistic semasiology, if it deals with meanings of morphemes, or the meanings of parts of a compound word, or with the meaning of a word, a phrase, a sentence and so on.

The same can be stated with regard to onomasiology (or: onomatology), the theory of naming. The difference between the two (semasiology and onomasiology) is as follows. If we take a linguistic unit of any rank (from morpheme upwards) and begin our inspection of its meaning (more often, meanings), of its combinability with other units of identical rank, or, to put it in a different manner, if we proceed from form to meaning, this is the way a semasiologist acts. If, on the contrary, we have an idea aroused in our mind by an external object and search for the wording, for the adequate 'name' of the idea or, more generally, if a linguist proceeds from meaning to form, his standpoint conforms to the general principle of onomasiology.

From these statements we can draw far-reaching conclusions. First: semasiology treats semantic structures of linguistic units, yet certainly having their spheres of use in view. Onomasiology treats problems of choice of linguistic units for naming extralingual objects (things, proper­ties, relations, situations). Again, as previously, the comparative analysis of choice cannot go on without the data of semasiology. They are dialectically interwoven; their differentiation shows merely the general directions of research.

Further, it can be seen that a semasiological approach fits for analysing separate units of the vocabulary, e.g. words, especially their historical development. Hence the problems of eventual degradation or elevation of meanings, their extension or narrowing, metaphoric and metonymic changes of meaning of words — all that constitutes the usual problems of etymological studies in lexicology.

By contrast, the problem of what word, phrase, sentence was chosen and used to characterize a certain object in the text, what transfer of the name occurred in using it with reference to an unusual object (whether it was, for instance, a metaphor, metonymy, or irony) is the business of

onomasiology, and if the specificity of the sublanguage has been considered, the business of stylistic semasiology.

Paradigmatics and syntagmatics. These terms are derived from the words paradigm and syntagma. The former should be known to the learner from such combinations as the paradigm of declension (or conjugation). The expression denotes all the grammatical forms of a noun (verb) that co-exist at the moment and could be presented in the form of a list to select from (this concerns mostly inflected languages, such as Latin, German, or Russian). The latter (syntagma) has also been used in phonetics or in syntax. It usually denotes a combination of words in speech and text, a linear sequence of lexical units.

The derivative paradigmatics, often used by Russian linguists, denotes the totality of units of which language (or sublanguage) has at its disposal. Or, otherwise, the units taken together make a paradigm.

As distinct from it, the term syntagmatics, implies a totality, or a certain number of sequences of units, of chains of units following one another. Here, the units do not co-exist simultaneously ready to be chosen by the speaker (writer) for his communicative purposes, but on the contrary, each unit enters into combinations with its neighbours, with what precedes it and follows it.

Certain linguists have said that paradigmatics represents language as a system, while syntagmatics characterizes speech as a process in its development, or text, which indeed has a linear form. At first glance this explanation seems correct and even self-evident, but on closer inspection of the matter we shall come to a different result.

In fact, what is a paradigm? Only separate phonemes, or morphemes, or separate words? Of course not. Word combinations, sentences (or sentence patterns), paragraphs, and even types of texts, if arranged together as possibilities from which one selects the necessary form (the stylistically suitable variety) make up their own paradigms, too.

And, further, can it be true that language-as-a-system is only a paradigm, and that syntagmata occur only in speech? Where should they come from if they have not existed previously in our minds, i.e. in our lingual systems (the systems that we acquired in early childhood by listening to the speech of our elders)? It would really be absurd to presume that a person who knows only words could make sentences without knowing how sentences are made. Hence: language itself, language-as-a-system has both paradigmatic and syntagmatic aspects.

What would a paradigm of only elementary units be without the knowledge of how the elements are combined in syntagmata? It would not only be useless, but even meaningless. We know the meaning of an elementary unit solely due to the knowledge of its distribution. And vice versa, what could a syntagma be without its place in the paradigm? We

should know nothing about it. Hence it follows that both syntagmatics and paradigmatics are non-existent without each other. It is true, of course, that in speech (or text), syntagmatics comes to the foreground: we perceive only syntagmata; yet in language — let us say it again — both paradigmatics and syntagmatics are represented. Perhaps we had better say that language is a system in which syntagmata as well as their constituents (elements) are presented paradigmatically.

Both are connected and interwoven; neither is thinkable without the other. For stylistics, however, strict delimitation of them is of great im­portance, as we are going to see further.

The most general way of subdividing stylistics is not into level-forming branches as was tentatively assumed above. It is connected with the opposition of paradigmatics to syntagmatics.

The nature, the essence of stylistic phenomena is radically different in cases where a unit itself (of whatever length — a phoneme, morpheme, word, phrase, sentence pattern, paragraph structure) is analysed as chosen out of the paradigm (and potentially opposed to those left unchosen), from cases when we try to explain the effect produced by a given pattern of combining units (also of whatever rank) in speech and text. In the former approach to the material we pass our judgement on what a unit is worth by itself; in the latter, it is the result, the stylistic value of the combination that the analyst is after. Here, the important point is that the units 'co-appear', 'co-occur' in the same text, either close to one another, or at a distance from one another. To put it in a more explicit manner, in the co­occurrence of units it is their interrelation that is stylistically relevant.

Two groups of examples will suffice to illustrate the difference be­tween the two aspects of stylistics, the two branches of stylistics, the two systems of stylistics, in fact.

The first. When we use the word guy (instead of man), the form ain't (instead of have not, am /is /are not), the word combination real good (instead of really good), the sentence John here? (instead of Is John here?), it is one unit used instead of another (or others) which could also be employed (but they were not employed). This is what illustrates the paradigmatic branch of stylistics. Practically, in the whole of our previous discussions of stylistic problems we dealt only with cases of paradigmatic choice.

The second group of examples is of a somewhat new kind: they have not yet been discussed above. Here, we will mention only a few instances of what pertains to syntagmatics.

In the utterance I ask you, I pray, I beseech you.' it is not the verbs or their meanings that are stylistically conspicuous, but the interrelation between the meanings expressed: pray is stronger than ask; beseech is the strongest of all three. The expression really and truly contains two

synonyms, therefore we observe equality of the two notions. In the combinations, however, like life and death, black and white, now or never the notions are contrasted, opposed to each other. The famous Shakespearian paradoxes loving hate and heavy lightness show quite a different relation between logically incompatible notions: the writer treats them as if they were compatible, and he does it not without a purpose of his own.

Those were examples showing interrelations of meanings, of semantic units. But stylistic phenomena (stylistic means, or stylistic devices) also arise due to interrelation of one-level units, e.g. speech sounds. Here is a sample of repetition (recurrence) of the same consonant in the beginning of several words, either following one another or co-occurring at a small distance:

Through florescence and feud, frosts and fires it followed the laws of progression even in the Forsyte family... (Galsworthy)

The vocabulary of the text likewise participates in determining its stylistic links with the syntagmatic aspect. A stylistically coloured word predominates over its neutral environment, imparting to the context its own stylistic value. The syntagma (sentence) This man is dippy is low colloquial as a whole owing to the presence of the low colloquial word dippy (cf. the neutral variant This man is crazy). The result of combining a stylistically coloured element with neutral ones is always the same: the non-neutral (specific) clement imparts its colouring to the context. More complicated are cases of co-occurrence of units with stylistically opposite connotations (e.g. 'high' and 'low') in the same utterance:

Philoprogenitiveness, says we, is strong in semi-rural communi­ties. (O. Henry)

The first and the two last words are presumptuously bookish, while the form says we is illiterate, not likely to be used by educated people. The effect is that of a stylistic mixture, in fact a stylistic collision. Neither stylistic party wins, and the utterance is comical.

It may also be mentioned briefly that syntagmatical (syntactic) pat­terns, following each other in the text, are in most cases formally dif­ferent. But quite often the second utterance is syntactically assimilated to the first, resulting in parallelism: The cock is crowing. The stream is flowing... (Wordsworth). Two or more contiguous sentences or paragraphs not infrequently have identical beginnings or identical endings (all this will be discussed in detail further).

What has been said in this section leads us to the only feasible conclusion. Paradigmatic choice of units and types of co-occurrence of units in syntagmata (sequences) ought to make two separate branches of stylistics. Yet, this differentiation has seldom if ever been made. At best,

English linguists discern 'tropes' and 'figures of speech', the former being transfers of names, i.e. results of choice (paradigmatics), the latter, combinations of meanings or formal units (syntagmatics). But there is no consequent differentiation of cases when the individual unit is the object of research, and cases when the interrelation of co-occurring units is what the linguist is after. Very often tropes are classed indiscriminately as figures of speech, too, along with genuine 'figures', i.e. configurations consisting of several words.

The careless attitude of many scholars to problems of classification is best seen in their enumerations or arrangements of items discussed. Thus M. Deutschbein, a German researcher of English stylistics, provides one of the chapters of his book with the title " Simile, Metaphor, and Quotation". Obviously, the simile (expressive comparison) is a. syntagmatic phenomenon, the metaphor is a trope, a paradigmatic re-naming. As for quotations, they are literary devices very remotely con nected with stylistics. Still more amazing is the enumeration of stylistic phenomena we find in An Outline of French Stylistics (" Precis do stylistique franc, aise") by J. Marouzeau: " ellipsis, anacoluthon, metaphor litotes". The reader may take this enumeration with equanimity, but, imagine a professional grammarian arranging the chapters of his book as follows: " The predicate, the possessive case, the past continuous tense, the article, the adverbial modifier". What should we take this author for? And where should he be taken?

Summing up, we must say that the material presented in this section urges us to divide the whole of stylistics into two parts: