Notebook, 1993-

Excerpts from: Frye, Northrop. Anatomy of Criticism, Four Essays. Princeton, NJ: Princeton University Press, 1957.

Introduction - Archetypal - Ethical - Historical - Rhetorical

Third Essay.
Archetypal Criticism: Theory of Myths

In the art of painting it is easy to see both structural and representational elements. A picture is normally a picture "of" something: it depicts or illustrates a "subject" made up of things analogous to "objects" in sense experience. At the same time there are present certain elements of pictorial design: what a picture represents is organized into structural patterns and conventions which are found only in pictures. The words "content" and "form" are often employed to describe these complementary aspects of painting. "Realism" connotes an emphasis on what the picture represents; stylization, whether primitive or sophisticated, connotes an emphasis on pictorial structure. Extreme realism or the illusive or trompe l'oeil type is about as far as the painter can go in one kind of emphasis; abstract, or, more strictly, non-objective painting is about as far as he can go in the other direction. 'The phrase "non-representational painting" seems to me illogical, a painting being itself a representation.' The illusive painter however cannot escape from pictorial conventions, and non-objective painting is still an imitative art in Aristotle's sense, and so we may say without much fear of effective contradiction that the whole art of painting lies within a combination of pictorial "form" or structure and pictorial "content" or subject.

For some reason the traditions of both practice and theory in Western painting have weighed down heavily on the imitative or representational end.... [tendency to mirror-like trompe l'oeil reflections of human imagery in the Classical, Renaissance development of perspective to suggest 3 dimensions in a two-dimensional medium, and the strength and persistence of the feeling that to achieve recognizable likeness in a subject, and to make this likeness the primary thing in his picture is a moral obligation on the painter - as we eavesdrop in a modern art gallery . . . . ] p. 131

An original painter knows, of course, that when the public demands likeness to an object, it generally wants the exact opposite, likeness to the pictorial conventions it is familiar with. Hence when he breaks with these conventions, he is often apt to assert that he is nothing but an eye, that he merely paints what he sees as he sees it, and the like. His motive in talking such nonsense is clear enough: he wishes to say that painting is not merely facile decoration, and involves a difficult conquest of some very real spatial problems. But this may be freely admitted without agreeing that the formal cause of a picture is outside the picture, an assertion which would destroy the whole art if it were taken seriously. What he has actually done is to obey an obscure but profound impulse to revolt against the conventions established in his own day, in order to rediscover convention on a deeper level. By breaking with the Barbizon school, Manet discovered a deeper affinity with Goya and Velasquez; by breaking with the impressionists, Cézanne discovered a deeper affinity with Chardin and Masaccio. The possession of originality cannot make an artist unconventional; it drives him further into convention, obeying the law of the art itself; which seeks constantly to reshape itself from its own depths, and which works through its geniuses for metamorphosis, as it works through minor talents for mutation.

Music affords a refreshing contrast to painting in its critical theory. When perspective was discovered in painting, music might well have gone in a similar direction, but in fact the development of representational or "program" music has been severely restricted. Listeners may still derive pleasure from hearing external sounds cleverly imitated in music, but no one asserts that a composer is being a decadent or a charlatan if he fails to produce such imitations. Nor is it believed that these imitations are prior in importance to the forms of music itself, still less that they constitute those forms. The result is that the structural principles of music are clearly understood, and can be taught even to children.

Suppose, for example, that the present book were an introduction to musical theory instead of poetics. Then we could begin by isolating, from the range of audible sounds, the interval of the octave, and explain that the octave is divided into twelve [p. 132] theoretically equal semitones, forming a scale of twelve notes which contains potentially all the melodies and harmonies that the reader of the book will ordinarily hear. Then we could abstract the two points of repose in this scale, the major and minor common chords, and explain the system of twenty-four interlocking keys and the conventions of tonality which require that a piece should normally open and close in the same key. We could describe the basis of rhythm as an accentuation of every second or every third beat, and so on through the whole list of rudiments.

Such an outline would give a rational account of the structure of Western music from 1600 to 1900, and, in a qualified and more flexible but not essentially different form, of everything that the user of the book would be accustomed to call music. If we chose we could lock up all the music outside the Western tradition in the solitary confinement of a prefatory chapter, before we got down to serious business. Someone might object that the system of equal temperament, in which C sharp and D flat are the same note, is an arbitrary fiction. Another might object that a composer ought not to be tied down to so rigidly conventionalized a set of musical elements, and that the resources of expression in music ought to be as free as the air. A third might object that we are not talking about music at all: that while the Jupiter Symphony is in C major and Beethoven's Fifth is in C minor, explaining the difference between the two keys will give nobody any real notion of the difference between the two symphonies. All these objectors could be quite safely ignored. Our handbook would not give the reader a complete musical education, nor would it give an account of music as it exists in the mind of God or the practice of angels--but it would do for its purposes.

In this book we are attempting to outline a few of the grammatical rudiments of literary expression, and the elements of it that correspond to such musical elements as tonality, simple and compound rhythm, canonical imitation and the like. The aim is to give a rational account of some of the structural principles of Western literature in the context of its Classical and Christian heritage. We are suggesting that the resources of verbal expression are limited, if that is the word, by the literary equivalents of rhythm and key, though that does not mean, any more than it means in music that its resources are artistically exhaustible. We doubtless have objectors similar to those just imagined for music, saying [p. 133] that our categories are artificial, that they do not do justice to the variety of literature, or that they are not relevant to their own experiences in reading. However, the question of what the structural principles of literature actually are seems important enough to discuss; and, as literature is an art of words, it should be at least as easy to find words to describe them as to find such words as sonata or fugue in music.

In literature, as in painting, the traditional emphasis in both practice and theory has been on representation or "lifelikeness." When, for instance, we pick up a novel of Dickens, our immediate impulse, a habit fostered in us by all the criticism we know, is to compare it with "life," whether as lived by us or by Dicken's contemporaries. Then we meet such characters as Heep or Quilp, and, as neither we nor the Victorians have ever known anything much "like" these curious monsters, the method promptly breaks down. Some readers will complain that Dickens has relapsed into "mere" caricature [as though caricature were easy]; others, more sensibly, simply give up the criterion of lifelikeness and enjoy the creation for its own sake.

The structural principles of painting are frequently described in terms of their analogues in plane geometry [or solid, by a further reach of analogy]. A famous letter of Cézanne speaks of the approximation of pictorial form to the sphere and the cube, and the practice of abstract painters seems to confirm his point. Geometrical shapes are analogous only to pictorial forms, not by any means identical with them; the real structural principles of painting are to be derived, not from an external analogy with something else, but from the internal analogy of the art itself. The structural principles of literature, similarly, are to be derived from archetypal and analogic criticism, the only kinds that assume a larger context of literature as a whole. But we saw in the first essay that, as the modes of fiction move from the mythical to the low mimetic and ironic, they approach a point of extreme "realism" or representative likeness to life. It follows that the mythical mode, the stories about gods, in which characters have the greatest possible power of action, is the most abstract and conventionalized of all literary modes, just as the corresponding modes in other arts---religious Byzantine painting, for example--show the highest degree of stylization in their structure. Hence the structural principles of literature are as closely related to mythology and comparative religion as [p. 134] those of painting are to geometry. In this essay we shall be using the symbolism of the Bible, and to a lesser extent Classical mythology, as a grammar of literary archetypes.

In the Egyptian tale of The Two Brothers, thought to be the source of the Potiphar's wife story in the Joseph legend, an elder brother's wife attempts to seduce an unmarried younger brother who lives with them, and , when he resists her, accuses him of attempting to rape her. The younger brother is then forced to run away, with the enraged elder brother in pursuit. So far, the incidents reproduce more or less credible facts of life. Then the younger brother prays to Ra for assistance, pleading the justice of his cause; Ra places a large lake between him and his brother, and, in a burst of divine exuberance, fills it full of crocodiles. This incident is no more a fictional episode than anything that has preceded it, nor is it less logically related than any other episode to the plot as a whole. But it has given up the external analogy to "life": this, we say, is the kind of thing that happens only in stories. The Egyptian tale has acquired, then, in its mythical episode, an abstractly literary quality; and, as the story-teller could just as easily have solved his little problem in a more "realistic" way, it appears that literature in Egypt, like the other arts, preferred a certain degree of stylization.

Similarly, a medieval saint with a huge decorated halo around his head may look like an old man, but the mythical feature, the halo, both imparts a more abstract structure to the painting and gives the saint the kind of appearance that one sees only in pictures. In primitive societies, a flourishing development in myth and folk tale usually accompanies a taste for geometrical ornament in the plastic arts. In our tradition we have a place for verisimilitude, for human experience skillfully and consistently imitated. The occasional hoaxes in which fiction is presented, or even accepted, as fact, such as Defoe's Journal of the Plague Year or Samuel Butler's The Fair Haven, correspond to trompe l'oeil illusions in painting. At the other extreme we have myths, or abstract fictional designs in which gods and other such beings do whatever they like, which in practice means whatever the story-teller likes. The return of irony to myth that we noted in the first essay is contemporary with, and parallel to, abstraction, expressionism, cubism, and similar efforts in painting to emphasize the self-contained pictorial structure. Sixty years ago, Bernard Shaw stressed the social significance of the themes in Ibsen's plays and his own. Today, [p. 135] Mr. Eliot calls our attention to the Alcestis archetype in the Cocktail Party, to the Ion archetype in The Confidential Clerk. The former is of the age of Manet and Degas; the latter of the age of Braque and Graham Sutherland.

We begin our study of archetypes, then, with a world of myth, an abstract or purely literary world of fictional and thematic design, unaffected by canons of plausible adaptation to familiar experience. In terms of narrative, myth is the imitation of actions near or at the conceivable limits of desire. The gods enjoy beautiful women, fight one another with prodigious strength, comfort and assist man, or else watch his miseries from the height of their immortal freedom. The fact that myth operates at the top level of human desire does not mean that it necessarily presents its world as attained or attainable by human beings. In terms of meaning or dianoia, myth is the same world looked at as an area or field of activity, bearing in mind our principle that the meaning or pattern of poetry is a structure of imagery with conceptual implications. The world of mythical imagery is usually represented by the conception of heaven or Paradise in religion, and it is apocalyptic, in the sense of that word already explained, a world of total metaphor, in which everything is potentially identical with everything else, as though it were all inside a single infinite body.

Realism, or the art of verisimilitude, evokes the response "How like that is to what we know!" When what is written is like what is known, we have an art of extended or implied simile. And as realism is an art of implicit simile, myth, is an art of implicit metaphorical identity. The word "sun-god," with a hyphen used instead of a predicate, is a pure ideogram, in Pound's terminology, or literal metaphor, in ours. In myth we see the same structural principles [not similar ones] fitting into a context of plausibility. [Similarly in music, a piece by Purcell and piece by Benjamin Britten may not be in the least like each other, but if they are both in D major their tonality will be the same.] The presence of a mythical structure in realistic fiction, however, poses certain technical problems for making it plausible, and the devices used in solving these problems may be given the general name of displacement.

Myth, then, is one extreme of literary design; naturalism is the other, and in between lies the whole area of romance, using that term to mean, not the historical mode of the first essay, but the [p. 136] tendency, noted later in the same essay, to displace myth in a human direction and yet, in contrast to "realism," to conventionalize content in an idealized direction. The central principle of displacement is that what can be metaphorically identified in a myth can only be linked in romance by some form of simile: analogy, significant association, incidental accompanying imagery, and the like. In a myth we can have a sun-god or a tree-god; in a romance we may have a person who is significantly associated with the sun or trees. In more realistic modes the association becomes less significant and more a matter of incidental, even coincidental or accidental, imagery. In the dragon-killing legend of the St. Goerge and Perseus family, of which more hereafter, a country under an old feeble king is terrorized by a dragon who eventually demands the King's daughter, but is slain by the hero. This seems to be a romantic analogy [perhaps also, in this case, a descendant] of a myth of a waste land restored to life by a fertility god. In the myth, then, the dragon and the old king would be identified. We can in fact concentrate the myth still further into an Oedipus fantasy in which the hero is not the old king's son-in-law but his son, and the rescued damsel the hero's mother. If the story were a private dream such identifications would be made as a matter of course. But to make it a plausible, symmetrical, and morally acceptable story a good deal of displacement is necessary, and it is only after a comparative study of the story type has been made that the metaphorical structure within it begins to emerge.

In Hawthorne's The Marble Faun the statue which gives the story that name is so insistently associated with a character named Donatello that a reader would have to be unusually dull or inattentive to miss the point that Donatello "is" the statue. Later on we meet a girl named Hilda, of singular purity and gentleness, who lives in a tower surrounded by doves. The doves are very fond of her; another character calls her his "dove," and remarks indicating some special affinity with doves are made about her by both author and characters. If we were to say that Hilda is a dove-goddess like Venus, identified with her doves, we should not be reading the story quite accurately in its own mode; we should be translating it into straight myth. But to recognize how close Hawthorne is to myth there is not unfair. That is, we recognize that The Marble Faun is not a typical low mimetic fiction: it is dominated by an interest that looks back to fictional romance and forward to the [p. 137] ironic mythical writers of the next century--to Kafka, for instance, or Cocteau. This interest is often called allegory, but probably Hawthorne himself was right in calling it romance. We can see how this interest tends toward abstraction in character-drawing and if we know no other canons than low mimetic ones, we complain of this.

Or, again, we have, in myth, the story of Prosserpine, who disappears into the underworld for six months of every year. The pure myth is clearly one of death and revival; the story as we have it is slightly displaced, but the mythical pattern is easy to see. The same structural element often recurs in Shakespearean comedy, where it has to be adapted to a roughly high mimetic level of credibility. Hero in Much Ado is dead enough to have a funeral song, and plausible explanations are postponed until after the end of the play. Imogen in Cymbeline has an assumed name and empty grave, but she too gets some funeral obsequies. But the story of Hermione and Perdita is so close to the Demeter and Prosserpine myth that hardly any serious pretence of plausible explanations is made. Hermione, after her disappearance, returns once as a ghost in a dream, and her coming to life from a statue, a displacement of the Pygmalion myth, is said to require an awakening of faith, even though, on one level of plausibililty, she has not been a statue at all, and nothing has taken place except a harmless deception. We notice how much more abstractly mythical a thematic writer can be than a fictional one: Spenser's Florimell, for instance, disappears under the sea for the winter with no questions asked, leaving a "snowy lady" in her place and returning with a great outburst of spring floods at the end of the fourth book.

In the low mimetic, we recognize the same structural pattern of the death and revival of the heroine when Esther Summerson gets smallpox, or Lorna Doone is shot at her marriage altar. But we are getting closer to the conventions of realism, and although Lorna's eyes are "dim with death," we know that the author does not really mean death if he is planning to revive her. Here again it is interesting to compare "The Marble Faun", where there is so much about sculptors and the relation of statues to living people that we almost expect some kind of denouement like that of The Winter's Tale. Hilda mysteriously disappears, and during her absence her lover, the sculptor Kenyon, digs out of the earth a statue that he associates with Hilda. After that Hilda returns, with a plausible reason eventually assigned for her absence, but not without some rather pointed and petulant remarks from Hawthorne himself to the effect that he has no interest in concocting plausible explanations, and that he wishes his reading public would give him a bit more freedom. Yet Hawthorne's inhibitions seem to be at least in part self-imposed, as we can see if we turn to Poe's Ligeia, where the straight mythical death and revival pattern is given without apology. Poe is clearly a more radical abstractionist than Hawthorne, which is one reason why his influence on our century is more immediate.

This affinity between the mythical and the abstractly literary illuminates many aspects of fiction, especially the more popular fiction which is realistic enough to be plausible in its incidents and yet romantic enough to be a "good story," which means a clearly designed one. The introduction of an omen or portent, or the device of making a whole story the fulfillment of a prophecy given at the beginning, is an example. Such a device suggests, in its existential projection, a conception of ineluctable fate or hidden omnipotent will. Actually, it is a piece of pure literary design, giving the beginning some symmetrical relationship with the end, and the only ineluctable will involved is that of the author. Hence we often find it even in writers not temperamentally much in sympathy with the portentous. In Anna Karenina, for instance, the death of the railway porter in the opening book is accepted by Anna as an omen for herself. Similarly, if we find portents and omens in Sophocles, they are there primarily because they fit the structure of this type of dramatic tragedy, and prove nothing about any clear-cut beliefs in fate held by either dramatist or audience.

We have, then, three organizations of myths and archetypal symbols in literature.

First, there is undisplaced myth, generally concerned with gods or demons, and which takes the form of two contrasting worlds of total metaphorical identification, one desirable and the other undesirable. These worlds are often identified with the existential heavens and hells of the religious contemporary with such literature. These two forms of metaphorical organization we call the apocalyptic and the demonic respectively.

Second, we have the general tendency we have called romantic, the tendency to suggest implicit mythical patterns in a world more closely [p. 139] associated with human experience. Third, we have the tendency of "realism" [my distaste for this inept term is reflected in the quotation marks] to throw the emphasis on content and representation rather than on the shape of the story. Ironic literature begins with realism and tends toward myth, its mythical patterns being as a rule more suggestive of the demonic than of the apocalyptic, though sometimes it simply continues the romantic tradition of stylization. Hawthorne, Poe, Conrad, Hardy and Virginia Wolf all provide examples.

In looking at a picture, we may stand close to it and analyze the details of brush work and palette knife. This corresponds roughly to the rhetorical analysis of the new critics in literature. At a little distance back, the design comes in to clearer view, and we study rather the content represented: this is the best distance for realistic Dutch pictures, for example, where we are in a sense reading the picture. The further back we go, the more conscious we are of the organizing design. At a great distance from, say, a Madonna, we can see nothing but the archetype of the Madonna, a large centripetal blue mass with a contrasting point of interest at its center. In the criticism of literature, too, we often have to "stand back" from the poem to see its archetypal organization. If we "stand back" from Spenser's Mutabilitie Cantoes, we see a background of ordered circular light and a sinister black mass thrusting up into the lower foreground--much the same archetypal shape that we see in the opening of the book of Job. If we "stand back" from the beginning of the fifth act of Hamlet, we see a grave opening on the stage, the hero, his enemy, and the heroine descending into it, followed by a fatal struggle in the upper world. If we "stand back" from a realistic novel such as Tolstoy's Resurrection or Zola's Germinal, we can see the mythopoetic designs indicated by those titles. Other examples will be given in what follows.

We proceed to give an account first of the structure of imagery, or dianoia, of the two undisplaced worlds, the apocalyptic and the demonic, drawing heavily on the Bible, the main source for undisplaced myth in our tradition. Then we go on to the two intermediate structures of imagery, and finally to the generic narratives or mythoi which are these structures of imagery in movement.

Theory of Archetypal Meaning [1]

Theory of Archetypal Meaning [2]

Theory of Archetypal Meaning [3]

Theory of Mythos: Introduction
The Mythos of Spring: Comedy
The Mythos of Summer: Romance
The Mythos of Autumn: Tragedy
The Mythos of Winter: Irony and Satire

[Frye, Northrop. Anatomy of Criticism, Four Essays. Princeton, NJ: Princeton University Press, 1957.]



The contents of this site, including all images and text, are for personal, educational, non-commercial use only. The contents of this site may not be reproduced in any form without proper reference to Text, Author, Publisher, and Date of Publication [and page #s when suitable].