Why Fiction May Be Twice as True as Fact

Fiction as Cognitive and Emotional Simulation

by Keith Oatley (August 2010)

Abstract. Although fiction treats themes of psychological importance, it has been excluded from psychology because it is seen as flawed empirical method. But fiction is not empirical truth. It is simulation that runs on minds of readers just as computer simulations run on computers. In any simulation coherence truths have priority over correspondences. Moreover, in the simulations of fiction, personal truths can be explored that allow readers to experience emotions — their own emotions — and understand aspects of them that are obscure, in relation to contexts in which the emotions arise.

The uses of fiction

Fiction means “something made,” even “something made-up.” As compared to “things found,” such as the data of science, people are suspicious of it. This suspicion leaks into common usage: “fiction” has come to mean “falsehood.”

The work from which has sprung much literary criticism, and some psychology of reading and writing, is Aristotle’s Poetics (circa 330 BCE). In it poetry, like fiction, meant “something made,” covering all literary works, including drama. In what follows, I use the term “fiction,” rather than “poetry,” but with the same inclusive sense. The Poetics was of a genre that has remained somewhat rare: it combined literary criticism and psychology, indeed it assumed an easy relationship between the two. In modern times psychology has become empirical science, as our textbooks remind us. As such, then, it seems to offer no serious role for fictional literature — for material that is just made up. The point scarcely seems to need arguing.

And yet . . . separation of literature from psychology has been a loss to both. My proposal is (a) that consideration of literature within psychology allows the inclusion of psychological content, not just process with which so much psychological theorizing is concerned, and (b) that fiction is not defective empirical description, but a species of simulation; just as computer simulation has augmented theories of language, perception, problem-solving, and connectionist learning, so fiction as simulation illuminates the problematics of human action and emotions.

I offer this as a new psychological theory although linking it to old sources. As a move in bringing psychology closer to literary criticism, the argument is not in the form of inference from empirical data, but of inference from literary quotation.

The content of fiction as the problematics of human agency

Fiction is about the content of what people do, and think, and feel. As Bruner (1986) has argued, narrative is that mode of thinking in which human agents with goals, conceive plans that meet vicissitudes. Not all narrative is fiction. Aristotle made the distinction: history is about the particular, about what has happened. By contrast poetry (fiction) is about the universal, about what can happen.

So we can add to Bruner’s proposal: fictional narrative is that mode of thought in which the planful agent, on meeting vicissitudes, experiences emotions. Typical fictional narrative is based on the following schema (Rumelhart, 1975): Agent with goals and a plan that typically involve other agents—>vicissitude—>emotion, which maps onto Aristotle’s proposal that a story has a beginning, a middle, and an end.

As to the content of human goals, plans, and vicissitudes, it was a discovery by Aeschylus and Sophocles that human action, no longer directed by gods, often produces effects that cannot be foreseen. When gods direct they, of course, have perfect mental models and unlimited agency. They are, in other words, omniscient and omnipotent. But when we humans act, our mental models are usually imperfect and invariably incomplete. Our agency is limited by the constraints of our embodiment. So — necessarily — we start with aspirations (goals) and from them contrive actions (plans). Our actions generate consequences we can not foresee (vicissitudes). As authors of these actions we experience consequences, and must take responsibility for them, suffering changed personal and interpersonal states (emotions) that occur.

Aristotle’s example of the perfect tragedy was Sophocles’s King Oedipus. Oedipus had come as a prince from Corinth, and saved Thebes by solving the riddle of the deadly Sphinx that menaced the city. He was received by the Thebans with joy, became their king and married Jocasta, widow of Laius the former king. At the start of the play, the city is again in affliction. Jocasta’s brother says it is caused by the pollution of harboring the killer of Laius. Oedipus forms the goal of again saving the city. He will find the murderer, and starts a plan to do so. The play is full of metaphors of seeing and blindness, of gods knowing and foretelling the future, of humans rejecting these oracles and longing to shape their own fate, thinking they know but being ignorant.

What Oedipus does not know is that he had been adopted by the King of Corinth, and that he himself killed Laius in a skirmish on his journey to Thebes. As searcher and suspect, he gradually uncovers the secrets of his own birth, of his having been put out to die, pinioned by his ankles on a hillside, by a shepherd – servant of Laius, to escape the oracle that Laius’s son would kill his father and marry his mother. As the narrative of his own guilt is pieced together, as finally the shepherd is summoned, and tells that he was unable to condemn the infant Oedipus to death, the full discovery occurs. Oedipus in frenzy, calls for a sword, and goes to find Jocasta “Where is that wife, no wife of mine.” As the revelations began to be pieced together, when she heard the shepherd was to be questioned, she recognized the terrible truth, which had been hidden from her too. The chorus recounts how she fled to her bed chamber, locking the doors, crying to Laius in despair, and hanging herself. Oedipus burst the doors.

The King saw too, and with heart-rending groans

Untied the rope, and laid her on the ground.

But worse was yet to see. Her dress was pinned

With golden brooches, which the King snatched out

And thrust, from full arm’s length, into his eyes —

Eyes that should see no longer his shame, his guilt . . .

On re-reading this play, for perhaps the fourth or fifth time, for the writing of this article, I find myself once again moved, find tears in my eyes. It is not the psychological description of a process that moves us. It is its content that gives meaning to the tragic vision of human life, and which leads to the play’s famous closing words of acceptance but with a certain kind of subdued dignity of our shared human condition.

Then learn that mortal man must always look to his ending,

And none can be called happy until that day when he carries

His happiness down to the grave in peace.

The content of fiction as including the experience of emotions

As hinted in this brief synopsis of Oedipus, the problematics of human agency are linked to emotions. The recent growth in understanding emotions will no doubt continue in psychology; my suggestion is that it will be helped if research does not become constrained by any one methodological hegemony. Let me make this argument by a comparison with the psychology of perception.

Visual perception is the best understood aspect of psychology. Apart from receiving the attention of psychology’s principal intellectual giant, Helmholtz (1866), its theories have been embodied in artificial intelligence programs of both symbolic and connectionist kinds. The study of perception allows many methodologies, and has a long tradition of demonstrations in which the enquirer experiences color mixing, geometrical illusions, apparent motion, random dot stereograms, and so forth.

By contrast, some areas of psychology, confined to behaviorism or experimentalism excluded appeal to the anchors of personal experience, and hence have been more subject to academic fashions. Instead of accumulating understanding, mapping back and forth to our culturally derived and experientially based folk-theories so that these folk theories can develop, whole sections of carefully collected psychological knowledge became, once a fashion was past, no longer of much interest to anyone.

Philip Sidney (1595) the Elizabethan poet explored the idea that a poet can provide something that a philosopher (we would now say “scientist”) does not:

A perfect picture, I say, for he [the poet] yieldeth to the powers of the mind an image of that whereof the philosopher bestoweth but a wordish description: which doth neither strike, pierce, nor possess the sight of the soul so much as that other [the poet] doth (p. 119).

We can read experiments, or tables of questionnaire responses, on emotions in the psychological literature, but when we think of experiencing and understanding an emotion, something like Sidney’s defence is needed.

…let but Sophocles bring you Ajax on a stage, killing and whipping sheep and oxen, thinking them the army of Greeks, with their chieftains Agamemnon and Menelaus, and tell me if you have not a more familiar insight into anger than finding in the Schoolmen [its] genus and difference (p. 119).

Sophocles, especially in Oedipus (from which I have offered something of the point Sidney made), was the poet to whom Aristotle devoted most attention in the Poetics. In that book, one of Aristotle’s principal psychological arguments was that in tragedy the audience is moved to experience pity and fear, which may undergo katharsis (a term that occurs just once in the book). It has been regarded as obscure, and translated as “purification” or “purgation.” Nussbaum (1986), however, says there is nothing obscure about it. The term has the same root as kathairo, and katharos, used commonly by Plato and Aristotle, with the meaning of clearing away obstacles. Katharsis could be translated as purification, but such a meaning would be the secondary one of clearing away spiritual impediments. Similarly, it could have the derived medical meaning, purgation; clearing away the obstruction of constipation. It is likely, argues Nussbaum, that what Aristotle really meant was neither of these, both of which express a Platonic distrust of the emotions and imply that the theater might be a place apart from ordinary life in which such unfortunate elements could be transformed into something better, or expelled by inner cleansing. She argues for a straighforward cognitive meaning of katharsis: “clarification.” Here is Nussbaum’s translation of the sentence in the Poetics (1449b) that includes the term: “The function of tragedy is to accomplish through pity and fear, a clarification (or illumination) concerning experiences of the pitiable and fearful kind” (p. 391).

Literature allows not just learning about emotions, but experiencing them, although in a form in which they may be clarified and better understood. The function is equivalent to visual demonstrations in the study of perception.

Aristotle’s mimesis

Mimesis is the central concept in the Poetics. Shakespeare’s version of the idea was that the purpose of drama is “to hold, as ’twere the mirror up to nature,” (Hamlet, III, ii, l. 22). With this metaphor Shakespeare continued the tradition of regarding mimesis as imitation. To Sidney, Shakespeare’s contemporary, mimesis meant “a representing, counterfeiting, or figuring forth” (p. 114), and recent translations follow this tradition.
Once again we may have been misled by such translations. In the theory of art as representation, especially in its metaphor of art as an image in a mirror, we contemplate the ephemeral and the permanent. If, although life is short, art is to be long — the artist must not only hold up the mirror but offer an image that lasts. The first experiments in photography occurred about two hundred years ago. There followed other copying technologies: film, audio-recording, video-recording, which imitate more faithfully than can any human. They leave something that endures so, unintentionally, the idea of art as a lasting image of nature was finished.

Also two hundred years ago came the romantics. Wordsworth (1802) wrote this.
Poetry is the spontaneous overflow of powerful feelings: it takes its origin from emotion recollected in tranquility: the emotion is contemplated till by a species of reaction the tranquility disappears, and an emotion, kindred to that which was before the subject of contemplation, is gradually produced, and does itself actually exist in the mind (p. 611).

So the problem of art as mimesis might have been made to go away. Among philosophers in this tradition was Collingwood (1938). He saw representation as the mark of the technical. One only needs to represent if working to a plan: representation = blueprint. In narrative this is the idea of formula, to achieve a predetermined end of arousing specific emotions in the reader, such as anxiety in the thriller, warm feelings in the romantic love story. When we see deliberate elicitation of specific emotions in other societies we call it magic, which we tend to regard as pseudo-science. But this is wrong, says Collingwood. Magic is legitimate social activity, we see it in our own society in the shaping of preference by advertisement, in a skilled speaker arousing an audience, as well as in mass-market paperbacks and movies. Such emotional — magical — effects are not pseudo-science, but pseudo art. Art proper is the creative expression of an emotion in a particular medium such as words. Because emotions are problematic they demand a creative response. Being creative can not be done to formula in relation to any representation, or to determine any specific effect —Collingwood rejects Aristotle’s Poetics as technical hints for hacks.

At first, questions were of how writing could represent the world. In the Renaissance, rhetorical issues became prominent: how the writer addresses the reader via the text. Romanticism was a shift of interest to how the writer produces what is written. Now the focus has shifted to reader-response, how the reader interprets a text.

Despite all this, the question of the relation of words to world has not gone away. Also, despite its limitations and the naggings of post-modernism, we know that science can by elaborate procedure bring words and other symbols into correspondence with things in the world. But this has left the words of fiction with no role in relation to truth.

I wish to offer two accounts which do not evade the relation of fiction to the world, but which offer a different idea than representation or imitation. One is in the humanist tradition, the other is from cognitive science. Both show how we can think of art as illuminating or clarifying both the problematics of human action, and the emotions.

Fiction as dream: a humanist metaphor

In “The Art of Fiction” Henry James (1884) spoke in favor of representation: a novel, he wrote, is “a direct impression of life.” Robert Louis Stevenson (1884) replied that it was no such thing. The novel is an abstraction. We cannot do without the idea of a perfect circle, but it exists only in the abstract realm of geometry. We cannot do without the idea of character, but we encounter it in pure form only in fiction. Stevenson went on:

Life is monstrous, infinite, illogical, abrupt and poignant; a work of art, in comparison, is neat, finite, self-contained, rational, flowing, and emasculate. Life imposes by brute energy, like inarticulate thunder; art catches the ear, among the far louder noises of experience, like an air artificially made by a discreet musician (p. 182).

In 1888, Stevenson, continuing to ponder the issue, wrote this:

The past is all of one texture — whether feigned or suffered — whether acted out in three dimensions or only witnessed in that small theatre of the brain which we keep brightly lighted all night long, after the jets are down, and darkness and sleep reign undisturbed in the remainder of the body (p. 189)”

Fascination with dreams was common for writers of the romantic era. Here is a French contemporary of James and Stevenson, Hippolite Taine (1882):

So our ordinary perception is an inward dream which happens to correspond to things outside; and, instead of saying that a hallucination is a perception that is false, we must say that perception is a hallucination that is of the truth. (p. 13, my translation, emphasis in original).

(Departing for a moment from the humanist metaphor, Taine’s theory of perception restates Helmholtz’s. It also restated in recent connectionist theory, that the mechanism of visual perception depends on a kind of constructive dreaming, Hinton et al., 1995).

Gardner (1984) agrees with the dream idea: what a novel does for a reader is to provoke a certain kind of dream. Think of it like this: close your eyes, then instead of your experience being guided by signals from the visual system, think of the novelist speaking softly into your ear, providing for you purely mental materials to start up, construct, and sustain the dream.

What human minds do generally is to make models that parallel the workings of the world (Craik, 1943). Our language is thick with ideas of this function: simile, metaphor, schema, parallel, analogy, theory, hypothesis, explanation, model, and so on. The dream is just one example of the constructive machinery running to produce events that parallel those of the world. It is a metaphor of the idea of fiction as imagination.
Abrams (1953) has proposed that we think of the writer of fiction as drawing attention to something. The metaphor of the dream of imagination shows how attention is indeed directed. A text guides us to dream this dream, and not some other one. Abrams suggests that for the mirror we substitute the metaphor of the lamp that illuminates, and clarifies. The lamp is that which lights the “small theatre of the brain.”

Fiction as simulation: a metaphor from cognitive science

We can counterpoise to James’s idea of the novel as “a direct impression of life” another metaphor that was not available until recently (Oatley, 1992): a play runs on the minds of the audience as a computer simulation runs on a computer. With the metaphor of mimesis as simulation, however anachronistic it may seem, we can re-read Poetics and see that something like this was what Aristotle really meant. He says correspondences are inessential; great drama is often in poetic verse, although no-one ever spoke verse in real life. Aristotle asks what is it that makes a play moving? It is the plot which is “. . . the heart and soul, as it were, of tragedy . . . [it] simulates persons primarily for the sake of their action” (1450b 2-3, my translation, emphasized, of mimesis). What tragic drama achieves is a focus on essentials of human action. Non-essentials are excluded. As Aristotle put it: “A poetic mimesis, then, ought to be . . . unified and complete, and the component events ought to be so firmly compacted that if any one of them is shifted to another place, or removed, the whole is loosened up and dislocated” (1451a, 30-36).

Metaphors draw attention to some facet of direct experience and carry it over to something new, or to something difficult to understand. Dreaming is familiar to most people, programming computers only to a few, so as a metaphor, simulation works well only for a minority. I wish nonetheless to press it further.

If art were representation, then literary art would simply figure forth a world and actions within it. But since the Russian formalists, two aspects of narrative have been recognized, the fabula (material or story) and the siuzhet (the aesthetic working of it or plot). Following Brewer and Lichtenstein (1981) let us call these the event-structure and the discourse structure. So, the writer must:

1. Provide an event structure — a world model that does have aspects of representation (semantics), and which must be self-contained.

2. Guide the reader, by means of a discourse structure that includes speech acts (pragmatics) and cues to the reader which are like the operations which by which a computer program is controlled, cues that make the simulation run.

Look at these two streams of information in this fragment of a program I wrote, following Sharples et al., (1989) in the language Pop 11, as a prototype for students to augment in an artificial intelligence course. The program simulates a conversation partner that can answer questions about the best route from anywhere to anywhere else in the downtown Toronto public transit system. In this fragment, two variables are declared by the command “vars.” Then these variables (“travtime” and “changetime”) are given values. Then comes the start of the declaration of a procedure “setup (),” built on a database which is a list of lists, of subway and streetcar stations in Toronto, each joined by “connects” to indicate which station directly connects with which other.

vars travtime, changetime;

2 -> travtime; 5 -> changetime;

define setup ();

[[[BloorSubway spadina] connects [BloorSubway st george]]

[[BloorSubway st george] connects [BloorSubway bay]]

[[BloorSubway bay] connects [BloorSubway bloor yonge]] etc., etc.

In this simulation one stream of information is a mode that exhibits correspondences with the world. Part of it can be seen in the data-base over which the program computes within the procedure “setup ().” There are names of lines (e.g. “BloorSubway”) and stations (e.g. “st george”) that correspond to names of lines and stations in the real Toronto. But the other stream of information has its reality only in the computer world. It activates processes to make the simulation run: declaring variables, the operator for assignment ->, defining a procedure, the list of lists, etc.

For a writer of fiction it would be no good simply offering something like “event+event+event,” you must also offer cues to running the simulation. If, for instance, you write a suspense story, there are three basic elements: (i) get the reader attached to likeable protagonist, then (ii) create a believable threat to this character from, (iii) some person or agency to whom the reader will feel antagonistic (Vorderer, 1996). At the start you may make a few pages cover weeks or years of event structure. Once the threat is established, you can devote many pages to each hour of event structure. As this supenseful version of the simulation runs, the reader feels duly anxious, and turns and turns the pages until he or she can feel the relief of safety again.

The writer invites the reader to enter, Alice-like, through the looking glass, into the story world. This is like entering a particular kind of social interaction, as Goffman (1961) has shown, through a semi-permeable membrane, inside which is a world with its own history, and its own conventions from which meanings are constructed.

Revisiting the idea of empirical truth as faithful correspondence

Equipped with the idea of simulation, let us consider more closely why a faithful copy such as an audio recording of a conversation would never be part of literature (well hardly ever). This section will also show what is missing from correspondences, and what must be provided to make them comprehensible to us humans.

Here is a transcript of part of a purposeful conversation. A complex new photocopier had been delivered to a Psychology Department so two colleagues and I (Oatley, Button and Draper, in preparation) took the opportunity to record some conversations. Our only input was to give a person (who plays the role of Xavier) the goal of finding someone (whom we call Yolande) to show him how to use the copier. Apart from inserting this goal into his mind, everything else flowed naturally, nothing was contrived.

Xavier:           (1) Could you show me how to do the photocopying?
Yolande:        (2) Double sided?
Xavier:           (3) Eh. Yer. I want to do, to do double sided.
Yolande:        (4) Uhm, I don’t know, some or . . .
Xavier:           (5) Sorry, what do you do here? [Points to buttons.]
Yolande:        (6) This one. [Selects button “Duplex 2.”] But, eh, some turn the                                          other way round. You must have it.

Writers avoid imitating all aspects of such conversations: “Eh. Yer . . . uhm,” needless repetitions (“to do, to do”), and lapses of grammar. Note, moreover, that after the first three utterances, this real conversation is largely incomprehensible. Let me write it as it might occur in a novel. I have preserved as many words from the transcription as I could, and I have made each paragraph correspond to each utterance in the transcript.

(1) “Could you show me how to do the photocopying?” asked Xavier.

(2) Yolande knew that Xavier must be able to do straightforward copying, he must want to do something more, perhaps learn advanced features of this irritating machine, recently delivered to the Psychology Department, which she had spent a good deal of time learning how to operate. Xavier only had one piece of paper: maybe he wanted her to use it to show how the machine did different kinds of copying. “You want to do double-sided?” she asked.

(3) “Yes, I want to do double sided,” said Xavier.

(4) “It’s more complicated than you might think.”

(5) “Sorry. What do you do here?” asked Xavier. He wanted to get started.

(6) “You press this button, but usually you have to think about how many copies you want, and whether you have got single or double-sided originals, and sometimes you have to worry about whether the copy on the second side will come out the right way round.”

Notice how the simulated (novelistic) version does not really imitate. Instead I showed — as Aristotle says — the essentials of human action, which included characters’ goals and interpretations. Also I had to create a discourse structure that would help you, the reader, construct a working event model.

Recently Scheff (1997) has proposed a new methodology for the social sciences. He argues that records such as transcripts of conversation need to be studied in great detail as biologists study single specimens. The “specimens” are interpreted under the illumination of theories up to the widest level of generality. The process reminds one of literary criticism, in which a text is given context that enables it to be better understood. In two chapters of his book, indeed, Scheff works with literary examples, a Shakespeare play and Jane Austen’s novels. He works in comparable way with transcripts of conversations of children, and with telegrams between national leaders during the preliminaries to World War I. We come to understand the conversations as negotiations in the sustaining of self — with its possibilities of intimacy and the threat of shame. Scheff’s interpretations offer another equivalence to the novelist’s guidance about how to understand events in the story world. They are equivalent too, to the stream of information in programs that guides the computer in running a simulation.

So, not only do many imitations require explication, but fiction does not really imitate. It does not even copy conversation, which many assert to be the form of life most indigenous to fiction. To give a further example, here is one of the world’s greatest novelists, Tolstoy, in some conversation from the novel that many consider his greatest, Anna Karenina. We are in the novel’s central moments: Anna and her husband are watching a cavalry officers’ race when a rider falls.

The officer brought the news that the rider was unhurt but that the horse had broken its back.

On hearing this Anna quickly sat down and hid her face behind her fan. Karenin saw that she was crying, and that she was unable to keep back either her tears or her sobs that were making her bosom heave. He stepped forward so as to screen her, giving her time to recover.

“For the third time I offer you my arm,” he said . . . (p. 210)

Then, in the carriage on the way home:

. . . “Perhaps I was mistaken,” said he. “In that case I beg your pardon.”

“No, you were not mistaken,” she said slowly, looking despairingly into his cold face. “You were not mistaken. I was, and can not help being, in despair. I listen to you, but I am thinking of him. I love him, I am his mistress, I cannot endure you; I am afraid of you, and I hate you” (p. 212).

What is wrong with this? Empirically, almost everything. It presupposes an observer, either with supernatural qualities or with a tape recorder hidden in the carriage. In any case the whole passage is too stagey; people don’t talk like this. The whole thing is made up. It is obviously better written than ordinary people could manage, but it has scant relation to real conversation.

If you want to know the correspondence-truth of what goes on between married people in conflict, you bring a proper sample of married couples into the laboratory, as Gottman and Levenson (1992) have done. You have each person complete psychological instruments to measure marital satisfaction. You wire them up to polygraphs, and videotape them to record speech and facial expressions. You ask them to talk with each other, first about a neutral subject, then about a topic of conflict between them. To obtain records of subjective emotions, you show each participant the video next day, and have the participant indicate what he or she was feeling at each point.

The point is this: many, perhaps most, transcriptions would not be intelligible. There are two routes. Either transcription must be accompanied by interpretative analysis (e.g. of the kind that Gottman and Levenson offer) so that the reader can understand it. Or, if utterances are to understood by the reader at the time of reading, they must be altered — fictionalized. They may then become a simulation in which meanings can become comprehensible. Fiction provides context to understand the elliptical. It offers the context of goals and a direction of plans present in the minds of interactants. It gives a sense of how actions lead to vicissitudes. It allows, too, the reader to experience something of emotions that can arise. All these are omitted from a faithful, empirically unexceptionable, recording of real life.

In fiction based on simulation it is important to have some imitative correspondence with the real world; some aspects are represented in the event-model, just as the names of subway stations were represented in my simulated conversation partner. Without such correspondence fiction founders. But piece-by-piece correspondence is in the service of coherence within the larger structure, and of the relationship with the reader.

Dream and simulation as alternative metaphors of construction

Dream and simulation are alternative metaphors; one rather than the other is likely to be preferred by each reader. The essence of both is that mental life is constructive: the machinery of construction is within us.

For the literary experience of fiction, Japanese Bunraku (Inoura & Kawatake, 1981) is instructive. In Bunraku the characters are puppets nearly two-thirds life-size. Each is worked by three puppeteers who appear on the stage dressed in black cowls. The chief puppeteer holds the puppet’s torso; he manipulates its right arm and expressions of its face. One assistant manipulates its left arm, and the other its legs. The words of the play are poetry read by a narrator who sitting in full sight at the side of the stage with an open book. For these dramas stage settings are colourful and striking, and the plays are accompanied by music. Japan experienced a humanistic renaissance comparable with the European renaissance, and Bunraku was a important in it, with plays being written by great poets, including Chikamatsu, “the Japanese Shakespeare.”

The elements of this drama — the actions of puppets, the partly hidden forces that control them, the metrical poetry read by a narrator, and the music — could all seem disparate, formal devices. But the plays treat themes of power, revenge, and adultery, and were popular as Japan emerged from feudalism. The audiences integrate the elements, creating for themselves a moving experience.

Lest this idea seems strained, note that film, that seemingly most realistic form is shot often with several cameras in the same scene, some of which zoom, or track smoothly on rails, others of which can be located high in the air, with the shots edited together to form sequences that no single live human observer could ever experience. The editing is designed not to mimic what a real human observer might ever see, but to offer the impression of being in exactly the right place, at the right moment, to follow a plot. The staging and editing are the equivalent of the discourse structure of a literary text, the housekeeping of a program that allows the simulation to run.

Three theories of truth

The idea of simulation allows one to ask whether there is more than one sense of the term “truth.” Three senses, relevant to this argument, are depicted in Table 1.

Table 1. Types of truth, methods to approach them, and criteria for their recognition

Type                      Method         Criteria
Correspondence Empirical     Valid discovery of facts, reliability, practical usefulness
Coherence             Simulation   Produces emergent effects without ad hoc intervention
Personal                Reflection     Recognition, insight

Empirical psychology obeys criteria of the first type of truth. Fiction fails this criterion, but fulfils the other two. One could say, then, that fiction is twice a true as fact.

Truth as correspondence

In terms of science the methods of fiction cannot escape the sampling problems of any single, unsystematic, biased, observer. Empirical truth means bringing descriptive statements into valid and reliable correspondence with specified aspects of the world, for purposes that include generalization and useful technology.

The epistemological debate as to how to discover truth as correspondence derives from the debate between Plato and Aristotle — Plato won, and scientists are his descendents. To find truth as correspondence you must join a special society. Plato called it the Academy, nowadays the University. You must learn mathematics. You have to devote yourself to empirical truth, the methods of which are demanding.

Truth as coherence

If empirical methods prioritize correspondence, simulation prioritizes coherence. If fiction is simulation, it is not to be dismissed as “unreliable description.” Its value is in exploring interactions, to see what relationships allow a system to produce emergent properties from the interactions, which the parts alone can not produce. The sections on dream and simulation have been expositions of this idea.

In a simulation one includes only elements essential to the phenomena of interest, and sets them to interact. In a computer simulation of global warming (environmental warnings of this phenomenon all derive from simulations) the causal process of the production and absorption into biomass of carbon dioxide will be included but share prices on the New York Stock Exchange will not.

Personal truth

The idea of personal truth is generally discussed in terms of insight, and in the idea that a piece of fiction can be profound. So, a literary work can be truthful, not just generally but in relation to a specific reader and to that reader’s own understandings of self and others. In literary criticism one term is “the sublime.”

Longinus (1st Century CE) introduced the idea that some expressions or images seem to come from some realm beyond the merely human. They are sublime thoughts on fundamental questions, they involve an intensity of emotion, and they provide a permanent focus for reflection. Let us consider an illustration from Shakespeare — his most famous words. Compare two versions, starting with an un-sublime version, which comes from the First Quarto, probably pirated by the actor who played Marcellus in one of Shakespeare’s company’s productions. The pirate was accurate with his own lines, but elsewhere ran aground on the shoals of the ordinary.

To be or not to be, ay there’s the point,

To die, to sleep, is that all? Ay all:

No, to sleep, to dream, ay marry there it goes

For in that dream of death when we awake

And borne before an everlasting judge . . .

Now here is Shakespeare’s version from the collated Second Quarto and First Folio editions, thought to be authentic.

To be or not to be, that is the question:

Whether ’tis nobler in the mind to suffer

The slings and arrows of outrageous fortune,

Or to take arms against a sea of troubles,

And by opposing end them. (Hamlet, III, i, 56-60)

A poet who occasionally reached the sublime, and also wrote about it in his letters, was Keats. This was to John Taylor, dated 27 February 1818:

. . . In Poetry I have a few Axioms, and you will see how far I am from their Centre. 1^stI think Poetry should surprise by a fine excess and not by Singularity — it should strike the Reader as a wording of his own highest thoughts, and appear almost a Remembrance — 2^nd Its touches of Beauty should never be half way ther[e] by making the reader breathless instead of content: the rise, the progress, the setting of the imagery should like the Sun come natural to him — shine over him and set soberly although in magnificence leaving him in the Luxury of twilight . . .

Keats’s idea of “a fine excess” rather than “Singularity” (oddness), seems so apt, right too for mimesis as simulation: one simulates essentials, thus exaggerating them slightly at the expense of inessentials, for the sake of understanding. Just as striking is the idea that poetry should be a “wording of [the reader’s] own highest thoughts, and appear almost a Remembrance.” This was not a new idea in criticism, but Keats put it with such telling simplicity that some part of oneself leaps to meet it.

Emotions

Emotion was an aspect of Longinus’s theory as it was of Aristotle’s; being emotionally moved is important in responding to fiction. If fictional narrative is about human intention, its vicissitudes, and its emotions, these emotions seem to have been central from the beginning of literature, perhaps because many of them — our own and those of others — seem either inchoate or just beyond the edge of understanding.

I propose that the experience of emotions when reading is the aspect that shows most fully why the idea of fiction as simulation is productive.

The part of the mind on which the reader (or audience) runs the simulation of a novel, play, or film, is the planning processor. Ordinarily, we use it in conjunction with our mental models of the world to assemble actions into plans to attain goals, for instance to pick up some milk on the way home from work. By contrast, in reading a story the plot then takes over the planning processor. We tend to identify with the protagonist, adopt his or her goals, take on his or her plans. Then — straightforwardly in the terms of this theory — we experience emotions as events and outcomes of actions are evaluated in relation to the protagonist’s goals. But here is the extraordinary feature: although the goals and plans are simulated, the emotions are not. They are the reader’s own emotions. The fictional narrative allows us to experience emotions and to understand them more clearly than in real life. Understanding may be increased, just as it is by interpretations of the kind offered by Scheff in the discourse segments he treats. The commentaries that have been offered of canonical works by literary critics offer possiblities of yet further understanding.

As one enters through the semi-permeable membrane of the text, there are at least three different possible ideas of how emotions are produced, which can be seen as three modalities of emotional response. One is of identification, based on the simulation theory as described here; emotions arise as they do in real life, but here they arise not from the reader own life, but from evaluations of story-events in relation to the story-protagonists’ goals, inserted into the simulation which runs on the reader’s planning processor. A second is the idea of Tan (1994), that the writer describes patterns of events of the kind that cause emotions. From these, the reader attributes emotions to story characters, and the reader experiences sympathetic emotions towards these characters. A third idea is that of Scheff (1979) that narrative and drama allow the re-evocation of emotional memories from the autobiographical past. In ordinary life, Scheff argues, emotions can sometimes be too close, too overwhelming, so that their meaning is not recognized. In other instances, we distance ourselves from emotions, and suppress them. In either case the effects of such experiences can leave us with an emotional arrears, because important emotional events have occurred which we have not assimilated. Drama or novels allow evocation of such emotions, neither underdistanced nor overdistanced, but at what Scheff calls an optimal aesthetic distance at which they can be understood and assimilated.

Why should emotions be problematic? In part, the reasons are those described by Scheff, as mentioned in the previous paragraph, but the matter can be put more generally. In a study of subjects asked to keep diaries of incidents of emotion, some of which were followed up by interviews, Oatley and Duncan (1992) found that some six percent of emotions occurred without the subject knowing why, and (in different samples) between five and twenty-five percent of emotion incidents had aspects subjects did not understand. If we fall in love — why with that particular person? If we feel grief for someone lost, why do we sometimes feel angry as well?
Here is a more specific example. A subject in one of Oatley & Duncan’s diary studies was a nurse who worked by making home visits. One day, she felt an overwhelming sense of disgust as a patient cleared his throat. So strong was her disgust that she had to interrupt what she was doing, and her emotion made it difficult for her to perform the procedures. Thinking about her emotion afterwards, it made her wonder about whether she should give up her job. When interviewed, she said the man’s clearing his throat reminded her of her ex-husband who used to clear his throat in a similar way, but that she could not understand why the patient who was an inoffensive person, made her feel like this, or why the feeling was so strong.

So here we have the beginnings of a story of a woman, and her life, and of something to do with an emotion that has an element of the mysterious about it.

Emotions happen when events occur that relate to our goals (see the schema for stories offered above) — i.e. to those concerns that are important to us. So an emotion not understood means that something has happened with an obscure implication. Emotions are at the centre of literature because they signal what is personally important, and because they often have this element of the mysterious. The simulations which are novels, plays, movies and so forth, can allow us to explore emotions and find out more about the intimate implications of these elements in our lives.

How to write and read fiction

There is a genre of “How-To” books on writing fiction. They recommend that the writer should provoke flashes of life-like, visualizable, imagery. “Show, don’t tell” is the slogan. Anything else, they say, is dull stuff, and editors will reject it.

In the light of the idea of literary simulation as a kind of dream, one can see the point. But on considering it more closely it is sad advice, an admission that reading is not as good as watching a movie. Not only do the great writers not follow it, but psychologically, within the simulation theory, it misses the most important insight. What a writer does, is to provide cues for the simulation to run — cues that call up identifications, or indicate patterns of events so that we can feel sympathetic towards some character’s emotional plight, or prompt memories (consciously or unconsciously), or summon fragments of cultural understanding. All these allow inferences that fill the gaps in what the author offers, to make the simulation complete, to sustain the dream. When the simulation works well, our experience becomes a seamless whole.

The devices invented by writers of fiction are ways of programming our simulations. Each was invented, and put into use in some particular piece of narrative, then noticed, then further developed by subsequent writers. Here is a brief selection.

Conversation in fiction was an ancient invention, already present in the oldest stories, such as the 4000-year-old Sumerian stories of Gilgamesh. Later came the invention of the idea of personality, a coherent set of goals of a character, from which actions would flow: this is the accomplishment of J, the author of the first books of the bible. Later came the idea of action limited by human mental models — an invention of Aeschylus and Sophocles (discussed above). Later yet is the depiction of a character’s action as including conscious thinking, by which the character is affected. This idea too was an invention — of Shakespeare in the soliloquies. It was continued by George Eliot. Later, Henry James explained and used “point of view” an essential idea to any present-day writing. In the modern novel is the technique of “stream of consciousness” a phrase coined by Henry James’s brother, William, and introduced into fiction by Dorothy Richardson and James Joyce. These, and many other techniques allow certain kinds of simulation to run.

If you are a writer of fiction you equip yourself with a repertoire of such techniques, just as when, if you write a computer simulation, you learn the repertoire of ideas for accomplishing certain computational effects: procedures, list processing, recursion, pattern-matching, connectionist learning, and so forth. It is quite wrong, said T.S. Eliot (1919) discussing tradition in poetry, to think that one inherits a tradition: “It cannot be inherited, and if you want it you must obtain it by great labour. It involves, in the first place, the historical sense . . .” (p. 23).

“Novel” means something new, and this is perhaps why the novelist too needs the historical sense. But the invention of new techniques and genres is all part of the same attempt: to start up and sustain the reader’s dream-simulation in ways that enable particular novelistic purposes. So if a writer believes that plot makes it too easy for the reader simply to swallow up the pages, then the technique of plot is replaced. If the novel seems too monologic, then one creates a simulation with more than one centre of consciousness, and so forth.

Moreover, as Wolfe (1975) has shown, fictional techniques are so useful that they have been adopted into non-fiction. He names four: (a) scene-by-scene construction and avoidance of sequential description, (b) dialogue, (c) point-of-view, and (d) “the recording of everyday gestures, habits, customs, manners . . . symbolic, generally, of people’s status life,” (p. 47). Wolfe suggests that written text, being an indirect medium, relies more on prompting readers’ memories than in creating direct images.
On the one hand, then, narrative is the mode of communicated thought that slips most easily into the human mind, and is most easily apprehended (Turner, 1996). But on the other hand the simulation theory suggests that just as artificial intelligence programmers develop new techniquess, so do fiction writers. So not all fiction slips easily into the mind. Techniques need also to be recognized and understood by readers, else the simulation will not run for them.

For the literary novel, we arrive at the idea that not just writing but reading can be creative. Running a simulation on the mind can involve what Barthes (1975) called a “writerly reading,” in which the reader constructs his or her own version of the narrative. In such a reading, the modes of emotion, the effects of various techniques, each afforded by the writer, enable the reader’s interpretation to cross among modes of understanding, to create a reading uniquely personal although afforded by the text. Barthes proposed a range of interpretive modes, which move in and out of perspective as a reading continues: the hermeneutic, the thematic, the symbolic, the action-related, and the cultural. Or one might prefer the alternative set, proposed long ago by Dante (e.g. in the Convivo) in which he describes not just several modes, but progressively deepening interpretation as the reader engages with the text: first is the literal interpretation, then the allegorical which generalizes to a more profound plane, then comes the moral in which one’s own schemas of action are affected, then the anagogical or spiritual, in which human beings can glimpse (in Dante’s thought) grace, and the eternal.

In any event, readers who are prepared to use a fictional book to start up their dream-simulation can become writers of the understandings that they create as they read. The author will have constructed the text to be productive of certain truths of coherence and juxtaposition. The simulation will run if the reader can achieve a certain correspondence of mind to the writer. Although the text offers constraints, for literary fiction each reading is different because in the construction of a writerly reading it is personal meanings that emerge as one immerses oneself in the simulation.

For psychology these meanings center on the emotions prompted by the text, but which are the readers’ own — experienced and potentially better understood in the context provided by the writer than in some of the contexts that occur in ordinary life. In this way, for psychology, emotions prompted by fiction can take on roles like those of visual demonstrations in the study of perception, of being able to map between folk-theoretical and natural-scientific understandings. But in addition, because emotions are always personal, their meanings can be, for the reader, profound in the way that Longinus describes: in the constructed understandings, the reader’s emotions, and the reader too, can be transformed.

(This article is based on an address I gave as retiring Executive Officer of the International Society for Research on Emotions, at the Society’s Ninth Meeting in Toronto, 13-17 August, 1996. The writing and research for this article were supported by the Social Science and Humanities Research Council of Canada, whom I thank warmly.)

References:

Abrams, M. H. (1953). The mirror and the lamp: romantic theory and the critical tradition. Oxford: Oxford University Press.
Aristotle (c. 330 BC). Poetics (G.E. Else, Ed. & trans. 1970) Ann Arbor: University of Michigan Press.
Barthes, R. (1975). S / Z (R. Miller, Trans.). London: Cape.
Brewer, W. F., & Lichtenstein, E. H. (1981). Event schemas, story schemas and story grammars. In J. Long &. A. Baddeley (Eds.), Attention and Performance 9 (pp. 363-379). Hillsdale, NJ: Erlbaum.
Bruner, J. (1986). Actual minds, possible worlds. Cambridge: Harvard University Press.
Collingwood, R. G. (1938). The principles of art. Oxford: Oxford University Press.
Craik, K. (1943). The nature of explanation. Cambridge: Cambridge University Press.
Dante Alighieri (1304-1308) Il Convivio (The Banquet), Ed. & Trans. C. Ryan, (1989) Saratoga, CA: Anma Libri.
Eliot, T. S. (1919). Tradition and the individual talent. In J. Hayward (Ed., 1953), T.S. Eliot: Selected prose (pp. 21-30). Harmondsworth: Penguin.
Gardner, J. (1984). The art of fiction. New York: Knopf, reissued Random House, 1991.
Goffman, E. (1961). Encounters: Two studies in the sociology of interaction. Indianapolis: Bobbs-Merrill.
Gottman, J. M., & Levenson, R. W. (1992). Marital processes predictive of later dissolution: Behavior, physiology and health. Journal of Personality and Social Psychology, 63, 221-133.
Helmholtz, H. (1866). Treatise on physiological optics, Vol 3. (Ed. J.P.C. Southall, 1962) New York: Dover.
Hinton, G. E., Dayan, P., Frey, B. J., & Neal, R. M. (1995). The wake-sleep algorithm for unsupervised neural networks. Science, 268, 1158-1161.
Inoura, Y., & Kawatake, T. (1981). Traditional theater of Japan. New York: Weatherall.
James, H. (1884). The art of fiction. Longman’s Magazine, September. Reprinted in M.D. Zabel, (Ed., 1951) The Portable Henry James, (pp. 391-418). New York: Viking.
Keats, J. (1918) Letter to John Taylor. In R. Gittings (Ed., 1966). Selected poems and lettes of Keats, (p. 46). London: Heineman.
Longinus (1st Century CE). On the sublime. In T. S. Dorsch (Ed., 1965), Aristotle, Horace, Longinus: Classical literary criticism. (pp. 99-158) Harmondsworth: Penguin.
Nussbaum, M. C. (1986). The fragility of goodness: Luck and ethics in Greek tragedy and philosophy. Cambridge: Cambridge University Press.
Oatley, K. (1992) Best laid schemes: The psychology of emotions. New York: Cambridge University Press.
Oatley, K., & Duncan, E. (1992). Incidents of emotion in daily life. In K. T. Strongman (Ed.), International Review of Studies on Emotion (pp. 250-293). Chichester: Wiley.
Rumelhart, D. E. (1975). Notes on a schema for stories. In D. Bobrow &. A. Collins (Eds.) Representation and understanding: Studies in cognitive science New York: Academic Press.
Scheff, T. J. (1979). Catharsis in healing, ritual, and drama. Berkeley: University of California Press.
Scheff, T. J. (1997). Emotions, the social bond, and human reality: part/whole analysis. New York: Cambridge University Press.
Shakespeare, W. (1600). Hamlet. London: Methuen (current edition 1981).
Sharples, M., Hogg, D., Hutchison, C., Torrance, S., & Young, D. (1989). Computers and thought: A practical introduction to artificial intelligence. Cambridge, MA: MIT Press.
Sidney, P. (1595). An apology for poetry. In C. Kaplan (Ed., 1986), Criticism: The major statements, 2nd Edition. (pp. 108-147). New York: St Martin’s.
Sophocles (496-406 BCE). King Oedipus. In E. F. Watling (Ed., 1947), The Theban Plays (pp. 23-68). Harmondsworth: Penguin.
Stevenson, R. L. (1884). A humble remonstrance. Longman’s Magazine, December, Reprinted in C. Harman (Ed., 1992). R.L. Stevenson: Essays and Poems (pp. 179-188). London: Dent Everyman’s Library.
Stevenson, R. L. (1888). A chapter on dreams. Scribner’s Magazine, January. Reprinted in C. Harman (Ed., 1992) R.L. Stevenson: Essays and Poems (pp. 189-199). London: Dent Everyman’s Library.
Taine, H. (1882). De l’intelligence, Tome 2. Paris: Hachette.
Tan, E. S.-H. (1994). Film-induced affect as a witness emotion. Poetics, 23, 7-32.
Turner, M. (1996). The literary mind. New York: Oxford University Press.
Tolstoy, L. (1877). Anna Karenina. (C. Garnett, Trans., 1901). London: Heinemann.
Vorderer, P. (1996). Towards a psychological theory of suspense. In P. Vorderer, H. J. Wulff, & M. Friedrichsen (Eds.), Suspense: Conceptualizations, theoretical analyses, and empirical explorations (pp. 233-254). Mahwah, NJ: Erlbaum.
Wolfe, T. (1975). The new journalism. In T. Wolfe & E. W. Johnson (Eds.), The new journalism (pp. 13-68). London: Picador.
Wordsworth, W. (1802). Preface to Lyrical Ballads of 1802. In S. Gill (Ed., 1984), William Wordsworth (pp. 595-615). Oxford: Oxford University Press.

Dr. Oatley is Professor Emeritus, Department of Human Development & Applied Psychology, University of Toronto.