Art as Device

(1917)

Abbreviated

Victor Shklovsky

"Art is thinking in images." This maxim, which even high school students parrot is, nevertheless the starting point for the erudite philologist who is beginning to put together some kind of systematic literary theory. The idea, originated in part by Potebnya, has spread. "Without imagery there is no art, and in particular no poetry," Potebnya writes. And elsewhere, "Poetry, as well as prose, is first and foremost a special way of thinking and knowing."

Poetry is a special way of thinking; it is, precisely, a way of thinking in images, a way which permits what is generally called "economy of mental effort," a way which makes for "a sensation of the relative ease of the process." Aesthetic feeling is the reaction to this economy. Potebnya and his numerous disciples consider poetry a special kind of thinking—thinking by means of images; they feel that the purpose of imagery is to help channel various objects and activities into groups and to clarify the unknown by means of the known.

"Without imagery there is no art"—"Art is thinking in images." These maxims have led to far-fetched interpretations of individual works of art. Attempts have been made to evaluate even music, architecture, and lyric poetry as imagistic thought.

Nevertheless, the definition "Art is thinking in images," which means (I omit the usual middle terms of the argument) that art is the making of symbols, has survived the downfall of the theory which supported it. It survives chiefly in the wake of Symbolism, especially among the theorists of the Symbolist movement.

Many still believe, then, that thinking in images is the chief characteristic of poetry. Consequently, they should have expected the history of "imagistic art," as they call it, to consist of a history of changes in imagery. But we find that images change little; from century to century, from nation to nation, from poet to poet, they flow on without changing. Images belong to no one: they are "the Lord’s." The more you understand an age, the more convinced you become that the images a given poet used and which you thought his own were taken almost unchanged from another poet. The works of poets are classified or grouped according to the new techniques that poets discover and share, and according to their arrangement and development of the resources of language; poets are much more concerned with arranging images than with creating them. Images are given to poets; the ability to remember them is far more important than the ability to create them.

Imagistic thought does not, in any case, include all the aspects of art nor even all the aspects of verbal art. A change in imagery is not essential to the development of poetry. We know that frequently an expression is thought to be poetic, to be created for aesthetic pleasure, although actually it was created without such intent.

Thus a work may be (1) intended as prosaic and accepted as poetic, or (2) intended as poetic and accepted as prosaic. This suggests that the artistry attributed to a given work results from the way we perceive it. By "works of art," in the narrow sense, we mean works created by special techniques designed to make the works as obviously artistic as possible.

Potebnya’s conclusion, which can be formulated "poetry equals imagery," gave rise to the whole theory that "imagery equals symbolism," that the image may serve as the invariable predicate of various subjects. (This conclusion, because it expressed ideas similar to the theories of the Symbolists, intrigued some of their leading representatives—Andrey Bely, Merezhkovsky and his "eternal companions" and, in fact, formed the basis of the theory of Symbolism.) The conclusion stems partly from the fact that Potebnya did not distinguish between the language of poetry and the language of prose. Consequently, he ignored the fact that there are two aspects of imagery: imagery as a practical means of thinking, as a means of placing objects within categories; and imagery as poetic, as a means of reinforcing an impression: I shall clarify with an example. I want to attract the attention of a young child who is eating bread and butter and getting the butter on her fingers. I call, "Hey, butterfingers!" This is a figure of speech, a clearly prosaic trope. Now a different example. The child is playing with my glasses and drops them. I call, "Hey, butterfingers!" This figure of speech is a poetic trope. (In the first example, "butterfingers" is metonymic; in the second, metaphoric—but this is not what I want to stress.)

Poetic imagery is a means of creating the strongest possible impression. As a method it is, depending upon its purpose, neither more nor less effective than other poetic techniques; it is neither more nor less effective than ordinary or negative parallelism, comparison, repetition, balanced structure, hyperbole, the commonly accepted rhetorical figures, and all those methods which emphasize the emotional effect of an expression (including words or even articulated sounds) But poetic imagery only externally resembles either the stock imagery of fables and ballads or thinking in images. Poetic imagery is but one of the devices of poetic language. Prose imagery is a means of abstraction.

The law of the economy of creative effort is also generally accepted. Herbert Spencer wrote:

And Richard Avenarius:

Even Alexander Veselovsky acknowledged the principle of the economy of creative effort, a theory especially appealing in the study of rhythm, and agreed with Spencer: "A satisfactory style is precisely that style which delivers the greatest amount of thought in the fewest words." And Andrey Bely, despite the fact that in his better pages he showed the difficulties inherent in poetic epithets, also thought it necessary to speak of the law of the economy of creative effort in his book.

These ideas about the economy of energy, as well as about the law and aim of creativity, are perhaps true in their application to "practical" language; they were, however, extended to poetic language. Hence they do not distinguish properly between the laws of practical language and the laws of poetic language.

We must, then, speak about the laws of expenditure and economy in poetic language not on the basis of an analogy with prose, but on the basis of the laws of poetic language.

If we start to examine the general laws of perception, we see that as perception becomes habitual, it becomes automatic. Thus, for example, all of our habits retreat into the area of the unconsciously automatic; if one remembers the sensations of holding a pen or of speaking in a foreign language for the first time and compares that with his feeling at performing the action for the ten thousandth time, he will agree with us. Such habituation explains the principles by which, in ordinary speech, we leave phrases unfinished and words half expressed. In this process, ideally realized in algebra, things are replaced by symbols. Complete words are not expressed in rapid speech; their initial sounds are barely perceived.

This characteristic of thought not only suggests the method of algebra, but even prompts the choice of symbols (letters, especially initial letters). By this "algebraic" method of thought we apprehend objects only as shapes with imprecise extensions; we do not see them in their entirety but rather recognize them by their main characteristics. We see the object as though it were enveloped in a sack. We know what it is by its configuration, but we see only its silhouette. The object, perceived thus in the manner of prose perception, fades and does not leave even a first impression; ultimately even the essence of what it was is forgotten. Such perception explains why we fail to hear the prose word in its entirety and, hence, why (along with other slips of the tongue) we fail to pronounce it. The process of "algebrization," the over-automatization of an object, permits the greatest economy of perceptive effort. Either objects are assigned only one proper feature—a number, for example—or else they function as though by formula and do not even appear in cognition:

And so life is reckoned as nothing. Habitualization devours works, clothes, furniture, one’s wife, and the fear of war. "If the whole complex lives of many people go on unconsciously, then such lives are as if they had never been." And art exists that one may recover the sensation of life; it exists to make one feel things, to make the stone stony. The purpose of art is to impart the sensation of things as they are perceived and not as they are known. The technique of art is to make objects "unfamiliar," to make forms difficult, to increase the difficulty and length of perception because the process of perception is an aesthetic end in itself and must be prolonged. Art is a way of experiencing the artfulness of an object; the object is not important.

The range of poetic (artistic) work extends from the sensory to the cognitive, from poetry to prose, from the concrete to the abstract: from Cervantes’ Don Quixote—scholastic and poor nobleman, half consciously bearing his humiliation in the court of the duke—to the broad but empty Don Quixote of Turgenev. The meaning of a work broadens to the extent that artfulness and artistry diminish; thus a fable symbolizes more than a poem, and a proverb more than a fable. Consequently, the least self-contradictory part of Potebnya’s theory is his treatment of the fable, which, from his point of view, he investigated thoroughly. But since his theory did not provide for "expressive" works of art, he could not finish his book

After we see an object several times, we begin to recognize it. The object is in front of us and we know about it, but we do not see it – hence we cannot say anything significant about it. Art removes objects from the automatism of perception in several ways. Here I want to illustrate a way used repeatedly by Leo Tolstoy, that writer who seems to present things as if he himself saw them, saw them in their entirety, and did not alter them.

Tolstoy makes the familiar seem strange by not naming the familiar object. He describes an object as if he were seeing it for the first time, an event as if it were happening for the first time. In describing something he avoids the accepted names of its parts and instead names corresponding parts of other objects. For example, in "Shame" Tolstoy "defamiliarizes" the idea of flogging in this way: "to strip people who have broken the law, to hurl them to the floor, and to rap on their bottoms with switches," and, after a few lines, "to lash about on the naked buttocks." Then he remarks:

I apologize for this harsh example, but it is typical of Tolstoy’s way of pricking the conscience. The familiar act of flogging is made unfamiliar both by the description and by the proposal to change its form without changing its nature. Tolstoy uses this technique of "defamiliarization" constantly. The narrator of "Kholstomer," for example, is a horse, and it is the horse’s point of view (rather than a person’s) that makes the content of the story seem unfamiliar. Here is how the horse regards the institution of private property:

The horse is killed before the end of the story, but the manner of the narrative, its technique, does not change:

Thus we see that at the end of the story Tolstoy continues to use the technique even though the motivation for it [the reason for its use] is gone.

In War and Peace Tolstoy uses the same technique in describing whole battles as if battles were something new.

Anyone who knows Tolstoy can find several hundred such passages in his work. His method of seeing things out of their normal context is also apparent in his last works. Tolstoy described the dogmas arid rituals he attacked as if they were unfamiliar, substituting everyday meanings for the customarily religious meanings of the words common in church ritual. Many persons were painfully wounded; they considered it blasphemy to present as strange and monstrous what they accepted as sacred. Their reaction was due chiefly to the technique through which Tolstoy perceived and reported his environment. And after turning to what he had long avoided, Tolstoy found that his perceptions had unsettled his faith.

The technique of defamiliarization is not Tolstoy’s alone. I cited Tolstoy because his work is generally known.

Now, having explained the nature of this technique, let us try to determine the approximate limits of its application. I personally feel that defamiliarization is found almost everywhere form is found. In other words, the difference between Potebnya’s point of view and ours is this: An image is not a permanent referent for those mutable complexities of life which are revealed through it; its purpose is not to make us perceive meaning, but to create a special perception of the object—it creates a "vision" of the object instead of serving as a means for knowing it.

The purpose of imagery in erotic art can be studied even more accurately; an erotic object is usually presented as if it were seen for the first time. Gogol, in "Christmas Eve," provides the following example:

And Knut Hamsun has the following in "Hunger": "Two white prodigies appeared from beneath her blouse."

Erotic subjects may also be presented figuratively with the obvious purpose of leading us away from their "recognition." Hence sexual organs are referred to in terms of lock and key, or quilting tools, or bow and arrow, or rings and marlinspikes

But defamiliarization is not only a technique of the erotic riddle— a technique of euphemism—it is also the basis and point of all riddles. Every riddle pretends to show its subject either by words which specify or describe it but which, during the telling, do not seem applicable (the type: "black and white and ‘red’—read—all over) or by means of odd but imitative sounds ("‘Twas brillig, and the slithy toves/Did gyre and gimble in the wabe").

Quite often in literature the sexual act itself is defamiliarized; for example, the Decameron refers to "scraping out a barrel," "catching nightingales," "gay wool-beating work," (the last is not developed in the plot). Defamiliarization is often used in describing the sexual organs.

Such constructions as "the pestle and the mortar," or "Old Nick and the infernal regions" (Decameron), are also examples of the technique of defamiliarization. And in my article on plot construction I write about defamiliarization in psychological parallelism. Here, then, I repeat that the perception of disharmony in a harmonious context is important in parallelism. The purpose of parallelism, like the general purpose of imagery, is to transfer the usual perception of an object into the sphere of a new perception— that is, to make a unique semantic modification.

In studying poetic speech in its phonetic and lexical structure as well as in its characteristic distribution of words and in the characteristic thought structures compounded from the words, we find everywhere the artistic trademark—that is, we find material obviously created to remove the automatism of perception; the author’s purpose is to create the vision which results from that deautomatized perception. A work is created "artistically" so that its perception is impeded and the greatest possible effect is produced through the slowness of the perception. As a result of this lingering, the object is perceived not in its extension in space, but, so to speak, in its continuity. Thus "poetic language" gives satisfaction. According to Aristotle, poetic language must appear strange and wonderful; and, in fact, it is often actually foreign: the Sumerian used by the Assyrians, the Latin of Europe during the Middle Ages, the Arabisms of the Persians, the Old Bulgarian of Russian literature, or the elevated, almost literary language of folk songs. The common archaisms of poetic language, the intricacy of the sweet new style ,³⁵ the obscure style of the language of Arnaut Daniel with the "roughened" forms which make pronunciation difficult—these are used in much the same way.

Just now a still more characteristic phenomenon is under way. Russian literary language, which was originally foreign to Russia, has so permeated the language of the people that it has blended with their conversation. On the other hand, literature has now begun to show a tendency towards the use of dialects. And currently Maxim Gorky is changing his diction from the old literary language to the new literary colloquialism of Leskov. Ordinary speech and literary language have thereby changed places. And finally, a strong tendency, led by Khlebnikov, to create a new and properly poetic language has emerged. In the light of these developments we can define poetry as attenuated, tortuous speech. Poetic speech is formed speech. Prose is ordinary speech— economical, easy, proper, the goddess of prose is a goddess of the accurate, facile type, of the "direct" expression of a child. I shall discuss roughened form and retardation as the general law of art at greater length in an article on plot construction.

Nevertheless, the position of those who urge the idea of the economy of artistic energy as something which exists in and even distinguishes poetic language seems, at first glance, tenable for the problem of rhythm. Spencer’s description of rhythm would seem to be absolutely incontestable:

This apparently conclusive observation suffers from the common fallacy, the confusion of the laws of poetic and prosaic language. In The Philosophy of Style Spencer failed utterly to distinguish between them. But rhythm may have two functions. The rhythm of prose, or of a work song like "Dubinushka," permits the members of the work crew to do their necessary "groaning together" and also eases the work by making it automatic. And, in fact, it is easier to march with music than without it, and to march during an animated conversation is even easier, for the walking is done unconsciously. Thus the rhythm of prose is an important automatizing element; the rhythm of poetry is not. There is "order" in art, yet not a single column of a Greek temple stands exactly in its proper order; poetic rhythm is similarly disordered rhythm. Attempts to systematize the irregularities have been made, and such attempts are part of the current problem in the theory of rhythm. It is obvious that the systematization will not work, for in reality the problem is not one of complicating the rhythm but of disordering the rhythm—a disordering which cannot be predicted. Should the disordering of rhythm become a convention, it would be ineffective as a device for the roughening of language. But I will not discuss rhythm in more detail since I intend to write a book about it.