THREE EMOTIONAL STORIES

IN A 1995 ESSAY ON memory, “Yonder,” I wrote the following sentence: “Writing fiction is like remembering what never happened.”1 It seemed to me fifteen years ago, and still seems to me today, that the mental activity we call memory and what we call the imagination partake of the same mental processes. They are both bound up with emotion and, when conscious, they often take the form of stories. Emotion, memory, imagination, story—all vital to our subjective mental landscapes, central to literature and psychoanalysis and, much more recently, hot topics in the neurosciences.

Ever since Plato banned poets from his Republic, philosophers have debated the role of imagination and its link to memory. Traditionally, imagination referred to the mental pictures we conjure in our minds, as opposed to direct perception. For thinkers as diverse as Aristotle, Descartes, Kant, and Hegel, the imagination occupied a middle zone between the bodily senses and intellect. Augustine connected imagination to both emotion and will. The will directs internal vision toward memories, but it also transforms and recombines them to create something new. “My memory,” he writes in the Confessions, “also contains my feelings, not in the same way as they are present to the mind when it experiences them, but in a quite different way that is in keeping with the special powers of the memory.” The emotions are all there, Augustine tells us—desire, joy, fear, and sorrow—and they can all be called to mind, but “if we had to experience sorrow or fear every time that we mentioned these emotions, no one would be willing to speak of them. Yet we could not speak of them at all unless we could find in our memory not only the sounds of their names, which we retain as images imprinted on the memory by the senses of the body, but also the idea of the emotions themselves.”2

Surely, Augustine’s thoughts remain cogent: remembering is not the same as perceiving. We remember what we have perceived, although we need ideas or concepts and names, language, to recognize and organize the material we have brought to mind. The seventeenth-century Italian philosopher and historian Giambattista Vico regarded memory and imagination as part of the same faculty rooted in sense perceptions. The imagination, he wrote, is “expanded or compounded memory,” and memory, sensation, and imagination are skills of the body: “It is true,” he insisted, “that these faculties appertain to the mind, but they have their roots in the body and draw their strength from it.”3

Vico’s comment is startlingly like the phenomenology of the twentieth-century French philosopher Maurice Merleau-Ponty (1908–1961), who understood imagination as an embodied reality, dependent on sensory perceptions, but which nevertheless allows us entrance into possible, even fictive spaces, l’espace potentielle, “potential space.” D. W. Winnicott used the same term in relation to his thoughts on play and culture. Unlike other animals, human beings are able to inhabit fictional worlds, to turn away from the phenomenal present and imagine ourselves elsewhere. Rats, even sea snails, remember, but they don’t actively recollect themselves as characters in the past or hurl themselves into imaginary futures.

How do we conceive of the imagination now? From Galen in the second century to Descartes and his pineal gland where he located phantasie or the imagination in the seventeenth, to phrenology in the nineteenth and Freud’s abandoned Project near the dawn of the twentieth, thinkers have sought the anatomical sites of mental functions. We have not solved the mystery of the mind’s eye, or what is now framed as a brain/mind problem. Terms such as neural representations, correlates, and substrates for psychological phenomena do not close the explanatory gap; they reveal it. There is a vast literature on this, and the debates are ferocious. A solution does not seem imminent. Suffice it to say that our inner subjective experience of mental images, thoughts, memories, and fantasies bears no resemblance to the objective realities of brain regions, synaptic connections, neurochemicals, and hormones, however closely they are connected.

I am not going to solve the psyche/soma problem here, but I can put some pressure on that old sentence of mine: Writing fiction is like remembering what never happened, or to rephrase it: How are remembering and imagining the same and how are they different?

The novelist, the psychoanalyst, and the neuroscientist inevitably regard memory and imagination from different perspectives. For the novelist, the story does all the work. When I am writing fiction, I am concerned with what feels right and feels wrong. I see images in my mind as I work, just as I do when I remember. Often I use landscapes, rooms, and streets that actually exist as backdrops for the actions of my fictional characters. I am directed by the story, by the creation of a narrative that resonates for me as emotionally, rather than literally, true. The novel develops an internal logic of its own, guided by my feelings.

For the analyst, a patient’s personal memories are crucial, but so are fantasies and dreams. They exist within the dialogical atmosphere of the analytic room and the abstract conceptual framework the psychoanalyst brings to his work. When listening to a patient’s memory, a psychoanalyst would keep in mind Freud’s idea of Nachträglichkeit, what James Strachey translated as “deferred action.” The adult patient may have memories from when he was five, but those memories have been reconfigured over time. The analyst would be alert to repetitive themes and defenses in his patient’s speech, but also voice cadences, hesitations, and, if his patient is looking at him, the motions and gestures of a body. What is created between analyst and patient is not necessarily a story that represents historical fact, but one that reconstructs a past into a narrative that makes sense of troubling emotions and neuroses. For patient and doctor, as for the novelist, the narrative must also be felt; it must resonate bodily as emotionally true.

The neuroscientist is trained to conceive of subjective memory and creative acts through objective categories, which she hopes will unveil the neurobiological realities of a self that both remembers and imagines. Following Endel Tulving and others, she will probably divide memory into three categories: 1) episodic memories, conscious personal recollections that can be localized to a specific place and time; 2) semantic memories, impersonal recall of information—cats have fur, Kierkegaard wrote under pseudonyms—and 3) procedural memories, unconscious learned abilities—riding a bike, reaching for a glass, typing.4 As a memory researcher, she would be aware of Joseph LeDoux’s work on the enduring synaptic connections formed by emotion in memory, fear in particular,5 and she would know that memories are not only consolidated in our brains, they are reconsolidated. Although it is unlikely that our neuroscientist has read Freud carefully, she would unwittingly agree with him that there is no original “true” memory; autobiographical memories are subject to change. Finally, theoretically, at least, her subjective feelings are irrelevant to her work.

In these three practices, we find the two modes of human thought which William James in his essay “Brute and Human Intellect” called narrative thinking and reasoning.6 Jerome Bruner, using a philosophical term, has called them two natural kinds, that is, essentially different.7 Novelists think in stories. Analysts use both narrative thought and the categorical thinking of reasoning. Scientists may employ a case history as an illustration, but their work proceeds without story. Reasoning is sequential but not dependent on a representation of time. Narrative is embedded in the temporal. Unlike the flux that characterizes narrative, scientific categories are static. Memory and imagination have to be approached from a third-person perspective and placed in a broader taxonomy. In the reasoning mode, definitions become all-important, and therefore a frequent battleground. First-person experience is vital to narrative because there is always an agent whose subjectivity and intentionality are part of the story’s movement, narrated from one perspective or another. In science the subject is nameless and normative.

What is the third-person point of view? Scientists cannot jump out of themselves and become God any more than the rest of us can. Through a largely unexamined agreement about what Thomas Kuhn in The Structure of Scientific Revolutions calls a paradigm—the bottom line of accepted theory, which changes over time, often in great convulsions—and an explicit consensus about methodology, scientists aim to avoid subjective bias.8 Although Freud never used the word neutral himself, the idea of the neutral analyst is a direct importation from the natural sciences. The omniscient narrator in some novels plays this role as he looks down from on high at his characters and their follies, but we readers know that clever as he may be, Henry Fielding’s narrator in Tom Jones is not God. Indeed, the cool “I-less” voice of nearly all academic writing adopts the pose of the third person. There is an author or authors, but they vanish as persons from the text. Still, as Kuhn argues in his book, there is no such thing as perceptual neutrality. The histories of science, psychoanalysis, and, of course, the novel, make this abundantly clear. This truth does not impede either discovery or innovation; it merely qualifies epistemology.

As Augustine points out, if we didn’t have names and ideas for things, we couldn’t speak of them at all. Both the reasoning and narrative modes of thought create linguistic representations. They exist on what the linguist Emile Benveniste refers to as a pronominal axis of discourse.9 The I implies a you, even if that I is just listening to one’s inner self. The language wars are as fierce as the brain/mind wars. Is there a universal grammar? as Noam Chomsky argued. Wasn’t Wittgenstein right that there is no such thing as a private language? How is language acquired, and exactly what does it have to do with our memories and imaginations? There is no consensus. I am sympathetic to A. R. Luria’s position that the advent of language reorders the mental landscape.10 I do not subscribe to the postmodern notion that it is the mental landscape. Nevertheless, whatever innate abilities we may have to learn it, language, which is both outside and inside the subject, plays a crucial role in our reflective self-consciousness, in how we become creatures of not only “I remember,” but “What if…?”

We codify perceptual experiences in conscious memory through both placement—where and when it happened—and interpretation, what it means in the larger context of my life. In our autobiographical memories, as in a mirror, we become others to ourselves. Even if we don’t see ourselves in the third person, we have projected the self elsewhere in time. As Merleau-Ponty notes in The Phenomenology of Perception: “Between the self which analyzes perception and the self which perceives, there is always a distance.”11 There is a difference, he argues, using Hegel’s distinction, between the “in itself” and the “for itself” (für sich).12 When I actively recall something from my past, what Augustine called “will” is involved. This is exactly how Aristotle distinguished human from animal memory. Only we people will ourselves backward in time.

And the episodic memories we recall have mostly been turned into stories. If narrative is, as Paul Ricoeur argues in Time and Narrative, a “grasping together” of various temporal actions or episodes13 in both life and fiction into a whole that has meaning, I believe that meaning is crucially grounded in emotion. It makes sense that narrative, a ubiquitous form of human thought, would, mimicking memory itself, focus on the meaningful and leave out the meaningless. What I am indifferent to, I mostly forget. The stories of memory and fiction are also made by absences—all the material that is left out.

As early as 1895, the psychologists Alfred Binet and Victor Henri tested children’s memories for lists of unrelated words as opposed to meaningful paragraphs.14 The children remembered the meaningful passages far better, but they reported them back to their examiners in their own words. They retained, to borrow a term from the Russian formalists, the fabula. Cinderella can be told in many different ways, and the details may vary, but the fabula, the bones of the story, remain the same. The narrative mode contextualizes the meaning or valence inherent in every emotion. It pulls together and makes sense of disparate sensory and affective elements.

Augustine’s insight that emotion dims in memory, however, is overwhelmingly true of our episodic memories. The cooling of the emotions that belongs to such recollections is built into the nature of this kind of memory, because it is quickly turned into narrative. The raw affective material of memories is restructured and then told as stories from a remove. Much, if not all, of this restructuring takes place unconsciously. When I remember, for example, that in 1982 I was hospitalized for eight days with an excruciating migraine that had lasted for many months, I do not reexperience either my pain or my emotional distress, although the pictures in my mind are colored gray for sadness.

I no longer remember what happened every day in the hospital, only a few highlights—a nurse who seemed to believe migraineurs were either neurotics or malingerers, the interns who asked me over and over who the president was, and my doctor who seemed exasperated that I didn’t get well. (All of their facial features are now visually dim.) I remember lying in the hospital bed, but I no longer see the room clearly. Still, I have a mental image that probably combines several hospital rooms I’ve visited or seen in the movies. We deposit memorable emotional events into a visual setting that makes sense, but what we see in our minds may bear little resemblance to what actually was. What I’ve retained is the story and a few serviceable mental pictures, but much is missing from that verbal account.

The fact that I used that hospital stay in my first novel, The Blindfold, further complicates matters because I turned an episode from my life into fiction, an episode I had already, no doubt, fictionalized in memory. Both the memory story and the novel were created unconsciously. Furthermore, I don’t truly recall my twenty-seven-year-old self. Too much time has intervened. I can easily shift the scene and see myself in the third person, a wan, blond young woman pumped full of Thorazine, staring at the ceiling. The hospital chapter of The Blindfold was turned into a movie, La Chambre des Magiciennes, by the French filmmaker Claude Miller. My experiences in Mount Sinai in 1982 generated three stories with the same fabula: my own narrative memory of an actual event, my character’s story in the novel based on that event, and my character’s story in the film, embodied by the actress Anne Brochet. Each one is different, and each one is constructed as a narrative, which partakes of the imaginary, the fictionalizing processes inherent to memories that are reflectively self-conscious.

Neuroscience research on the imagination is limited. However, a 2007 paper on patients with bilateral hippocampal lesions found they suffered from impaired imaginations as well as memory. A paper that same year by the same team, D. Hassabis et al., was published in The Journal of Neuroscience. This paper, “Using Imagination to Understand the Neural Basis of Episodic Memory,” based on fMRI scans, concludes, “… we have demonstrated that a distributed brain network, including the hippocampus, is recruited during both episodic memory recall and the visualization of fictitious experiences.”15 The activated part of the brain is a large cortical network, which has been implicated in “high level” cognitive functions, not only episodic memory, but future thinking, spatial navigation, theory of mind, and the default network. The participants were given three tasks that fell under the rubrics recall, re-create, and imagine. Notably, they were asked to keep all of these scenarios emotionally neutral.

The authors divided episodic memory into what they call “conceptual components,” among them a sense of subjective time, narrative structure, and self-processing. Although I enthusiastically endorse such research and believe episodic memory and imagination are fundamentally connected, I would like to focus on just one of their components: self-processing. The authors hypothesize that there will be less self-processing in imaginary other-oriented scenarios than in autobiographical ones, a reasonable thought until one asks oneself exactly what self-processing is. How exactly does an imaginary story I am generating about you, or her or him, not involve me? Aren’t all of these narratives—recalled, re-created, or imagined—related to my self, a part of my subjective experience? Furthermore, aren’t these narratives represented, at least in part, in language, and so necessarily located on the axis of discourse? There is no pronominal I without a you. When I think of you, are you not a part of me? What is being processed here? Shouldn’t neuroscience look to other disciplines to refine this vague idea: self-processing? Isn’t phenomenology’s concept of an embodied self useful in this regard? And what about psychoanalytic theory, with its internal objects, transference, and countertransference? Even with a cooling effect, can episodic memory and imagination really be entirely divorced from emotion?

In a 2009 comprehensive review of neuroimaging studies on self-processing versus other-processing in the Psychological Review, Dorothée Legrand and Perrine Ruby state: “The authors of the aforementioned studies … hypothesized that a given cerebral substrate should be systematically more activated for the self than for the nonself. Our review demonstrates that the cerebral network they identified does not exhibit such a functional profile.”16 I politely suggest that many of the researchers reviewed by Legrand and Ruby have lost themselves in the philosophical wilderness of selfhood. At the explicit representational level of episodic and imaginative narration, a distinction between self- and other-processing strikes me as entirely artificial.

In a fascinating 2011 paper “Bodily Self: An Implicit Knowledge of What Is Explicitly Unknown,” Frassinetti, Ferri, Maini, and Gallese conducted two experiments to untangle the following question: “We directly compared implicit and explicit knowledge of bodily self to test the hypothesis that bodily self-advantage, i.e., facilitation in discriminating self compared to other people’s body effectors, is the expression of an implicit body-knowledge.”17 In the first experiment, subjects were confronted with three photographs, one on top of the other, of their own and other people’s hands and feet, as well as objects belonging to them or others—mobile phones and shoes. They were asked to match the lower or upper image to the center “target” photograph. In this task a distinct self-advantage showed itself. In other words, people were considerably better at matching their own body parts than matching other people’s. No such advantage was present with the objects. In the second experiment, there was no target image, just an empty white box in the center. This time, the subject was asked which of the two remaining images was his or her own hand, foot, mobile phone, or shoe. Not only was there no self-advantage in this case; there was a self-disadvantage in recognizing one’s own body parts, one that was not seen in the recognition of one’s own objects.

The hypothesis is that an unconscious motor representation of our bodies is at work in the first task, while what the authors call “body identity” must be summoned for the explicit task. Body identity (or what Shaun Gallagher in How the Body Shapes the Mind calls body image) is a conscious, not an unconscious idea.18 It is the self perceived as an other. In the explicit task, the response is not automatic; the person has to think about it; and thinking often involves a linguistic construction as well as a visual one. Is that my foot? Is it someone else’s? The authors’ conclusion is worth quoting, “Taken together, our results show for the first time that the representation of our body-effectors is not only different from the way we represent inanimate objects, but—more importantly—it is accessible in a least two different ways: one way is implicit, while the other is detached, third-person like.”19 This unconscious/conscious distinction is paramount to understanding what neuroimagers call “self-processing.”

William James said that all personal memories have a “warmth and intimacy,” a quality of one’s own.20 To use the Latin word for selfhood or identity, my memories have ipseity. But so do my fantasies, the vicarious experiences I have while reading, my thoughts about others, my feelings about my fictional characters, and my dreams. James’s “warmth and intimacy,” that sense of ownership, is not emotionally neutral. And, as Freud stressed in The Interpretation of Dreams, however irrational or bizarre our dream plots may be, the emotions we feel are not fictional. He quotes the dream researcher Stricker: “If I am afraid of robbers in a dream, the robbers, it is true, are imaginary—but the fear is real.”21

While Proust’s tea-soaked bit of cake has become a facile reference for just about everybody, what is interesting is not that the petite madeleine opens the narrator to memories of his childhood, but rather that at first, the taste produces only feeling. “… this new sensation having had on me the effect which love has of filling me with a precious essence; or rather this precious essence was not in me, it was me.” It is only after the swell of high feeling has passed that Proust’s narrator asks himself this: “What did it mean?”22 That meaning, explored in the seven-volume first-person narrative of Remembrance of Things Past, lies in the fluctuations of subjective experience—of an emotional self in space and time. First, the narrator perceives and feels. He is immersed in the prereflective consciousness of a sensual reality that is also somehow remembering. Only later does he reflect on it, and that reflection requires that he conceive of himself as an object to himself in the same way he conceives of others. Isn’t it reasonable, then, that “self-processing” cannot be distinguished from “other-processing” at the explicit conscious level of storytelling?

The narrative self is the self in time. We are immersed in time, not clock time necessarily, although we adults refer to it, and certainly not the time of physics. We live in subjective time, the sequential time of our consciousness, and what happens before becomes the template for what we expect to happen later. Through repetition, past perceptions create future ones. In one of his 1925 lectures on phenomenological psychology, Edmund Husserl writes that “each … momentary perception is the nuclear phase of a continuity, a continuity of momentary gradated retentions on one side, and a horizon of what is coming on the other side: a horizon of protention, which is disclosed to be characterized as a constantly graded coming.”23 We are continually retaining and projecting, and the present always carries in it the thickness of before and after. Husserl, who was influenced by William James, argues that the experience of time, this perceptual stream is always pregiven in a first-person perspective. When he was five, my nephew Ty sat in the family car and made a startling discovery. Looking at the road behind him, he cried out, “That’s the past!” Turning to the road ahead, he crowed, “That’s the future!” The locus of that streaming reality of time and space, was of course, Ty himself.

We now know that a form of time, or rather, timing is also part of infancy. Psychoanalysis, attachment studies, and infant researchers, such as Daniel Stern, have been vital to our notation of what might be called the intersubjective music of early life, the preverbal melodies of the first human interactions. As John Bowlby postulated, these rhythms of attachment are crucial to affect regulation later in life. In an empirical study of adult-infant vocal interactions, Rhythms of Dialogue in Infancy, the authors proceed from a dyadic view of early communications. The nuanced analysis of the rhythmic dialectic between mother and child provides a foundation for a child’s ongoing social and cognitive experiences by forming, as the authors put it, “temporal expectancies.”24 These bodily, emotional expectations form the ground for the axis of discourse and the narrative self. An infant’s prereflective conscious perceptions are not yet for herself in an articulated story. Nevertheless, these deeply established corporeal metrics, the motor-sensory beats of self and other, merge with genetic temperament in the dynamic synaptic growth that accompanies early emotional learning. In his book Affective Neuroscience, Jaak Panksepp writes,

From a psychological perspective, I would say that the main thing that develops [in a child’s interactions with his world] in emotional development is the linking of internal affective values to new life experiences. However, in addition to the epigenetic processes related to each individual’s personal emotional experience leading to unique emotional habits and traits, there is also a spontaneous neurobiological unfolding of emotional and behavioral systems during childhood and adolescence.25

We are creatures of a subjective time founded in the wordless dialogues of infancy, which is further developed in language and its natural consequence, story. As important as the narrative self is, however, I am in complete concordance with Dan Zahavi, who writes in his book Subjectivity and Selfhood, “Is it legitimate to reduce our selfhood to that which can be narrated?” He goes on to add, “The storyteller will inevitably impose an order on the life events that they did not possess while they were lived.”26 Proust’s “precious essence,” which he claims is himself, resonates with Panksepp’s revision of Descartes’ famous cogito ergo sum to I feel therefore I am. In “The Neural Nature of the Core SELF,” Panksepp locates his core SELF in the brain stem: “… the ability to experience raw affect,” he argues, “may be an essential antecedent to foresight, planning, and thereby willful intentionality.”27 I add narrative to that list. Antonio Damasio, in his book Self Comes to Mind, discusses his protoself that produces “primordial feelings” reflecting the body’s homeostatic reality, “along the scale that ranges from pleasure to pain, and they originate at the level of the brain stem rather than the cerebral cortex. All feelings of emotion,” he continues, “are complex musical variations on primordial feelings.”28

In Instincts and Their Vicissitudes (1915) Freud proposed his own homeostatic model of primitive selfness in an organism that can discriminate between outside and inside through “the efficacy of its muscular activity.” For Freud, the regulation of internal drives and external stimuli are the origin of all feelings. “Even the most highly evolved mental apparatus,” he writes, “is automatically regulated by feelings belonging to the pleasure pain series.”29 It is out of this core feeling self that a reflectively conscious, remembering, imagining narrative self develops.

Shaun Gallagher also posits a minimal self or a “primary embodied self” already present in the motor-sensory corporeal reality of an organism that is aware of its own boundaries.30 Infant studies on imitation and deferred imitation give credence to the idea that a newborn has a greater awareness of his separateness from others and the environment than was thought earlier.31 Exactly how memory develops in babies is controversial. What effects do an immature hippocampus and forebrain and incomplete myelination have on that development? What exactly do implicit and explicit memory mean in a preverbal infant? What roles do imitation, mirroring, and language play? How do we frame the reality of infant consciousness? How is it related to a minimal or core self? When does in itself become for itself? What is the neurophysiology of time perception and how does it develop? All of these questions remain unanswered.

Narratives from the Crib, edited by Katherine Nelson, focuses on the monologues of Emily Oster taped before she went to sleep between the ages of twenty-one and thirty-six months. These soliloquies are remarkable illustrations of what Vygotsky called private speech,32 the stage before inner speech takes over. We witness the chattering play-by-play announcer who has not yet gone underground. Here is a monologue from when Emily was twenty-one months old. She is talking to her doll. I have truncated it slightly.

Baby no in night

Cause baby crying

Baby no eat supper in in in this

No eat broccoli no

So my baby have dinner

Then baby get sick

Baby eat no dinner …

Broccoli carrots cause rice

Emmy eat no dinner

Broccoli soup cause

No baby sleeping

Baby sleeping all night33

There are no fixed tenses here that situate past, present, and future, no pronominal “I.” There is a third-person baby and a third-person Emmy, characters that mingle in what might be called a protonarrative. Emily verbally represents herself as an agent to herself, and describes a series of actions in order to make sense of her emotion: the memory of not feeling well, not eating, and not being able to sleep. The third-person “Emmy” precedes the first-person “I” because reflective self-consciousness, “for-itself” reality, emerges from seeing herself as others see her, those vital others who recognize Emmy as an agent and actor in the world. In a later monologue, at twenty-eight months, the little girl imagines herself in a fictional place, the future, to master her anxiety about what lies ahead. “We are gonna at at the ocean/ ocean is a little far away … /I think it is a couple of blocks away.”34 After an associative stream that includes a fridge submerged in water and a river, the child imagines sharks biting her. The fantasy is driven by emotion, but her speech allows the flowering of creative speculation while she is still safely in her crib, away from the sharks in her mind. Emily’s monologues are heavily analyzed in the book, but two points go unmentioned, perhaps because they are too obvious: having a narrator, external and voiced or internal and silent, is a way of keeping company with one’s self. In language, the self is always touched by otherness, if only because it is represented.

Some memories have no narrator and no time except the present. In 1961, when my cousin Nette was one year old, she traveled to Africa with her parents and sister. Her father, my uncle, was a doctor who practiced in Bambuli in what was then Tanganyika. Nette learned Swahili, a language she later forgot. When she was three, she returned home to Norway with her family. Nette retained no conscious memories of Africa, but in 2007, she and her husband Mads visited Tanzania. As soon as she set foot in Bambuli, she was overwhelmed by sensations of familiarity. The smells, the colors, the sounds all contributed to a heady feeling that she had come home. One afternoon, Nette and Mads met some schoolgirls on the road, and although the two groups shared no common language, they communicated with smiles, laughter, and gestures. Mads suggested Nette hum a melody she remembered from childhood, a song the family had once sung together, the words to which had disappeared. When the girls heard the tune, they began to sing and, to her own amazement, Nette joined them. One after another, the lost Swahili lyrics returned to her, verse after verse, and Nette sang loudly and joyfully. In that moment of exuberant recall, forty-one years seemed to collapse. The forty-four-year-old woman and the small child met.

This memory is not episodic and, although I have told it as a story, the recovery of the lyrics and the flood of joy my cousin experienced is not a narrative, but a form of involuntary memory. The nineteenth-century neurologist John Hughlings Jackson called this kind of repetitive, learned knowledge automatisms. The automatism is proprioceptive, related to my bodily orientation in space, what Merleau-Ponty called a body schema, and it engages my motor-sensory capacities. The perceptual context—visual, auditory, and olfactory—acted as cues, and the once-learned but lost Swahili words came back automatically. Nette’s eruption of memory accompanied by a flood of joy has meaning in itself. Affect marks experience with valence, positive or negative, part of the pleasure-pain series. It is purely phenomenal and prereflective until we ask ourselves: What did it mean?

By far the most dramatic form of bodily prereflective, involuntary memory is the flashback. After a car accident, I had flashbacks four nights in a row that shocked me out of my sleep. Rigid, repetitious, horrifying, this memory was a visuo-motor-sensory reexperiencing of the crash. As the psychoanalysts Françoise Davoine and Jean-Max Gaudillière argue in their book History Beyond Trauma, this form of traumatic memory is outside time and language.35 It is not in the past. It is the kind of memory Augustine said nobody would want to have. In a 1993 paper, the neurobiologists van der Kolk and Saporta make the same argument. “These experiences may then be encoded on a sensorimotor level without proper localization in space and time. They therefore cannot be easily translated into symbolic language necessary for linguistic retrieval.”36 Translation into words means location in space and time; it also means distancing and, perhaps ironically, greater mutability in memory. This very mutability, however, serves the cooling and creative aspects of narration, whether in memory or in fiction.

In Beyond the Pleasure Principle, Freud cites Kant, for whom “time and space are ‘necessary forms of thought,’” and then goes on to say, “We have learnt that unconscious mental processes are in themselves ‘timeless.’ This means in the first place that they are not ordered temporally, that time does not change them in any way and that the idea of time cannot be applied to them.”37 Unlike secondary process, what Freud called primary process does not distinguish past, present, and future. We glimpse this form of archaic thought in dreams, which are more concrete, emotional, and associative than waking thought, and in Emily’s early monologues in which subjective time is not yet fully codified in language.

That creativity is mostly unconscious is hardly surprising. Psychoanalysis has long known that we are strangers to ourselves, and the idea of unconscious perception has been with us at least since Leibniz in the seventeenth century. All creativity in both modes of thought—reasoning and narrative—can be traced to this timeless dimension of human experience or, I would say, a dimension with motor-sensory timing, but not self-reflective time. In a letter to Jacques Hadamard, Albert Einstein wrote that neither language nor “any other kinds of signs which can be communicated to others” were important features of his thought. His work, he said, was the result of “associative play,” was “visual and motor” in character, and had an “emotional basis.”38 In 1915, Henri Poincaré, the great mathematician, pointed to the unconscious origins of his own work:

The subliminal self plays an important role in mathematical creation … we have seen that mathematical work is not simply mechanical, that it could not be done by a machine, however perfect. It is not merely a question of applying rules, of making the most combinations possible according to fixed laws. The combinations so obtained would be exceedingly numerous, useless and cumbersome.39

Every once in a while a formula, a poem, an essay, a novel bursts forth as in a waking dream. The poet Czeslaw Milosz once said: “Frankly all my life I have been in the power of a daimonion, and how the poems dictated by him came into being, I do not quite understand.”40 William Blake said his poem “Milton” “was written from immediate dictation … without premeditation and sometimes against my will.”41 Nietzsche described thoughts that came to him like bolts of lightning. “I never had any choice about it.”42 The last pages of my novel The Sorrows of an American were written in a trance. They seemed to write themselves. Such revelations may well be based on years of laborious living, reading, learning, and cogitating, but they come as revelations nevertheless.

A retreat to nineteenth-century science is needed to frame this creative phenomenon. F. W. H. Myers was a renowned psychical researcher and a friend of William James, who is now mostly forgotten. His magnum opus was called Human Personality and Its Survival of Bodily Death,43 a title which no doubt hastened his oblivion. Still, he was a sophisticated thinker who applied the idea of automatisms to creativity. Unlike Jackson’s habitual automatisms or the pathological dissociations of hysteria studied by Pierre Janet or Freud’s idea of sublimation, Myers argued that subliminally generated material could suddenly find its way into consciousness, and that this eruption was not necessarily the product of hysteria, neurosis, or any other mental illness.

The definition of creativity in neuroscience research I have stumbled over again and again is: “the production of something novel and useful within a given social context.”44 Useful? Was Emily Dickinson’s work considered useful? Within her given social context, her radical, blazingly innovative poems had no place. Are they useful now? This research definition must be creativity understood in the corporate terms of late capitalism. Another component of creativity featured in these studies is divergent thinking, or DT. In one study, subjects’ brains were scanned as they “produced multiple solutions to target problems.” The more solutions, the more creativity, but this is obtuse, as Poincaré pointed out so succinctly. We are not machines or computers but embodied beings guided by a vast unconscious and felt emotions.

I have often asked myself, Why tell one fictional story and not another? Theoretically, a novelist can write about anything, but she doesn’t. It is as if the fabula is already there waiting and must be laboriously unearthed or suddenly unleashed from memory. That process is not exclusively the result of so-called higher cognition; it not purely cognitive or linguistic. When I write, I see images in my mind, and I feel the rhythms of my sentences, embodied temporal expectancies, and I am guided by gut feelings of rightness and wrongness, feelings not unlike what has happened to me in psychotherapy as a patient. After my analyst’s interpretation, I have felt a jolt of recognition, which is never merely an intellectualization but always has a felt meaning: Oh my God, that’s true, and if it’s true, I have to rewrite my story.

Fictions are born of the same faculty that transmutes experience into the narratives we remember explicitly but which are formed unconsciously. Like episodic memories and dreams, fiction reinvents deeply emotional material into meaningful stories, even though in the novel, characters and plots aren’t necessarily anchored in actual events. And we do not have to be Cartesian dualists to think of imagination as a bridge between a timeless core sensorimotor affective self and the fully self-conscious, reasoning and/or narrating linguistic cultural self, rooted in the subjective-intersubjective realities of time and space. Writing fiction, creating an imaginary world, is, it seems, rather like remembering what never happened.

2010