As O starts the book, the voice of the first narrator, Lockwood, supplants his own internal narrative voice and takes over the verbal stream of his consciousness. Georges Poulet articulated this experience as a form of colonization: “Because of the strange invasion of my person by the thoughts of another, I am a self who is granted the experience of thinking thoughts foreign to him . . . my consciousness behaves as though it were the consciousness of another.”1 The writer’s narration replaces the reader’s inner speech and, as he reads, images are produced spontaneously. He does not labor to create them. If O has never been to West Yorkshire, where the novel is set, his mental hills will look different from those of the reader who can summon a vivid memory of a countryside she visited last month.
As Mikel Dufrenne points out in The Phenomenology of Aesthetic Experience, “The represented world [of the book] also possesses, in its own fashion, the spatiotemporal structure of the perceived world. Space and time here fill a dual function. They serve both to open up a world and to ordain it objectively by creating a world common to the characters and the readers.”2 In order for O to invent the book he reads, to fill in what is not there, he must to some degree share its world, even if that world is a fantasy invented by the writer. And yet, the time and space of reading are not the same as the represented time and space of the novel. O reads for hours, but the events of Wuthering Heights take place over years.
When O arrives at page 20, Lockwood, the novel’s first narrator, has retired for bed inside a curious piece of furniture in a room of Heathcliff’s house: a casement that opens and closes; it resembles at once a tomb and a book. Cloistered in this small space, Lockwood notices that the name “Catherine” has been carved over and over into its inner walls. He finds and reads the youthful Catherine’s diary and then, as he fitfully tries to sleep, her ghost appears at the window. The panicked “I” narrates: “Terror made me cruel; and, finding it useless to attempt shaking the creature off, I pulled its wrist onto the broken pane, and rubbed it to and fro till the blood ran down and soaked the bedclothes: still it wailed, ‘Let me in!’ ”3 O shivers in his chair. Catherine’s ghost is a fiction (doubly so, in fact, because inside the world of the novel, Lockwood is probably dreaming), but the shiver, the fear O feels, is real. Poulet cites his emotional responses as proof of the power of the writer’s consciousness to become his own. “For how could I explain, without such take-over of my innermost subjective being, the astonishing facility with which I not only understand but even feel what I read.”4
Although most readers take for granted that a novel can amuse, torment, or bring them to tears, this truth has become a vexing conundrum for analytical philosophers, who have tied themselves in knots over the “paradox of fiction.” In 1975, Colin Radford argued that the feelings evoked by fictional characters are irrational, incoherent, and inconsistent.5 Isn’t it unreasonable to feel pity for Anna Karenina, awe for Heathcliff and Catherine, sympathy for David Copperfield, and admiration for Sherlock Holmes, persons who do not, in fact, exist? The roots of the paradox can be traced to the idea that logic is not a property of the human mind but eternal and objective, that propositions are the fundamental units of thought and meaning, and to multiple versions of the correspondence theory of truth. Bertrand Russell’s stark exposition may stand as exemplary: “Thus a belief is true when there is a corresponding fact, and is false when there is no corresponding fact.”6 The paradox turns on the nature of a propositional attitude, which is the mental state someone has in relation to a proposition, a belief, fear, or desire, which can then be declared true or false. I will restate the standard version of the paradox using myself:
Response condition: Siri Hustvedt, S.H., feels for David Copperfield, D.C.
Belief condition: S.H. knows D.C. is imaginary.
Coordination condition: In order for S.H. to have feelings for D.C., she cannot believe that D.C. is purely imaginary.
In order to solve the logical impasse, the philosopher Kendall Walton has proposed a theory of pretend or make-believe. Although readers believe they are feeling actual emotions for characters, and Walton does not dispute they may experience physiological changes of various kinds, what they feel are not real but “quasi-emotions.”7 I am afraid of a wolf on the road because I know it may eat me, but to fear a wolf in a fairy tale makes no sense. Therefore my emotion in relation to the fairy-tale wolf must be unreal or qualitatively different from the one I have in the presence of the actual wolf. I mention the paradox, which I think misconstrues our involvement with fiction, not only because it illuminates the constricted frame of analytical reasoning but because it offers insight into a genuine question: Why do we read fiction, and why do we care about fictional people? The ontology of fictional characters and the emotions they elicit in us clearly depend on how one understands the human mind and the imagination.
The word “imagination” has a labile history and is not easy to define. In Western philosophy, it traditionally served as a halfway house between sensation and intellect, a link between body and mind. In Critique of Pure Reason, Kant also conceives of the imagination as a mediator between the senses and the conceptual intelligence, which he argues is involved in perception itself. The imagination is preconceptual or “blind” and engages in “free play” with the understanding. I don’t know exactly what this Kantian free play is, but we can deduce that for the philosopher perception is not passive as it is for Hume, and that it includes unconscious “imaginative” processes, processes that reconfigure what we perceive and how we come to understand it. The following is Kant’s description of the imagination in the Critique of Judgment: “For the imagination (as a productive cognitive power) is very mighty when it creates, as it were, another nature out of the material that actual nature gives it. We use it to entertain ourselves when experience strikes us as overly routine . . . In this process we feel our freedom from the law of association (which attaches to the empirical use of the imagination); for although it is under that law that nature lends us material, yet we can process that material into something quite different, namely, into something that surpasses nature.”8 The mental transformation of one nature into another nature is a resonant idea. Imagination is, I think, part of all perception.
The biophysicist Hermann von Helmholtz, who was influenced by Kant to one degree or another (depending on whom you read), proposed the idea of “unconscious inference” in his 1867 Treatise on Physiological Optics. Through this inference human beings fill in the blanks and make sense of what they perceive through past experience, an idea now considered prescient in perception studies. As an illustration of unconscious inference, Helmholtz addresses the problem of the fictional character directly:
An actor who cleverly portrays an old man is for us an old man there on the stage, so long as we let the immediate impression sway us . . . He arouses fright or sympathy in us . . . and the deep-seated conviction that all this is only show and play does not hinder our emotions at all, provided the actor does not cease to play his part. On the contrary, a fictitious tale of this sort, which we seem to enter into ourselves, grips and tortures us more than a similar true story would do when we read it in a dry documentary report.9
Helmholtz recognizes not only that human emotions remain unaltered by a fictional state of affairs—in this case knowing that a young actor is playing an old man—but also that the style of the presentation affects the viewer. A well-done theatrical performance of Ibsen’s Hedda Gabler will have far more emotional impact on us than a newspaper story of a woman who committed suicide by shooting herself.
Unconscious inference relies on past perceptions and projects the old onto the new. Although some controversy remains, visual masking studies have shown over and over that people’s emotions are influenced by images that they do not consciously register.10 Our moods may sink into sadness or rise to elation without conscious explanation or with a false explanation. Infants are surely emotional, but it would be peculiar to argue that their fear upon heari
ng a loud noise, for example, is the result of propositional attitudes or self-reflective judgments about the sound. Joseph LeDoux has forcefully demonstrated that fear can be produced through subcortical routes that do not involve cortical areas necessary for conscious appraisals of a situation.11 Even before I know I am afraid, I may find myself frozen with fear. And it is hard to understand how a cognitive theory of emotion applies to other animals.
The problem with Walton and his like is that they have turned emotions into propositional attitudes or cognitions, which are amenable to formal logic. What amazes me is how influential this kind of analytical thinking has been in science. In their paper “Comparing Formal Cognitive Emotion Theories,” Koen V. Hindriks and Joost Broekens write, “Several formal emotion models have been proposed that derive emotions from the mental state of an agent using basic logic.”12 Hindriks and Broekens hope to take these logical formulations and create a formal computational model that can be implemented in a machine. I confess I don’t understand how a machine can mimic human feeling, but I may also be dense about their methods. It seems to me that a bodiless, nonorganic machine is a poor device to replicate the movements of a feeling self, unless emotion is turned into a form of symbolic thinking.
In his careful review of the many cognitive theories of emotion, Jesse Prinz writes, “I conclude that emotions are not in fact cognitive, most of the time. They are not generated by acts of cognition, and they are not conceptual . . . They do not decompose into meaningful, propositionally structured parts. They are not propositional attitudes.”13 A distinction between real and quasi emotions can exist only if emotions are like conscious beliefs, mental states that can be divorced from the sensory body.
Reading a novel obviously requires cognitive skills and appraisals of what is going on in the text, but once deciphering letters has been mastered, the act of reading itself is unconscious. O does not worry about sounding out the words on the page. The fear that visits him when Catherine’s ghost appears in the text is generated by the symbolic representations of an imaginary story, but his emotion is not a symbol of anything. Let us posit another reader, P, a woman. P begins reading Fanny Hill, John Cleland’s infamous erotic text. When P reads Fanny’s account of her mistress’s fingers wandering down to that part of her body our narrator refers to as “mount-pleasant,”14 P begins to feel an unmistakable sensation in the same part of her own body. Has P been gripped by a case of pseudo or quasi lust? The argument for quasi emotions is at once curiously static—it lacks a developmental trajectory—and it is weirdly disembodied; its reasoning veils a mind-body schism or rather it leaves the body altogether to argue for emotions as cognitive mental states, which can be judged real or quasi. That fundamental psyche-soma divide is not only ancient, it is weighted with contempt and fear for what has been viewed as the bottom, lower, debased half of the split—the body, long associated with femininity, unruly passion, unreason, and chaos.
Well over three hundred years before anyone thought to worry about what is now called “the paradox of fiction,” Margaret Cavendish proposed an anti-atomistic, antimechanistic (that is, anti-Hobbesian) dynamic, organic natural philosophy. In her 1666 Observations upon Experimental Philosophy, she writes, “I perceive man has a great spleen against self-moving corporeal nature, although himself is part of her, and the reason is his ambition; for he would fain be supreme, and above all other creatures.”15 Her insight is as mordant today as it was then. In Cavendish’s thought, perceptions are bodily imprints on sensitive matter, which are then reconfigured into thoughts, fancies, memories, and dreams.
But let us say for the moment that there is something illogical about caring about a fictional character. One may then ask if the very idea of a fictional character might not be broader than Walton and his fellow philosophers care to admit. When I remember an incident from my childhood—for example, the day I fell on the ice in the fifth grade and cut open my chin, an injury that required six stitches—should the images I conjure in my mind now of the child I was then be called real or imaginary? Memory may be reproductive, but it is notoriously unreliable. Through reconsolidation, conscious autobiographical memories are prone to all manner of transformation—two events with similar emotional impact may easily collapse into each other, for example, or we may unwittingly transfer a memory from one location to another. As in our dreams, our memories are subject to condensations and displacements. Freud’s Nachträglichkeit, which he introduced in a letter to Fliess in 189516 and later used to great effect in the case of the Wolf Man,17 anticipates the neurobiological concept of reconsolidation.
What we recover from the past is not an original memory but a revision, which it is fair to call an unconscious imaginative retranscription. In A Universe of Consciousness, Gerald Edelman and Giulio Tononi write, “If our view of memory in higher organisms is correct . . . every active memory is to some degree an act of imagination.”18 I do not know whether Edelman and Tononi’s emergent view of consciousness is correct, but that memory is imaginative seems undeniable. Is the memory of my self at ten and the mental images I summon in my mind less fictive than a fantasy of myself at seventy because I am convinced my ten-year-old self was once real? Would my sympathy for the child I recollect be true but my feelings for the older me I imagine be somehow false? Are the feelings I have for my dead father, a person who no longer exists, irrational because he is nonexistent? Or, for that matter, are the movie stars in the tabloids real people or fictions? Should the feelings of Schadenfreude they evoke in countless consumers be understood as quasi emotions? Doesn’t the art of fiction partake of precisely these linked processes—memory and imagination and those fleeting mental images that accompany them? Where does fiction begin and end?
So what does happen to O when he reads about the bleeding phantom and experiences that momentary shudder? Henry James famously wrote, “In the arts, feeling is always meaning.”19 The original ground of all meaning may well rest on the pain-pleasure spectrum of mammalian experience that is prelinguistic and preconceptual. Henry’s philosopher brother, William, did not treat aesthetics in a sustained way, but he subscribed to a somatic naturalism, in which the pleasures of art are connected to bodily instincts and appetites shaped by both our evolutionary and personal histories. “Aesthetic principles are at bottom such axioms as that a note sounds good with its third and fifth, or that potatoes need salt.”20 The Greek word aisthetikos, from which the word “aesthetic” is derived, means sensitive, perceptive, to perceive by the senses or by the mind. All works of art, including the novel, are animated in the body of the spectator, listener, or reader. Only an American pragmatist like James would bring up the homely image of salted potatoes in relation to aesthetics, but the image of tasty food is surely resonant in aesthetic considerations by returning the idea of aesthetic “taste” to the actual business of eating.
John Dewey, William James’s fellow Pragmatist, is emphatic about the embodied nature of art in his book Art as Experience. The existence of art, he writes, “is proof that man uses the materials and energies of nature with intent to expand his own life, and that he does so in accord with the structure of his organism—brain, sense-organs, and muscular system. Art is the living and concrete proof that man is capable of restoring consciously, and thus on the plane of meaning, the union of sense, need, impulse and action characteristic of the live creature.”21 In the tenth chapter, “The Varied Substance of the Arts,” of the same book, he makes the traditional argument that language reduces the infinitely subtle varieties of human experience to “orders, ranks, and classes that can be managed.”22 Then in his discussion of poetic language, Dewey quotes A. E. Housman, who had made the comment that poetry was “more physical than intellectual.” Paraphrasing the poet, Dewey writes that the effects of poetry can be recognized by “physical symptoms such as bristling of the skin, shivers in the spine, constriction of the throat, and a feeling in the pit of the stomach.” He makes an important distinction: “To be a thing and to be a sign of i
ts presence are different modes of being. But just such feelings, and what other writers have called organic ‘clicks,’ are the gross indication of complete organic participation, while it is the fullness and immediacy of this participation that constitutes the esthetic quality of an experience, just as it is that which transcends the intellectual.”23 In this view, O’s shiver is an organic click, triggered by signs, or to refine slightly what Dewey doesn’t articulate directly: without the reading body, the signs on the page are dead. They are animated only in the reader.
Undoing the paradox of fiction requires a revision of theories of language and meaning that suppress its physicality, of which emotion is a part. Language remains a subject of immense controversy. If language is understood as a logical system of signs, in which arbitrary symbols are manipulated according to syntactical rules, then the taste of salted potatoes and other clicks must be integrated into this disembodied linguistic view that binds formal logic, language, and computation. In some versions of language acquisition, an innate grammar, a language module, or instinct determines its course through development.24 Computational models have been devised to mimic reading comprehension, some of which involve computer simulations of what happens when we read. The following sentence appears in a book on computational models of reading comprehension: “These models involve processes that are assumed to occur during comprehension.”25 Their assumptions turn out to be substantial and, although forms of computation are clearly involved in reading comprehension, emotional responses to texts are treated nowhere in the book, not even in relation to meaning retention.
Embodied theories of language, following Dewey, oppose the claim that language production and comprehension can be divorced from lived, corporeal experience in the world. Imaginative thought is given a physical foundation. In their influential work on metaphor in 1980, Lakoff and Johnson situated conceptual thought in the perceptual and motor functions of the body.26 Since then a growing body of work in cognitive linguistics and neuroscience has implicated the sensorimotor cortices in semantic comprehension.27 As John Kaag writes in his paper “The Neurological Dynamics of the Imagination,” the research as a whole overturns the notion of language use as pure mentalist conceptualization. “This is not merely to make the claim that one needs a body to think, but rather the stronger claim that our bodies, and their relationship with their environmental situations, continually structure human thinking.”28 An even stronger claim than Kaag’s is this one: if this is the case, then the very idea of conceiving of a human subject in isolation from his world and others is nonsensical. The tripartite formula—subject, propositional attitude, proposition—begins to look very strange indeed.
A Woman Looking at Men Looking at Women: Essays on Art, Sex, and the Mind Page 53