Dreaming by the Book

By Elaine Scarry

A pathbreaking work about the way literature teaches us to use our imagination.

We often attribute to our imaginative life powers that go beyond ordinary perception or sensation. In Dreaming by the Book, the noted scholar Elaine Scarry explores the apparently miraculous but in fact understandable processes by which poets and writers confer those powers on us: how they teach us the work of imaginative creation.
Writers from Homer to Heaney, Scarry argues, instruct us in the art of mental composition even as their poems progress: just as painters understand paint, composers musical sounds, and sculptors stone or metal, verbal artists understand and deploy the only material in which their creations will get made - the backlit tissue of the human imagination. In her brilliant synthesis of cognitive psychology, literary criticism, and philosophy, she explores the five principal formal practices by which writers bring things to life for their readers; she calls them radiant ignition, rarity, dyadic addition and subtraction, stretching, and floral supposition. The transforming power of these mental practices can be seen in their appearance in great literature, of course, but also in applying them to - and watching how they revise - our own daydreams.
Dreaming by the Book is not only an utterly original work of literary analysis but a sequence of on-the-spot mental experiments.

• Language: English
• Publisher: Macmillan Publishers
• Release date: May 21, 2013
• ISBN: 9781466845527

Elaine Scarry

Elaine Scarry, a professor of English and American literature at Harvard University, is the author of The Body in Pain. She lives in Cambridge, Massachusetts.

Book preview

The author and publisher have provided this e-book to you for your personal use only. You may not make this e-book publicly available in any way. Copyright infringement is against the law. If you believe the copy of this e-book you are reading infringes on the author’s copyright, please notify the publisher at: us.macmillanusa.com/piracy.

Contents

Title Page

Copyright Notice

Dedication

PART ONE: Making Pictures

1: On Vivacity

2: On Solidity

3: The Place of Instruction

4: Imagining Flowers

PART TWO: Moving Pictures

5: First Way: Radiant Ignition

6: Second Way: Rarity

7: Third Way: Addition and Subtraction

8: Fourth Way: Stretching, Folding, and Tilting

9: Fifth Way: Floral Supposition

PART THREE: Repicturing

10: Circling Back

11: Skating

12: Quickening with Flowers

Conclusion: Teaching Made-up Birds to Fly

Acknowledgments

Notes

Index

Other Books by Elaine Scarry

Copyright

for Jack Davis

PART ONE

Making Pictures

1

On Vivacity

When we speak in everyday conversation about the imagination, we often attribute to it powers that are greater than ordinary sensation. But when we are asked to perform the concrete experiment of comparing an imagined object with a perceptual one—that is, of actually stopping, closing our eyes, concentrating on the imagined face or the imagined room, then opening our eyes and comparing its attributes to whatever greets us when we return to the sensory world—we at once reach the opposite conclusion: the imagined object lacks the vitality and vivacity of the perceived one; it is in fact these very attributes of vitality and vivacity that enable us to differentiate the actual world present to our senses from the one that we introduce through the exercise of the imagination. Even if, as Jean-Paul Sartre observed, the object we select to imagine in this experiment is the face of a beloved friend, one we know in intricate detail (as Sartre knew the faces of Annie and Pierre), it will be, by comparison with an actually present face, thin, dry, two-dimensional, and inert.¹

It seems that we tend to notice the inadequacy of daydreamed faces only when we are especially keen on seeing a specific person’s face, only when we desperately care to have it present in the mind with clarity and force. We then notice the deficiency, and, like Proust’s Marcel, who berates himself for his inability to picture the face of Albertine or the face of his grandmother, we conclude that the vacuity of our imagining is somehow peculiar to our feeling about this particular person and that there must be a hidden defect in our affection. But the vacuity is instead general, and all that is peculiar or particular to such cases is the intensity of wishing to imagine that makes us confront, with more than usual honesty, the fact that we cannot do so. It is when we are soaked with the longing to imagine that we notice, as John Keats confessed, the fancy cannot cheat so well / As she is fam’d to do.² By means of the vividness of perceptions, we remain at all moments capable of recovering, of recognizing the material world and distinguishing it from our imaginary world, even as we lapse into and out of our gray and ghostly daydreams. Aristotle refers to this grayness as the feebleness of images. Sartre calls it their essential poverty.

Of course, insofar as the imagination is enfeebled and impoverished, it is so only on sensory grounds. To complain that the imagined object lacks vivacity and vitality is only to complain that it is not a perceptual object, since vivacity and vitality are the very heart of perception. We should not be surprised that the sensory realm surpasses the imaginary realm on sensory ground; we should only be surprised that this does not always strike us with the force of tautology. Phrased another way, only by decoupling vividness from the imaginary (where we unreflectingly and inaccurately place it in many everyday conversations about aesthetics) and attaching it to its proper moorings in perception, can we then even recognize, first, that the imagined object is not ordinarily vivid, and second, that its not being vivid is tautologically bound up with its being imaginary.

Now it is a remarkable fact that this ordinary enfeeblement of images has a striking exception in the verbal arts, where images somehow do acquire the vivacity of perceptual objects, and it is the purpose of this book to trace some of the ways this comes about. The verbal arts are of particular concern here because they—unlike painting, music, sculpture, theater, and film—are almost wholly devoid of actual sensory content. There is nothing mysterious about the fact that a painting approximates or exceeds the vivacity of the visible world, since it is itself a piece of the visible world. A painting by Henri Matisse or one of the great Florentine colorists, for example, saturates our eyes with actual sensory experience. The airy yellow and ochre stripes of Interior at Nice* set off an unceasing succession of retinal events that—carrying us out across the white shimmer of curtain, or back into the golden sheen held cuplike in the olive-green taffeta chair—are, whatever else they are, starkly perceptual acts. The same is true of music (why should it not share the vividness of the audible world when it is itself audible?), of sculpture (which inhabits, hence participates in, the vividness of the tactile and visual realms), and of theater and film (brimming with auditory and visual commitments). But verbal art, especially narrative, is almost bereft of any sensuous content.³ Its visual features, as has often been observed, consist of monotonous small black marks on a white page. It has no acoustical features. Its tactile features are limited to the weight of its pages, their smooth surfaces, and their exquisitely thin edges. The attributes it has that are directly apprehensible by perception are, then, meager in number. More important, these attributes are utterly irrelevant, sometimes even antagonistic, to the mental images that a poem or novel seeks to produce (steam rising across a windowpane, the sound of a stone dropped in a pool, the feel of dry August grass underfoot), the ones whose vivacity is under investigation here.

To be clear, it might be useful to distinguish three phenomena. First, immediate sensory content: the light-filled surface of Matisse’s Interior at Nice, the sweet fleeting notes of Honeysuckle Rose on Fats Waller’s piano recording, or indeed the particular room one, at this moment, inhabits while reading. Second, delayed sensory content, or what can be called instructions for the production of actual sensory content. A musical score has no immediate acoustical content, only the immediate visual content of lines and dots and the immediate tactile content of the smooth, thin pages, but it does directly specify a sequence of actions that, if followed, produces actually audible content. The third case, in contra-distinction to the first two, has no actual sensory content, whether immediate or delayed; there is instead only mimetic content, the figural rooms and faces and weather that we mimetically see, touch, and hear, though in no case do we actually do so.

It probably makes sense in this third case, as in the second, to use the word instructions. When we say Emily Brontë describes Catherine’s face, we might also say Brontë gives us a set of instructions for how to imagine or construct Catherine’s face. This reformulation is accurate if cumbersome, in that it shifts the site of mimesis from the object to the mental act. We habitually say of images in novels that they represent or are mimetic of the real world. But the mimesis is perhaps less in them than in our seeing of them. In imagining Catherine’s face, we perform a mimesis of actually seeing a face; in imagining the sweep of the wind across the moors, we perform a mimesis of actually hearing the wind. Imagining is an act of perceptual mimesis, whether undertaken in our own daydreams or under the instruction of great writers. And the question is: how does it come about that this perceptual mimesis, which when undertaken on one’s own is ordinarily feeble and impoverished, when under authorial instruction sometimes closely approximates actual perception? In the poem Birthplace, Seamus Heaney describes a young boy named Seamus Heaney staying up all night to read for the first time a novel (Thomas Hardy’s Return of the Native) and at dawn not knowing whether the newborn sounds of bird and rooster and dog were coming to him from the surface of the field or from the surface of the page. The question is: by what miracle is a writer able to incite us to bring forth mental images that resemble in their quality not our own daydreaming but our own (much more freely practiced) perceptual acts?*

Each of the arts incites us to the practice of all three acts: immediate perception, delayed perception, and mimetic perception. But painting, sculpture, music, film, and theater are weighted toward the first, or (perhaps more accurately) they bring about the second and the third by means of their elaborate commitments to the first; whereas the verbal arts take place almost exclusively in the third. Within the verbal, a further distinction must be made. Both narrative prose and poetry devote themselves centrally to mimetic perception, but poetry retains a strong engagement with delayed perception, the second category: like the musical score, its sequence of printed signs contains a set of instructions for the production of actual sound; the page does not itself sing but exists forever on the verge of song. Poetry—again unlike narrative—even has immediate sensory content, since the visual disposition of the lines and stanzas provides an at once apprehensible visual rhythm that is a prelude to, or rehearsal for, or promise of, the beautiful regulation of sound to come.

William Wordsworth describes two fish imprisoned in a glass bowl that, though they lack the song of larks and bees, produce a type of sun-writing in their glittering motions (their golden flash and silver gleam):

How beautiful!—Yet none knows why

This ever graceful change,

Renewed—renewed incessantly—

Within your quiet range.⁴

Is this beautiful display closer to the scattering of light in the yellow stripes of Matisse’s Interior at Nice (or, for that matter, his many paintings of goldfish), or is it instead like the scattering of light in the silver flash and gleam of the sword dance in Far from the Madding Crowd, or the lightning dance of Gabriel Oak when, sensitive of every ray, he secures the hayricks in a midnight storm?

Although its tone and content overlap with Matisse’s, there can be no question that, in terms of the categories posed here, Wordsworth’s sun-writing and Hardy’s light-writing are the same. Matisse’s colors bright and pigment sensitive of every ray are physically present and engage us in a starkly actual perceptual act, whereas Wordsworth and Hardy produce in our minds sudden radiant ignitions that are vividly mimetic of actually seen light but are not themselves actually seen. Yet because of the sound of the poem, the palpable touch of the interior parts of the mouth glancing across one another even in silent reading, and because of the visual scanning of the lines, the material surface of the poem is closer to the material surface of Matisse’s painting than is Hardy’s prose: that is, while Wordsworth is much closer to Hardy than to Matisse, he is a little closer to Matisse than Hardy is. Everything that I wish to say in what follows is as true of poetry as of prose, but it is harder to say clear sentences about the subject because one has to stop and qualify. Both prose and poem take place in the realm of the non-actual, but the poem is a few inches to the left of the narrative since it has its metrical feet in the material world. Therefore, in looking at how vivacity is achieved in the imagination, I will at first stay with prose. Prose requires of us neither immediate perception nor delayed perception; it instead requires non-actual or mimetic perception.

We shall find that imaginary vivacity comes about by reproducing the deep structure of perception. On one level this is wholly unsurprising: if imagining is a mimesis of perception, then successful imagining will of course come about through the accuracy or acuity of the mimesis. Still, it seems amazing that what in perception comes to be imitated is not only the sensory outcome (the way something looks or sounds or feels beneath the hands) but the actual structure of production that gave rise to the perception;⁵ that is, the material conditions that made it look, sound, or feel the way it did.⁶ I will illustrate this startling phenomenon with a very specific example, then turn back to the global features of narrative that also illustrate it.

2

On Solidity

Until now, I have framed the question of vivacity in a way that is equally applicable to imaginary seeing, imaginary hearing, imaginary touch, imaginary taste, and imaginary smell. For the duration of this chapter, it will instead be framed in a way uniquely relevant to touch.

How is the solidity of an imagined object achieved? To bring this about is key since solidity is ordinarily a main feature distinguishing perceptual objects from their gauzy counterparts in our imagination, where they are, to return to Sartre, lacking in density, thin, two-dimensional. The imagined object is not incidentally two-dimensional. Its two-dimensionality is what it is: Aristotle says in De Anima that images are like sensuous content except in that they contain no matter;¹ Ashbery describes in Tapestry the way the inner eye Draws an outline, or a blueprint, / Of what was just there: dead on the line.² The perceptible world, in contrast, is not just incidentally but essentially solid. John Locke urged that there is no idea we receive more constantly from sensation than solidity,³ and to his observation we may add the counterpart that there is probably no idea we receive less constantly in the ordinary exercise of the imagination.

How, then, does a writer get us to imagine a solid surface: a wall, for example, or the four walls that will make up a room? In his description of his childhood room at Combray at the opening of Swann’s Way, Proust describes the way the bright images from a magic lantern would play across the walls, overlaying the wall’s opaqueness with their own impalpable iridescence the figure of Golo, now moving in a jerky trot, now stopping quizzically, his form adaptable to any part of the wall:

And nothing could arrest his slow progress. If the lantern were moved I could still distinguish Golo’s horse advancing across the window-curtains, swelling out with their curves and diving into their folds. The body of Golo himself, being of the same supernatural substance as his steeds, overcame every material obstacle—everything that seemed to bar his way—by taking it as an ossature and embodying it in himself: even the door-handle, for instance, over which, adapting itself at once, would float irresistibly his red cloak or his pale face, which never lost its nobility or its melancholy, never betrayed the least concern at this transvertebration.

The room at Combray occasions, quite famously, Marcel’s meditations on habit: The mere change of lighting was enough to destroy the familiar impression I had of my room; the anaesthetic effect of habit [was] destroyed.⁴ But more fundamental than Proust’s philosophic speculation on habit is what he does not openly remark on: the perceptual mimesis of the solidity of the room brought about by the impalpable iridescence of Golo fleeting across its surfaces. It is not that we need attend to the solidity of the walls—quite the reverse: we simply assume them and, unimpressed by our miraculous hold on their solidity, go on to the seemingly more philosophic and psychologically complex issue of habit. Taken in isolation, the walls, the curtains, the doorknob are for the reader (as opposed to Marcel inside the book) certainly as thin and impalpable as the bright colored images issuing from the magic lantern. Yet by instructing us to move the one across the surface of the other, the transparency of one somehow works to verify the density of the other.

While Proust drenches the magic-lantern passage in the philosophic issue of habit, what is actually occurring is the slow coaxing into solidity of the Combray wall; it could even be said that the issue of habit in this passage, however fascinating in its own right, is an excuse for lingering on the walls precisely so that there will be time for the reader to bring about their materialization. Our belief in the solidity of these walls permits a phenomenon analogous to what takes place in the perceptible world. Locke says that in the ordinary operations of perception, the idea of solidity hinders our further sinking downwards; solidity establishes the floor beneath us that, even as we are unmindful of it, makes us cavalier about venturing out. The same can be said of the fictional walls. The idea of the solid wall prevents not our further sinking downward but our further sinking inward. It provides a vertical floor for all subsequent imaginings that lets us perform the projective act without vertigo or alarm, and thereby lifts the inhibitions on mental vivacity that ordinarily protect us.

But why should it be the case that the mental image of a wall can be coaxed into solidity by the passing of a transparent surface over it? The instruction for this reproduces the way solidity is visually inferred in the perceptible world. The movement of Golo across window drapes, wall, and door conflates two different phenomena that J. J. Gibson identifies in his classic study of perception. The first is called kinetic occlusion. If one surface passes in front of another surface—a picture dropping from a wall (Gibson’s example), or, more simply, my hand passing over my face—the movement of the object progressively covers and uncovers the physical texture of [the object] behind it. There occurs at the front edge of the moving object what Gibson calls a wiping-out or shearing-away of what lies behind, followed by its restoration: the wiping-out and shearing-across of texture in the array specify depth at an edge in the world. Furthermore, they specify durability, the continued existence of a hidden surface.⁵ Proust’s instructions to us in the Combray passage combine the assertion of depth and duration through kinetic occlusion with a second phenomenon, since the moving Golo is transparent (unlike the falling picture or my moving hand) and therefore provides only partial wiping out, only incomplete occlusion of the surface over which he passes. Analyzing the information that lies in the structure of ambient light, Gibson describes the nature of the shadow, which does not occlude the texture of the background as the object does, nor wipe it out when it moves. Hence a shadow is usually distinguishable from [a solid] object and also from a stain on the background.⁶ Drapes and walls, even the doorknob, are visibly ingested into Golo’s body. Their ongoing recoverability certifies that Golo is only a shadow, as in turn and more important his shadowiness continually confirms the solidity, the grabability, of the doorknob beneath. The difference here between the soft objects over which Golo passes (like the drapes) and hard objects (like the walls and doorknob) does not matter since, as Locke observes, solidity requires only repletion and hence is distinct from the attributes of hardness and softness: a solid object may be either hard or soft; hardness and softness, Locke writes, are merely judgments that express the relation of the thing to the constitutions of our own bodies.

Because I will be concentrating for quite some time on this minute inflection of something filmy passing over a wall, it may be useful to step back for a moment to see the overall architecture of the argument. First, solidity—if we may trust Locke in consultation with our own aliveness—is the key experience for percipient creatures; solidity relies on touch to provide access not just to material surfaces but to deep haptic experience as well. Second, solidity is difficult to reproduce in the imagination because it entails touch, the sense whose operation is most remote to us in imagining: Thomas Hobbes argued that the imaginary is exclusively visual; and certainly vision and audition are, by almost every theorist, credited as the main sensory candidates for the mimetic operations of imagining. Third, the very difficulty of achieving in the imaginary realm tactilely or haptically confirmed solidity is matched by the importance of doing so. It is impossible to create imaginary persons if one has not created a space for them. Speaking of the perceptible world, Locke says that "space in itself seems to be nothing but a capacity or possibility for extended beings, or bodies, to exist; space is only the consideration of a bare possibility of body to exist."⁷ When in the late pages of Remembrance of Things Past, Marcel is deserted by Albertine, he confesses that only the specification of the setting permits him to imagine her and to suffer in imagining, with acuity: How I repeated to myself these words, shed, passage, drawing-room, renewing the shock at will.… No, I had not imagined it at all, except as a vague dwelling. I had suffered first of all when the place where Albertine was had acquired a geographical identity.⁸ The operations of this principle explain why the most vivid writers are often the writers most deeply associated with place: Hardy and Wessex, Proust and Combray, Seamus Heaney and the bogs. Fourth and finally, even more important than the provision of an inhabitable space for imaginary persons is the creation for the reader of a fiction’s vertical floor that, by promising to stop our inward fall, permits us to enter capaciously into the projective space without fear and therefore with the lifting of inhibitions on vivacity.

The passing of a filmy surface over another (by comparison, dense) surface is not the only way of solidifying walls. But it is a key way; it recurs with writers universally saluted for their vividness, and it occurs precisely at moments where the newborn fictional worlds are most fragile and at risk because they are just in the midst of coming into being. Proust, for example, gets us to construct the room at Balbec in precisely this way: by laying forth a transparent layer of glass two feet out from the wall, then gliding over that glacine surface the moving reflections from the changing sea. Proust starts the passage by celebrating its sensory difference from the room at Combray. Yet the underlying instructions on how to construct it mentally are identical:

Among the rooms which used most commonly to take shape in my mind during my long nights of sleeplessness, there was none that differed more utterly from the rooms at Combray … than my room in the Grand Hôtel de la Plage, at Balbec, the ripolin-painted walls of which enclosed, like the polished sides of a bathing-pool in which the water glows blue, a finer air, pure, azure-tinted, saline.… [The room] which I found myself occupying had set against the walls, on three sides of it, a series of low book-cases with glass fronts, in which, according to where they stood, by a law of nature which [the designer] had not perhaps foreseen, was reflected this or that section of the ever-changing view of the sea, so that the walls were lined with a frieze of sea-scapes, interrupted only by the polished mahogany of the actual shelves.⁹

As at Combray, here at Balbec a film passes over and hence coaxes into solidity the walls behind it. As at Combray, where Proust engaged us in speculations about habit and defamiliarization (rather than conscious issues of solidity, transparency, complete and incomplete kinetic occlusion), here at Balbec Proust engages us in speculation about the difference between unreal oceans in the glassy mobile reflections and the real oceans somewhere outside, unseen. Yet precisely through that lengthy speculation, he requires us to keep steadily available to our inner mind the glassy reflections and the wall behind, thereby conscripting us into the production of solid walls: The walls were lined with a frieze of sea-scapes, interrupted only by the polished mahogany of the actual shelves.

The Combray–Balbec passages enable us to recognize three key solutions to the mystery of how the verbal arts enlist our imaginations in mental actions that in their vivacity more closely resemble sensing than daydreaming. Each of the three will let us move from the local instance of solidity to the more generalizable features of narrative. The first is the specification of the material antecedents of the perception to be produced. Second, while this mimesis entails a coaxing of the imagination into outcomes that move it beyond its ordinary Aristotelian enfeeblement, this is most strikingly accomplished when that very enfeeblement can successfully be enlisted into the process. Third, by the peculiar gravitational rules of the imagination, two or more images that are each independently weightless can nevertheless confer weight on one another; just as by the geometry of the imagination two or more images, each independently two-dimensional, can nevertheless confer three-dimensionality on one another. A fourth observation specifies a rule that regulates the preceding three: these methods of construction, so liberally present in these writings, are flatly withheld from the characters in those writings when they are trying to imagine absent persons or places. When poets and novelists get us to imagine imagining, they almost never permit the exposure of the rules they are themselves capaciously using.

I want to elaborate each of these four. But I do not want to suggest that what we are watching is a specifically Proustian solution. Here, then, is a quick set of examples that turn on this same phenomenon of conflating kinetic occlusion with transparency. A visually brilliant Japanese film cartoon, Hayao Miyazaki’s The Castle of Cagliostro,¹⁰ is,