Criticism and Taste: As Applicable to Motion Pictures

by Mortimer J. Adler, Ph.D.

Two general questions remain for discussion: (1) what is a good motion picture? and (2) what is a likable motion picture? The first states the problem of criticism; the second the problem of taste. A critical judgment differs from an expression of taste in that it considers a work of art in terms of its nature and technique; it says whether the work is well done. The judgment of taste says merely that we do or do not like the work, that we do or do not prefer it to some other. It refers the work to our capacity for enjoyment.

Criticism is evaluative in terms of standards of technical accomplishment. Taste is appreciative in terms of the quantity of aesthetic pleasure. The two judgments are capable of being made independently, yet they are obviously related [1]. What we mean by good taste is taste critically cultivated. A person has good taste if he gets pleasure in proportion to the objective goodness of the work. The discussion of taste properly follows, therefore, an analysis of the principles of criticism. In the case of a popular art, such as the motion picture, there is the problem of popular taste, which may or may not be the same as good taste. To the extent that these two are not the same, the problem of the relation of an art to considerations of taste is independent of aesthetic criticism.

The principles of criticism are, however, not independent of the preceding analysis of the specific nature and technique of the motion picture. Our prior discussion has, for the most part, been analytical rather than evaluative, but there have been unavoidable anticipations of the critical problem, indications that the analysis inevitably leads to the discrimination of good from bad workmanship. Thus, in the discussion of pictorial technique, the elements of the medium were isolated in such a way that it is now possible for us to distinguish good from bad style.

Similarly, in the discussion of cinematic narration, what is proper in the handling of plot, character and thought was suggested. The task now is to make such indications and suggestions explicit in an effort to formulate systematically the canons of criticism applicable to motion pictures.

One general insight explains the implications which the prior analysis has for criticism, the insight that an artist should not try to do more than he can with the elements of his medium, and should not be willing to do less. This is a critical insight. It is the most general principle of technical criticism. It marks the goal of workmanship in any field of art. And it explains why it is necessary to understand the specific nature and technique of a particular art in order to formulate critical standards specifically applicable to it. The rules of an art express the mature artist’s discipline and are prescriptive for the novice, forming his habits. These rules can become norms and it is through this transformation of requirements into standards that we pass from technique to criticism.

We are first concerned with the type of criticism which we have called technical, to distinguish it from (1) extrinsic or political criticism, and (2) the sort of intrinsic moral criticism which is the other part of aesthetic criticism. Here as before we are considering the work of the primary artist, the director, and perhaps also the work of one of his subordinates, the scenarist, with whom he must be closely united. If they are not the same person, it is the motion picture as their collaborative work which is to be criticized.

To the extent that the director should control the contributions of all the other auxiliary artists, their work is indirectly criticized through holding him responsible for it. The critical problem is divisible into two subordinate questions of style. “Style” is probably the best word to name all of the technical accomplishments of an artist. The work of any artist has style, but the style is not always good. When we say that work lacks style, we do so because we have identified style and good style. I shall make this distinction between two separable questions of style in terms of the arts of fiction. Analogous distinctions can probably be made for arts having other objects of imitation.

And I shall consider, first, the literary arts of fiction, which imitate action in the medium of language, be cause in this field the distinction is generally recognized between (1) narrative style and (2) linguistic style. The poet as a story teller is both a maker of plots and a maker in language. If there is any priority of the former to the latter sense in which he is a maker, it is because the object of imitation is prior to the medium. In any case, relative to these two respects in which he is a maker, he may be more or less technically accomplished. Whether these two styles are independent, whether the poet who has great narrative gifts may nevertheless write badly, is a difficult question. But the criteria of good style are, at least, analytically separable into these two dimensions.

The elements of narrative style are those elements of any art of fiction determined by the object of imitation: plot, character, thought. The elements of linguistic style are those elements of the literary arts of fiction determined by the medium of imitation: the elements of language. All of these elements are referred to by Aristotle’s single word “diction.” Diction is common to all literature. The elements of spectacle and song are peculiar to dramatic literature, not as literature but as produced theatrically. It may be asked whether there is any dimension of style determined by the manner of imitation. The answer is that the manner of imitation is involved in both narrative and linguistic style. The difference between the dramatic and the epic manner is not only a difference in the use of language but a difference in the treatment of plot and character.

If there is anyone for whom this analytical separation of narrative and linguistic style is not clear, he can be aided by the following consideration. Let us suppose a bilingual writer, a writer who has equal mastery of English and French. Such a writer must first decide whether he is going to write a novel or a play, after which he can conceive his narrative in terms of plot, character and thought, starting with a rough sketch of these elements and gradually increasing the detail. His conception may be relatively complete before he starts writing, and to this extent his narrative style is determined. But he cannot start writing without choosing his language. It is this choice which determines the appropriate problems of linguistic style, just as his decision to write a novel or a play previously determined the appropriate problems of narrative style. To whatever extent the choice of language and the actual writing alters the preconceived narrative, the two dimensions of style are not independent. It may even be that the preconception of the narrative influences to some extent his choice of the language in which to write. It is not being maintained that these two sorts of style are absolutely independent, but only that they are actually somewhat independent, as well as analytically separable.

The distinction made in terms of the literary arts of fiction holds perfectly for the motion picture as a non-literary member of this group of arts. The director is subject to criticism on two separate counts of style. Like any other worker in the field of fiction, he has a narrative style, good or bad. He must handle plot, character and thought in the cinematic manner. But he is not only a maker of plots, but a maker in the complex medium of pictures, words and sounds. Treating this manifold medium as integrated, we shall speak of filmic style as the analogue of linguistic style. Film is here understood as including both the photographic record and the sound track.

The analysis of filmic style is more complicated than that of linguistic style be cause of the complexity of the medium. We must distinguish pictorial style from the style of the sound track; each has its elements and its montage. Furthermore, filmic style involves the problem of total montage, the organization of the different components of the complex medium into the single continuity which is the motion picture. There is nothing analogous to this complexity in linguistic style. We do not, for instance, consider the dramatist’s linguistic style and the producer’s theatrical style as integrated parts of the same effort.

It is necessary to repeat once more the warning already given, that all of these separations are analytical only. The work of art is a unity of all its constituent elements, both those determined by the object of imitation and those determined by the medium. The writer, who is not our supposititious master of different tongues, does not actually separate his making of plot and his making in language. The director must think of the problems of story-telling and the problems of film-making at the same time.

The minimum condition of good style in all the arts of fiction is, therefore, to make good narrative sense, and this means a proper handling together of all of the elements which now, for the purposes of analysis, we shall separate. It is only the critic who is an analyst and therefore makes such separations. The artist is a creator, not an analyst. Nor is the audience, excepting the critic, analytical. The judg-ment of taste is an appreciation of the work of art as a whole because it is as a whole that it is enjoyed. But a critic must pay attention to the parts and the elements. It is for this reason that an over-developed critical faculty often hampers an artist or spoils enjoyment.

We shall deal first with the standards of good narrative style. For the most part, these are the same for all of the arts of fiction. In briefly summarizing them, our aim, therefore, must be to emphasize those aspects which are peculiar to cinematic narration.

Primacy of plot

To understand this principle, we must first distinguish between the theme of the plot and the developed plot. The theme of the plot is the matter of the story: the particular action being imitated. [2] Only two points are needed to determine the theme: a beginning and end, the beginning stated by the problem of the action and the choice, the end by the ultimate consequences of this choice. What lies between these two points is, as we shall see, the body of the story. The theme of a plot can be stated in a sentence or two. The fully developed plot cannot be stated except by the whole narrative.

  • Thus, Aristotle states the theme of the Odyssey as follows: “A certain man is absent from home for many years; he is jealously watched by Poseidon and left desolate. Meanwhile his home is in a wretched plight — suitors are wasting his substance and plotting against his son. At length tempest-tost, he himself arrives; he makes certain persons acquainted with him; he attacks the suitors with his own hand, and is himself preserved while he destroys them.” [3]
  • The theme of Crime and Punishment can be even more briefly stated: A man commits a crime and, after a period during which there is a growing suspicion of his guilt, is apprehended and punished.

It is clear that many different narratives may have the same theme. When it is said that there is only a small finite number of original plots, themes are meant. An enumeration of the themes of fiction could probably be made. It would be a classification of the particular types of action which fiction can imitate. The number of themes would be small because the number of moral problems is small and the variety of consequences following upon moral choices is limited.

A story-teller’s originality depends upon the way in which he develops the theme he has taken. When Aristotle says that the poet is a maker of plots, he does not mean that he is a maker of the themes of fiction, but of their development. The themes are not made. They are discovered or selected The poet makes a plot by taking a theme and adding to it all the particulars of narration: the incidents and episodes of the action the delineation of character, the expression of thought.

The best illustration of this point can be found in the Greek tragedies, many of which have the same theme. The three Electra-Orestes plays reveal that the work of Aeschylus, Sophocles and Euripides as plot-makers consisted in the development of the same theme: a son and daughter facing the problem of knowing or suspecting that their mother, aided by her lover, murdered their father, choosing revenge and reaping the consequences of matricide. This is the theme of Hamlet also, and of countless other plays, notably in recent years O’Neill’s Mourning Becomes Electra. These dramatic narratives differ in their treatment of the same theme: they all have differently developed plots.

This distinction between theme and development is important, negatively, in showing that the criticism of fiction should not be concerned with the theme. It is not a relevant point in criticism to say of a story that its plot is not original, if what is meant is its theme. Since an artist is a maker, originality is relevant to the criticism of his work, but the originality we must look for in fiction is in the plot-development.

It should be noted in passing that ethical criticism is similarly misdirected if it judges a story in terms of its theme. On the positive side, this distinction is important as showing that although theme is separable from character and thought, as well as from the incidents of the action — the developed plot involves not only the addition of all the incidents, but also of character and thought. The incidents of the action cannot be detailed without revealing character and thought. Aristotle at one point makes the extreme statement that a tragedy is possible without character, but not without action. This must be interpreted to mean that a plot cannot be developed without detailed incidents of action, but that the character of the agents or their thought need not be similarly detailed.

Illustrations of such plot development can be found in most melodramatic narratives, in detective and mystery stories. The Russian film Potemkin is an extraordinary example of a well developed plot without character or thought or, at least, with a minimum development of these elements. The extremity of Aristotle’s statement is, however, merely a way of insisting upon the primacy of plot. It should not be interpreted to mean that character and thought are not integrally related to the development of the plot through its constituent incidents.

That the plot is the primary element in narration depends upon the object of imitation. If a story were not the imitation of action in the political dimension, and of character and thought only as they are involved in such action, plot would not be primary. As we have seen, action necessarily includes character and thought as its causes and effects. Character and thought are habits or intrinsic operations which express themselves in the extrinsic behavior.

The primacy of plot means, therefore, that character and thought should be revealed by means of external action, and not directly and apart from action. But it may be objected that the primacy of plot depends not merely on the object of imitation, but also upon the manner of imitation; that it must be primary in the drama, but need not be so in the epic.

We have previously considered the psychological novel as an exception, or better as a paradoxical species of fiction in which character and thought become the primary objects of imitation because the epic manner affords the writer direct means of describing the introspective realm. A psychological novel is possible without action, or with the incidents of action given in a minimum of detail.

But whatever be the solution of the question whether such novels constitute a proper species of epic narration or whether they are violations of the art of fiction, it is at least clear that the dramatic manner of narration makes the psychological play either impossible or undesirable. The critical problem in the case of motion pictures is thereby solved. The cinematic manner of narration is more like the dramatic in that it cannot, without great difficulty, reveal thought and character directly. It should do so by means of the incidents of action, including, of course, the speeches. [4]

It follows, therefore, that the first criterion of good narrative style in motion pictures is the primacy of plot: it is an imitation of action through the incidents of action, and of character and thought subordinately by indirect means of revelation.

Unity of plot

Since we are concerned with the motion picture, we shall henceforth ignore narratives, such as the psychological novel, in which plot is not primary. The unity of the plot depends upon the unity of the action. This can be under stood negatively. A plot is not properly unified if it depends upon the unity of its hero or the unity of a problem or the unity of a period. In all of these cases, the plot development is bad because episodic. If its unity depends upon the singleness of its hero, any incidents are admitted into its structure so long as they are incidents in which the hero is an agent and whether or not they are causally related as the parts of a single action. Such narratives are like biographies, the unity of which is the life of a single person. The psychological novel may be like a biography, but the drama and the cinema should not be. [5]

The same can be said for the other types of inappropriate unity, illustrated by stories in which a number of different individuals and actions are put together because they are parallel instances of the same human problem, such as intolerance or ingratitude, or by stories in which the only unity is that all the events and persons occur at a certain time or at a certain place or somehow cross each other’s paths in space and time.

Positively, unity of plot can be understood in terms of the way in which the parts of a single action are organized into a whole. Unity of plot involves a unity in time, but not a unity of time: the action need not occur at one time, but the parts of it must be ordered sequentially in time. The principle of this ordering defines the unity of plot. The plot is divisible in two ways.

First, it can be divided into a beginning, a middle and an end. The beginning is constituted by the problem of the action, and by the choice among alternative courses of action which is made by the protagonist. The middle is constituted by the complications which follow upon this choice: the further choices which the protagonist makes because of the consequences of his first choice, and the consequences in turn of each of these choices.

It is in this part that most of the incidents and episodes of the action occur, that character is gradually revealed in greater detail and thought is more fully expressed. The inner complications of the action become interwoven with extraneous events in the outer world, which can be summarized as the good or bad fortune attending the career of the protagonist. This is what Aristotle means by saying that “incidents extraneous to the action are frequently combined with a portion of the action proper to form the complication.”

The progressive complication finally reaches a climax, a turning point in the story. After this point is the end, constituted by the denouement, a catastrophe or a benign resolution according as the story is tragic, comic or melodramatic, and an aftermath. This division of the parts of a single action indicates that the unity of the plot depends upon causality in the ordering of the incidents — not all the incidents because the extraneous ones happen as if by chance or fortune, but those incidents which proceed from the character and thought of the protagonist. Furthermore, the unity is emphasized by the fact that the denouement is the ultimate consequence of the original choice made with respect to the initial problem. It is this which binds the beginning, middle and end of a story together into a single whole.

The other division of the plot is into two parts: the complication and the unraveling, the former including everything from the beginning to the turning point of the action, the latter being what happens thereafter. This division shows the unity in terms of the crucial turning point, which must be the consequence of what precedes and the cause of what follows. The significance of this second division will be seen later in the point about the magnitude of the plot: it must be large enough to include a turning point that is intelligible in the light of what has gone before and is illuminated by what follows. The first division indicates another necessary feature of the plot structure: the middle part should always be the largest part. A story cannot be well told if too large a part of it is involved in getting the problem stated and the first choice made. The beginning is too large if it is larger than the middle. The same is true of the end.

The probability of the plot

This point follows in part from the rule that the incidents of the action must be causally related. A causal consequence is that which either happens necessarily as the result of some prior happening or that which happens for the most part. The incidents are probable, therefore, if they occur as normally they would in terms of human nature and the nature of the physical world. In other words, the sequence and conjunction of events which constitute the unified action of the plot must be such that the story is a likely or probable one.

The rule of probability thus applies not only to the action of the protagonist, but to the portraiture of character and the expression of thought. Even if the character is inconsistent or the thought irrational, it must be consistently inconsistent and irrational. “A person of a given character should speak or act in a given way, by the rule of necessity or probability, just as this event should follow that by necessary or probable sequence.”

The rule further applies even to the extraneous events that enter into the complication. Though they appear to the protagonist to happen as if by chance and as signs of good or bad fortune because they are not foreseen or ordained by him, they must nevertheless be probable incidents. It is the violation of this rule of probability which makes episodic plot development bad, and similarly plots in which character and thought are in consistent.

While it is generally recognized in criticism that a good story must be a likely story, the rule of probability to be followed in good plot construction is misunderstood whenever it is supposed that the criteria of probability in a poem are the same as in science. Poetic truth is not logical truth. What Aristotle says of tragedy, that “the element of the wonderful is required,” applies to an fiction. The good story-teller is always one like Homer, gifted in telling lies skillfully. “Accordingly, the poet should prefer probable impossibilities to improbable possibilities.” This indicates that the rule of probability is not the same in fiction as in science.

For knowledge, the impossible can never be probable. The probability of a story does not depend on the nature of things alone as does the probability of knowledge. It depends upon the art of the story-teller. The rule of probability is, therefore, the requirement that he make his story appear to be a likely one, whether or not its separate elements, viewed from the standpoint of science, are impossible or absurd or slightly probable. The impossible and the absurd are intolerable in fiction only if the narrator fails to veil them with poetic charm, which is another way of saying that he fails to make them seem probable. “Once the irrational has been introduced and an air of likelihood has been imparted to it, we must accept it in spite of its absurdity.” Aristotle goes so far as to approve of Agathon’s dictum that in story-telling even an improbable event can be made to appear probable because, as he says, “it is probable that many things should happen contrary to probability.”

The importance of this insight into the nature of probability in fiction cannot be overemphasized in the light of the tendency of current criticism to misunderstand the point. Much of the criticism of motion pictures uses the canon of probability as if the likelihood of a story depended upon its being life-like in the simple-minded sense of conforming to reality as it is. [6]

There is probably no greater error which the artist or critic can make than this simple-minded realism or naturalism. If the rule of probability be interpreted as a requirement that art be realistic or naturalistic, it falsifies the nature of art as imitation involving both similitude and difference. Far from being better because it is highly probable — in the sense of realistic — such a story is bad as a work of art. A highly fanciful tale, a tale that the realists would despise, is much better fiction if it satisfies the sole condition of being invested with poetic likelihood by narrative skill. In short, the principle of probability in artistic imitation differing from the principle of probability in science, determines two extremes which are bad: improbable fantasy, on the one hand, and “scientific” realism, on the other.

The story must be probable, but it must also be a story, and not a piece of faithful reporting. In other words, fiction is like history, but it is not history. The difference resides in the different conditions of probability that apply in each case. The Odyssey with all its impossible adventures on sea and land is a good story because of Homer’s great gift in telling lies, a much better story as a work of art than an accurate historical narrative of just what actually did take place in the voyage of Odysseus from Troy to Ithaca.

There is, of course, one further paradox involved. Even the historian or the realistic novelist at his worst extreme cannot avoid being an artist in fiction. He is always telling a story whether or not he is willing to acknowledge that the conditions of good story-telling are not the conditions of science. In a sense, realism and fantasy are impossible extremes. They are never really reached. There is no story which is totally devoid of probability nor one which is not a work of the imagination. The limits, therefore, merely indicate that a good story combines in proper proportion the factors of the wonderful and the probable. The artist who tries to be realistic never succeeds, but in trying so hard to go in one direction, he may fail to achieve a good proportion of these factors.

It is evidence of the essential rightness of Pudovkin’s under standing of the technique of the cinema that he always recognizes the pitfalls of naturalism. The tendency toward simple minded naturalism is more insidious in film-making than in writing, because of the superficially realistic character of photographs. It is this which makes montage crucially important, for it is by montage that naturalism can be most effectively avoided. But the basic principle of montage requires that film sequences be composed in a probable order, not the kind of probability which consists in fidelity to the way things actually appear, but the imaginative probability of the way in which things might appear to an ideal observer. We shall return to this point later in a discussion of filmic style. Here it is important only to note the way in which the rule of probability relates narrative and filmic style in the making of a motion picture.

To be good, a motion picture, like any other work of fiction, must avoid the extremes of reportorial realism and the improbably fantastic. Criticism which fails to understand this principle is as bad as art which futilely seeks to reach either extreme.

The structure of the plot

This locus of criticism has a number of subordinate topics. First, the simplicity or complexity of the plot. In one sense of complexity, the more complex plot is better; it not only contains a great number of incidents, both internal to the action and extraneous, but requires greater subtlety on the part of the narrator in ordering the incidents to bring about a progressive complication and a more striking climax.

The aesthetic principle here applies to all arts and not only to fiction: the artist must achieve a unity, and his skill is greater according as there is a greater multiplicity of parts out of which he makes the whole. But in another sense of complexity, the less complex plot is sometimes better and sometimes worse. Here we are concerned with the problem of subordinate plots of the sort which occur in Shakespearean tragedy, or with the double thread of plot which issues in a double denouement: prosperity or success for one person, and defeat or misfortune for another.

The critical standard must be differently applied to works that seek tragic effects, on the one hand, or comic and melodramatic effects, on the other. Simplicity of plot heightens the tragic effect. It concentrates all attention upon one protagonist; it makes his ultimate fate the single consequence of the action. It is questionable, therefore, whether the subordinate and often parallel plots in Shakespearean tragedy are not blemishes on their plot structure, despite the position of certain modern critics that the tragedy is thereby enriched. But in comedy and melodrama in which a happy ending is appropriate, it is no less appropriate to have the villain defeated or end ignominiously. Here happiness of the ending is increased by the inclusion of misfortune for the antagonist along with prosperity for the protagonist.

Aristotle’s comment on the story which has a double thread of plot is worth noting here. “It is accounted the best,” he says, “because of the weakness of the spectators, for the poet is guided in what he writes by the wishes of his audience. The pleasure, however, thence derived is not the true tragic pleasure. It is proper rather to comedy” and, I would add, melodrama. That most motion pictures are either comedies or melodramas may be due to the wishes of the audience, their preference for the double thread of plot. We shall return to this point later in a discussion of the problem of taste. It has no significance for criticism.

There is another distinction which Aristotle makes between simple and complex plots in tragedy; in the former the change of fortune takes place without reversal of situation and without recognition or realization. Reversal of situation occurs when an event suddenly produces the opposite effect to that which it at first portends; recognition or realization is a change from ignorance to knowledge about some matter that is crucially relevant to the motives and passions of the person involved. Both reversal and recognition involve the factor of surprise. It is clear, therefore, why the more skillful narrative, whether tragedy, comedy or melodrama, will have a complex plot in this last sense.

Second, the magnitude of the plot. This point is closely related to the preceding one. It differs in that the critical problem turns not upon whether the story seeks a tragic, a comic, or a melodramatic effect, but upon the manner of narration. Thus, the plot of any story must imitate an action having sufficient magnitude to permit a turn of fortune, which means that it must be large enough to have a beginning, a middle and an end as separate parts, but not so large that these parts cannot be viewed as a single continuity. But, as we have seen, the magnitude of an epic can be greater than that of a drama. It can involve more persons, more incidents, each of these parts having its proper magnitude. In other words, a novel can, and should, have a more complex unity than a play, more complex in the first two senses: having more parts, and having subordinate and parallel plots.

The cinematic manner of narration combines the features of both the novel and the play. The rule of magnitude for the motion picture must, therefore, be stated as follows. The magnitude of the plot must be thought of in two dimensions: (1) extensity, or the number of persons, incidents, parallel actions, etc., involved, and (2) intensity, or the amount of detail in the development of a single incident.

The epic magnitude is primarily in the dimension of extensity, although it may develop some of its more important incidents with dramatic intensity. The dramatic magnitude is primarily in the dimension of intensity. A motion picture must combine epic extensity with dramatic intensity. To do this, it cannot achieve the magnitude of either in its primary dimension.

Its proper magnitude, therefore, is a proportion between something less than the largeness of the epic in extensity and something less than the definiteness of the drama in detail. The narrative style of a motion picture is obviously bad if it transgresses this rule. It either becomes diluted as a succession of episodes which are thin because they lack dramatic definiteness, or it becomes cramped and motion-less be cause it has overindulged the dramatic depiction of a few of its incidents.

In short, just as Aristotle said that the plot of a poem must either be epic or dramatic in structure and not both at once, so we must say that the plot of a motion picture must be both epic and dramatic in structure, and not either exclusively. In other words, it must be cinematic. A motion picture is bad in narrative style if its plot is taken either from a play or a novel without the transforming work of adaptation.

All of these points of criticism can now be summarized in the single principle that good narrative style has unity, clarity and coherence. A story is a whole made up of parts. It must be sufficiently complex and subtle to engage attention, achieve suspense and surprise, and excite emotion, but not so complex and subtle that the unity, clarity and coherence of its parts are lost. If anything, most motion pictures are too simple and obvious in their narrative style.

The achievement of unity, clarity and coherence is not a mark of skill unless it is accomplished by a mastery of complexity and subtlety. This general principle of criticism applies to all works of fiction. Anyone who can judge good narrative in a novel, can do so also in a drama and a motion picture, provided only that he is sensitive to their essential differences as well as their essential sameness.

The criteria of goodness in filmic style

Because of the analogy between the elements of linguistic and filmic style, we can profit by a brief review of Aristotle’s standards of good writing. His first principle has a generality which has never been surpassed in later criticism. Good writing must have clarity without meanness or, in other words, it must be intelligible without being ordinary.

A style is ordinary or mean which lacks elevation and distinction, which is not unusual in its use of words. In order to achieve such elevation, a writer must employ unusual and strange words, invent new idioms and constructions. [7] But with elevation also comes subtlety, and subtlety is in a sense the opposite of clarity. Subtlety at the expense of clarity is bad, as is clarity without subtlety, resulting in meanness.

This basic principle can, therefore, be stated as the requirement that language be so used as to preserve a balance between clarity and subtlety, the latter for the sake of sublimity and elevation, the former for the sake of intelligibility. This principle of linguistic style neatly parallels the principle of narrative style which requires a similar balance between unity and complexity.

The principle applies both to vocabulary and to syntax: to the choice of words, and to phrasing, the invention of idioms, sentence and paragraph structure. It is most easily seen in the case of vocabulary. As previously pointed out, there could be no problem of style in the field of vocabulary did not language have the richness of synonyms. Otherwise there could be no choice of words.

In any group of synonyms, two kinds of words can be distinguished: (1) the ordinary words of current popular usage, and (2) unusual, strange words, invented words, or ordinary words somewhat altered by lengthening or shortening. If words of the first sort predominate, the style is to that extent mean and commonplace; if words of the second sort predominate, the style is to that extent lofty and distinguished. But there is the danger in using too many words of the second sort, that the writing will become an unintelligible jargon. There must, therefore, be a proportional use of words of both sorts to achieve both clarity and elevation.

The same analysis can be made in the field of syntax. There would be no problem of style here were it not possible to say the same thing in many different ways, that is, if many phrases and sentences were not related in the same way as synonymous words. On the one hand, a writer may use the ordinary constructions and idioms of common speech; on the other hand, a writer may invent new idioms, depart from the usual constructions in the direction of greater brevity or greater explicitness, or employ other sorts of strange and unfamiliar phrasing. If he writes exclusively in the first way, he writes clearly but in a commonplace manner. If he writes exclusively in the second, he is likely to become unintelligible. He must avoid both extremes: the commonplace and jargon too difficult to understand. Only in this way can he write with clarity and distinction.

To this first principle, Aristotle adds a second concerning the use of metaphor, either metaphorical words or metaphorical phrases. [8] He says it is of paramount importance to have command of metaphor: “it is a mark of genius, for to make good metaphors implies an eye for resemblances.” The reason why the use of metaphor is so important is not far to seek. Language is here being considered as a medium of imitation. Imitation depends upon similitude and difference. The metaphor is the condensation of an analogy or, perhaps, a whole series of analogies, involving many likenesses and differences.

Metaphors are therefore compact units of imitation. In them language is best adapted to the task of imitation. But there is a danger of excess here as before. A liberal use of metaphor is desirable for enriching the imitative symbolism of language, but a style too metaphorical becomes a riddle, as too many strange words and unusual constructions produce a jargon.

A third principle might be added: the appropriateness of the language to the nature and magnitude of the part of the plot being narrated. Good linguistic style, in other words, must be adapted harmoniously and in a different way to different phases of the narrative. Thus, to give one illustration of such correlation, certain parts of the narrative must move more rapidly, others more slowly. The writing must correspondingly have the same tempo, gaining speed by concentration, the use of suggestion and ellipsis, contrast, and so forth. An earlier question about the relation of narrative and linguistic style is thus in part answered. They should be correlated; the needs of the narrative usually determine the devices employed in language.

It should be noted above all that in this discussion of linguistic style, Aristotle is nowhere concerned with the distinction between verse and prose. Rather he is thinking of good writing, the perfection to be achieved in using the medium of language. The distinction between good and bad writing is a distinction between poetical and prosaic writing. The former is elevated and clear. The latter is clear but commonplace, or may even be commonplace without being clear. For the most part, the trait of elevation is most frequently achieved in verse, the trait of clarity most frequently achieved in prose. This is probably the reason why good writing which is both elevated and clear is called poetical, and writing which is merely clear is called prosaic.

There is, of course, nothing analogous to the distinction between prose and verse in filmic style, but as we shall now see there is a clear analogy of good and bad filmic style with poetical and prosaic writing.

The analogy is most easily grasped if we consider, first, the pictorial medium by itself. The further problems of filmic style which involve the combination of the pictorial sequences with the sound track are more complex than anything in the use of language. Furthermore, the pictorial style of a motion picture is, in a sense, its basic style, first because the pictures usually carry the narrative continuity into which sound and verbal elements are set, and second because it is by means of the pictures that metaphor, suggestion, ellipsis, contrast, condensation are best achieved in the film. It is through its pictures and not through its words that the cinema wins distinction in style.

The analogue of synonyms in vocabulary is the variety of ways in which the camera can be set and operated for shooting the same piece of action or the same object. As we have seen there is as much, if not more, variety in camera settings than there are synonyms for most words. All the possible camera settings fall into two groups: (1) the usual position, distance, angle and speed of ordinary vision and (2) the unusual, strange, and almost impossible position, distance, angle and speed — impossible in the sense that the eye could not see in the way the camera is able to. Shots from above or below, moving shots, certain types of close ups and telescopic shots, slow motion, special focus, are examples of the latter group; the middle distance, motionless, horizontal shot is the best example of the former.

There is similarly in montage the analogue of the different types of verbal syntax. Ordinary idiom and construction are like that cutting and joining of pieces of film which provide a customary sequence. The invented idiom and variant construction, the new metaphorical expression, are like the unusual filmic orderings produced by inventiveness in montage. The first criterion of good linguistic style, therefore, applies perfectly to work in the pictorial medium: the pictorial material must maintain a proper balance of clarity and elevation or subtlety, both in vocabulary and syntax.

Some of the better German and Russian films, particularly the early inventive ones, went too far in the direction of the unusual and strange in camera setting and montage, with a resultant loss of clarity. They were aiming in one right direction, but failed to preserve their balance. Most of the American films which are produced are clear enough, but lack any distinction, subtlety or elevation. They, too, are aiming in one right direction, but unfortunately not in the other. The danger of trying to be poetical without safeguarding clarity is that one becomes “arty”; the converse danger is that one be comes dull and commonplace. Of the two evils, the latter is worse because the former is a sign of invention and originality that needs only to be moderated in order to achieve the proper perfection of pictorial style.

We can also discover an analogy between the use of metaphor in language and the condensation of similitudes that can be obtained in the pictorial medium by the use of the camera and by montage. Thus, to take a simple example, the use of a blurred focus is the pictorial metaphor of something as if seen through a haze. The use of a certain camera mask is the obvious metaphor of something as if seen through a keyhole, and so forth. By cutting and joining two pieces of film that have certain parallel elements, the filmic analogue of metaphorical phrasing is achieved. There are countless other ways in which by a skillful use of pictures, likenesses can be suggested. But here, as in language, metaphorical excess is destructive of clarity. The filmic pieces or sequences can be made unintelligible in the effort to make them too compact of similitudes.

Finally, the third principle of pictorial style is the appropriateness of the pictorial devices to the part of the narrative which they are made to convey. Just as language can be used in such a way that its speed, its gravity, its definiteness or indefiniteness are fitting to the given part of a narrative, so the pictorial elements can be and should be adapted to different narrative purposes.

The criteria of good pictorial style are thus seen to be generally the same as the criteria of good linguistic style. But pictorial style is only one part of filmic style. We must now consider the other parts, the dialogue and the sound effects, and the ultimate problem of the organization of all of these parts into a filmic unity, which is the last problem of filmic style.

With respect to dialogue by itself, there is nothing to add to the criteria of good linguistic style, though, perhaps, the criterion of appropriateness should here be stressed. With respect to the sound effects by themselves, there is little if anything to say. They only present a problem in style when they are considered in relation to the pictures. This is the problem of the total montage of the film: the cutting and joining of the sound track, on the one hand, the pictorial pieces, on the other, and the composition out of these of the finished film. In terms of the task of total montage, it may be asked of particular units of dialogue or particular sound effects whether they are suitable. It is important that they be good in themselves, but this is not enough: they must also be good as parts of a whole. These two requirements some times conflict. When they do, the second should dominate the choice.

The final principle of good filmic style is that the total montage preserve a balance between clarity and variety. Variety is achieved by shifting the basic narrative thread from the pictures to the words and sounds, or from the latter to the pictures. The technique of total montage is still too young to permit an explicit formulation of the rules of parallelism, harmony and counterpoint, governing the correlation of the pictures and the sound track.

Yet Pudovkin’s insight that there is in this correlation something analogous to the melody and the accompaniment in polyphonic music is at present sufficient to enable us to discriminate distinctive filmic style from what is ordinary and commonplace. His insight can be stated in another way. Simple clarity is achieved to the extent that the montage is realistic or naturalistic. Elevation and variety are achieved to the extent that the montage is imaginative and inventive.

This final criterion of good filmic style is, therefore, a proper balance between realism and fantasy. It is not easy to accomplish. If most American films are, on the one hand, lethargically naturalistic in their style, the outstanding foreign films, particularly those of Germany and Russia, are often too radically fanciful. The former try to appear as if they did not employ the technique of montage at all and thereby lose distinction; the latter try to carry the technique of montage too far, and thereby lose clarity.

This concludes our formulation of the standards of criticism applicable to the cinema.


  1. Both the critical judgment and the judgment of taste are casuistical; hence uncertain, disputable and never conclusively established by any appeal to norms and principles. Criticism, “though it can always derives inspiration from philosophical principles — always a good thing, but risky — remains on the same plane as the work and the particular” — Jacques Maritain.

  2. The theme stands to the episodic details and the incidents as the form of the plot to its matter. It is in this sense that the theme is said to be the essence of the plot. Many plots may be identical in theme, differing individually in their accidents.

The universality of the theme, in contrast to the singularity of the thematic development, indicates that the object of imitation is, as Aristotle says, a sort of universal. It is not, in the case of poetry, a particular action but a kind of action. The same theme is common to many plots. This means that many stories imitate the same type of action. A type of action is a uniformity experienced, a universal in the imagination resulting from many perceptions and memories. It is not an explicit universal, fully abstracted and intellectually grasped. Each of the many stories sharing the same theme is individuated by the accidental details of plot development, an individuation due, not to the object of imitation, but to the artist’s unique composition of the elements of his medium. The object of imitation is never individual. If, in the case of the narrative arts, it were individual, there would be no distinction between fiction and history. The possible, which fiction imitates, must be universal. The actual, which history reports, must be individual.

  1. Poetics, 17, 1455bl8-2Z. Aristotle adds; “This is the essence of the plot the rest is episode.”

  2. Direct discourse is, for the novelist, an indirect way of revealing thought and character. He can describe them directly by indirect discourse. Many commentators on the film disagree with this judgment about the restriction of the cinema to dramatic surfaces. They think that the cinema has much greater power than the stage for psychological penetration.

  3. This does not mean that the film cannot be used as a medium for biography. Recently it has been well used in this way. But biography and fiction in the medium of language must be guided by different principles and subject to different standards of judgment. Similarly in the case of the film.

  4. Dr. Edgar Dale, How to Appreciate Motion Pictures, New York, 1935: “One of the most important things the motion picture can do is to show truthfully the consequences that come from making certain choices in life” (p. 96). See also pp. 206-208.

  5. Poetics, 22, 1458b l-3: “Nothing contributes more to produce a clearness of diction that is remote from commonness than the lengthening, contraction and alteration of words.”

  6. Poetics, 21, 1457b 7-9: “Metaphor is the application of an alien name by transference either from genus to species, or from species to genus, or from species to species, or by analogy, that is, proportion.”

[Great Books of the Western World GBotWW=”1″]