About CA
Contact CA

Search Journal

Editorial Board

Permission to Reprint


Site Map


Harnessing the Imagination:
the Asymmetry of Belief and Make-Believe

  Elisa Galgut

“For many things which if they happened in real life could produce no pleasure can nevertheless give enjoyment in a play.” 

Freud, “The Relation of the Poet to Day-Dreaming”

Contemporary philosophical discussion on the nature of the imagination has been influenced by recent empirical work in cognitive science.  Our imaginative and emotional engagement with works of fiction has been explained by appealing to the similarities between our ordinary cognitive functioning and the workings of our imagination. Believing and imagining, it is argued, are governed by a “single code.”  I argue against this claim, and suggest that our imagination – and in particular our literary imagination – in many respects functions very differently from ordinary cognition.

Key words
emotional engagement with fictions, make-believe, pretend play, simulation theory


1. Simulation theory and the single code hypothesis 

Contemporary philosophical discussion on the nature of the imagination has been influenced, in both direction and substance, by recent empirical work in certain branches of psychology and cognitive science.  Accounts of our emotional involvement with fictional characters, such as empathy with and sympathy for the likes of Anna Karenina, have been explained as appealing to the similarities between our ordinary cognitive functioning and the workings of our imagination.  Believing and imagining, it is argued, are governed by a “single code.”  This claim is central to simulation theory; its proponents explain that we react to fictional stories and their characters in ways that parallel ordinary cognitive or emotional behavior.[1] 

We weep for Anna Karenina, fear for Othello at the hands of Iago, and pity King Lear, despite knowing that none of these characters actually exists.  Simulation theory purports to explain how we are able to engage imaginatively with the mental states of others; it is a theory of mind reading.  Not surprisingly, given its roots in cognitive science, simulation theory models the imagination on that of ordinary belief acquisition.  Imaginative states are referred to as “pretend beliefs” or “quasi beliefs” and, notwithstanding central differences between pretend- or quasi-beliefs and real beliefs, central structural similarities are preserved.  "The basic idea of the off-line simulation theory of behavior prediction," writes Stich, Nichols, et al, "is that the practical reasoning component is taken off line and used for predicting behavior."[2]

Confirmation of this view regarding the relationship between imagining and believing is found in some recent papers by Shaun Nichols, who argues that there are “striking similarities between imagination and belief” that are “partly explained by the hypothesis that imagination and belief are in a ‘single code.’”[3]  Moreover, Nichols maintains that the intimacy of the relationship between imagining and believing is attested to by mainstream cognitive science:  “the single code theory, in one form or another, is embraced by most contemporary cognitivist accounts of pretence and imagination.”[4]  The account developed by Nichols and others has been shown to be powerful in its scope of explanation and prediction; it covers pretend play in children as well as cases of various kinds of imaginings in adults. 

Moreover, Nichols goes on to show how the single-code hypothesis can usefully explain many puzzles about our engagement with fictions.  For instance, my pity for Lear is explained by the analogous functioning between my belief system and my pretend or “make-belief “system, with the difference being that the latter is run  “off-line”:  “Since the ‘pity’ response would be activated by the belief, it [the make-believe pity] is activated by the isomorphic pretence representation.”[5]  The off-line metaphor, and it is important to note that it is a metaphor, is meant to capture this isomorphism between believing and make-believing. 

Differences are then explained by appealing to differences of input between the two systems (beliefs vs. make-beliefs), which may, in turn, generate differences in output.  The account is essentially a functionalist one.   The single-code hypothesis is also invoked to explain the functioning of proper names in fictions:  “[O]ur intuitions about names of fictional characters parallel in a striking way our intuitions about names of real individuals with respect to uniqueness and reference;” tacit beliefs, such as the belief that “Gatsby … has fewer than one billion hairs on his head,”[6] and certain kinds of imaginative resistance, such as our unwillingness to imagine in a fiction that certain things, like murder, are morally acceptable.  Our moral code seems to hold constant across both fictional and real worlds because imagining that an action is immoral parallels our believing that it is.

2. Asymmetries

In a later paper,[7] Nichols raises a problem with the single-code hypothesis:  it fails to account for certain asymmetries between our responses to some imaginary situations and reality.  In other words, there appears to be some asymmetries between our imaginative responses to fictional situations and our ordinary responses to the real-life counterparts of those situations.  For example, my feeling of pity for Lear in relation to the death of Cordelia is not felt as deeply or as consistently as my pity for a personal friend whose daughter has died.  Nor does it lead me to action:  I do not attempt to comfort Lear in any way or wonder, after watching a performance of the play, whether I should visit Cordelia’s grave. 

There are also deeper asymmetries, what Nichols calls “discrepant affect,” such as the laughter evoked by black comedies towards events that would be horrific were they to occur in real life.  In real life, we would not find the murder of a shop assistant funny, yet it is precisely this situation depicted in an episode of Monty Python’s Flying Circus that does bring us to laughter.  Much of the work of film director Quentin Tarantino relies on this asymmetry in affective response towards fictional and real-life scenarios by offering us situations that we feel should offend or revolt us but instead make us laugh.  Notwithstanding the arguments from Gendler, Walton, and others[8] concerning our “imaginative resistance” in relation to the endorsement of morally problematic situations in fiction, it seems that in certain situations we may be led to adopt ethical systems we disapprove of in real life in our response to fictions.  Anecdotally, a friend of mine, who is normally mild-mannered and morally virtuous, reported that he found himself responding in Mafia-style to the fate of a character while watching The Godfather, agreeing with the don that a member of “the family” should be “silenced” for his insubordination.

In “Just the Imagination,” Nichols attempts to explain this asymmetry, and his solution hinges on the work that desires play in the formation of emotions.  Nichols writes the following:

When watching Dr. Strangelove, the set of inferences and activations we have about the imaginary scenario is shaped by our desires about what happens in that imaginary scenario.  When it comes to the real world, we have powerful and consuming desires for the survival of human life….When it comes to black-comedy, we typically do not have such powerful desires for the preservation of human life in the imaginary scenario.  Hence, we are not compelled to draw out disturbing inferences like billions of innocent people will die horrifically painful deaths. Rather, genre considerations make us want to focus instead on Slim Pickens’ exuberant missile ride.[9]

Nichols thus wants to explain the asymmetries between our real-life and fictive responses by appealing to the “differences in input” between belief and imagination, claiming that in black-comedy scenarios our desires for the preservation of human life are not as strong as they are when directed towards real-life people.  The problem is that this is not always the case; readers care very much about “preserving the life” of some fictional character, as the responses to the death of Sherlock Holmes testify.  Indeed, many disaster movies appeal specifically to these concerns about the preservation of human life, the absence of which would result in audiences  simply not caring about whether the inhabitants of the building in Towering Inferno make it to safety.  It is precisely because we do care about the death of Desdemona that we feel deeply moved by the tragedy of Othello. 

Such considerations led Dickens to change the ending of Great Expectations because he felt that an unhappy ending would justifiably upset readers.  Nichols appeals vaguely to “genre considerations” as crucial for determining the nature of emotional responses, but how such considerations enable or disable the usual emotional responses needs to be detailed more carefully.  It is not sufficient to appeal to differences in input because, of course, one could claim that fictive input is always different from factual input in that one is fictional and the other is not.   

Yet it is this difference that simulation theory supposedly does not care about other than that fictive input runs off-line.  The single-code simulation hypothesis purports to explain the similarity of responses between imagination and belief, with the focus placed on the representational content of beliefs and make-beliefs rather than their mode of presentation, that is, fictional or non-fictional.  This was regarded as a virtue of the account:  the single-code hypothesis cannot simply appeal to aesthetic notions like genre to explain asymmetries without undermining its own force as an explanatory theory.  At least, not without saying more about it.

3. More than one code?

There is much that is salient in the claims of the single-code hypothesis, and its power in explaining many of our responses to fiction is evidenced by its current popularity.  It appears to deal well with explaining certain kinds of imaginative engagement along with some kinds of pretend play in children.  But, for all that, I would like to suggest, in the discussion on the nature and functioning of the imagination, that it is not the only game in town.  In some ways, on a simulation account of the imagination we should not be very surprised to find deeply embedded symmetries between imagining and believing because the simulation account is modeled on belief, that is, the contents of the imagination are cashed out in the language of belief and cognitive psychology.

Before proceeding with an alternative account of the imagination, I would like to make the perhaps obvious point that both the “off-line” analogy and the “single-code” terminology are metaphorical.  There are instances when Nichols does not explicitly spell out the implications for the single-code hypothesis and instead the metaphor is employed to do the explanatory work. As noted above, he takes it as a counter-example to the hypothesis that there are asymmetrical emotional responses to many pretend situations and their real-life counterparts, but does not consider that the asymmetrical behavioral responses to pretend situations and their real-life counterparts constitute a challenge to the single-code hypothesis. 

The asymmetries are accounted for by the “off-line” nature of the pretend response, but we are given no guidelines for knowing which behavioral responses fall within the scope of the metaphor and which do not.  So the fact that we do not run up on stage to prevent the murder of Desdemona is explained by the “off-line” nature of our emotional response.  Yet other behaviors, such as crying or avoiding the dark after seeing a horror movie, behaviors that are on a par with responses to similar real-life situations, are seen as consistent with the “single-code” hypothesis.  Thus the appeal of the “single-code” or “off-line” metaphor is at times vague, and it is not always clear what kinds of responses, emotional or behavioral, are consistent with or contradict these hypotheses.  

4. Child’s play

I would like to suggest that the single-code hypothesis is not sufficient for explaining many of our emotional responses to fictions.  I argue that our imagination may at times work in ways that are inconsistent with our ordinary cognitive engagement:  imagining that-p may differ from believing that-p.  I would argue that this is particularly true of our imaginative engagement with many kinds of literary works that employ devices to pull our imagination in a different direction from our cognitive engagement.  I am not arguing that it is only literature or the arts that engage our imaginations in these ways; different artworks and genres may involve different kinds of issues.  But if the case can be made for our literary imagination, this may pave the way for further discussion on the other ways in which our imaginations are engaged. 

Let’s start by looking at different ways in which children engage in pretend play.  In arguing that belief and imagination are functionally isomorphic, Nichols outlines an experiment conducted by the psychologist Alan Leslie:

Leslie had young children watch as he pretended to pour tea into two (empty) cups.  Then he picked up one of the cups, turned it over and shook it, turned it back right side up and placed it next to the other cup. The children were then asked to point at the “full cup” and at the “empty” cup.  Both cups were really empty throughout the entire procedure, but the two-year-olds reliably indicated that the “empty cup” was the one that had been turned upside down and the “full cup” was the other one.[10]

From this Nichols concludes the following:

On the most natural interpretation of this, the child is imagining that the cup is empty.  But the child also, of course, believes that the cup is empty.  This suggests that the crucial difference between pretence representations and beliefs is not given by the content of the representation.[11]

And so:

The prevailing cognitivist view is that the pretence representations are processed by the same inference mechanisms that operate over real beliefs.[12] 

In contrast, the following examples from the psychoanalytic tradition differ from the cognitive one.  In psychoanalytic literature, play is often understood as a way of dealing with important emotions that the child is at a loss to articulate, emotions that he subsequently deals with through play.  In the first account, psychoanalyst Susan Isaacs presents the observations of a nursery school superintendent concerning her young charges: 

She observed in many cases that when a two-year-old child was left in the nursery school for the first time and was feeling lonely and anxious because of the parting with his mother and being in a strange world, the plaything which most readily comforted him was the ‘posting box’, a box into which he could drop through appropriate holes in the lid a number of small bricks, the lid then being taken off and the lost objects rediscovered inside.  The child thus seemed to be able to overcome his feelings of loss about his mother by means of this play, in which he felt lost and rediscovered objects at his own will.[13]

Here is another example provided by the psychoanalyst Hanna Segal:

A child’s normal play is a major way of working through a conflict.  I remember watching a boy of just under three, when his mother was away for a day giving birth to his first sibling.  First he made a complex rail track for his wooden toy train.  He filled it with little toy people and they had several crashes.  Then he brought in ambulances.  Soon he delineated some fields with his little bricks and filled them with toy animals.  A complicated play resulted in shifting male, female, and baby animals in and out of the fields and train.  There were fights and crashes; ambulances came to the rescue.  Throughout, he was telling himself stories.  He introduced bigger toy men to regulate the traffic, and so on.  In other words, he presented the birth of new babies and his conflicts about it in very many different ways.  For an interested observer it was fascinating to watch.[14]

Here the function of playing takes on a significance of a special kind, one that is not exhausted or fully explained in cognitivist terms.  The children’s play in these examples is symbolic in ways not apparent to the children themselves.  In the examples provided by Nichols, the children were very much aware of the symbolic nature of their play.  For instance, the children are aware that the empty cup in reality represents a full cup in the pretend game.  In a game of mud pies, a child pretends that a pile of mud is a pie, and the child may manipulate this prop in ways suitable for the game, perhaps by pretending to eat it, and so on.  But in the examples presented by Isaacs and Segal, there is no straightforward way in which the children are pretending that this is that. 

In the Isaacs example, the little boy is using the game of posting bricks as a way to overcome his feelings of loss at his mother’s absence.  It would be inappropriate to say that the child is drawing a conscious comparison between his play activity and his real world desires.  Seeing that the brick game is directed at or is concerned with the child’s relationship with his mother requires an act of interpretation.  But this does not negate the claim that what the game is concerned with, what thoughts and feelings dominate the child’s inner world in play, are his anxieties concerning his mother’s absence.  In this case, the child’s playing is expressive of his emotional states rather than being goal directed; in this way it is very different from the functions of belief-desire pairings, for example, which work to bring about a certain state of affairs in the real world.  Posting little bricks and then retrieving them is not easily translatable into a cognitive account in the way that pouring pretend-tea or pretending to bake mud pies is.

And here we see one difference between imagination and belief.  Given that beliefs are world-oriented, concerned with and sensitive to truth conditions, they are connected with the formation of intentions and the execution of action.  But activities of the imagination are often not primarily world-directed and need not be concerned with or sensitive to truth conditions.  This often results in a difference in the function of states of the imagination, such as play, and because play often occurs in symbolic guise, one thing stands for another and make-beliefs may neither be expressible as coherent propositions nor follow the logic of propositional states. 

One might think that it is only by appealing to psychological theories, such as psychoanalysis, that deal with deeply unconscious or repressed material that differences between believing and imagining can be found.  But this is not the case.  Here are some more differences between imagination and belief:

A child can pretend that a chair is a tank and yet not expect it to shoot real shells, and there have been experimental demonstrations that small children can keep alternative or changing beliefs in mind if they are doing a task in play rather than for real.  Vygotsky wrote:  ‘in play, the child is always above his average age, above his daily behavior; in play it is as though he were a head taller than himself.’  In the world of play it becomes possible partially to free representations from their referents and allow these freed representations to be modified, creating a more flexible mode of thought that encourages the emergence of latent mental structures.  Playfulness enables ways of functioning that are rarely used, developmentally surpassed, or only just being formed to occupy centre stage.[15]

This is at odds with Nichols’s claim that the content of a pretend belief does not differ from the content of a real belief in the same context.  To quote him again:  “[T]he crucial difference between pretence representations and beliefs is not given by the content of the representation.”[16]   What Nichols means by this is that in cases of both pretending and believing, the mental content is identical.  But is it?  Let’s take another look at the Leslie example.  The claim is that when the child pretends that “the cup is empty,” and when she believes that “the cup is empty,” the representational content, “the cup is empty,” is identical in both cases. 

But in the case of belief, the child’s belief that the cup is empty sits within a network of a host of other beliefs, such as the belief that the cup will not transform into a kettle, that it is not invisible, not likely to grow wings, or not filled with invisible gremlins.  But all of these representational contents may very well be consistent with the child’s pretend representation that the cup is empty, depending on the nature of the game being played.  And if, by appealing to the holism of the mental, the representational content of mental states is determined in part by relations with other representational contents, then it seems to follow that the representational content of a belief and the representational content of a make-belief are different.  In other words, differences in function in relation to pretending and believing may also bring with them differences in representational content. 

5. Literary engagement

It is important to point out that much of our ordinary thinking and imaginative activity may often be governed by both cognitive and imaginative codes simultaneously.  Even in cases of imaginative play heavily imbued with symbolism, it is clear that children continue to have ordinary beliefs and desires about the objects of their play.  The little boy in the Isaacs example, above, understood that the bricks he was placing in the posting box were bricks and not something else. 

Moreover, he also knew that ordinarily posting boxes are used for mailing letters, and it was these beliefs, associations, and the like that partially determined his play activity.  Conversely, in ordinary cognitive activity, we may often follow a train of thought whose links are associative rather than logical.  We may use a metaphor to convey an idea that could have been stated in literal language but would have lacked the emotional force of a more literary expression.  Thus, in practice, it is rare for either cognitive or imaginative thought processes to be governed solely by their respective “codes.”  This does not imply that the two different types of mental functioning are reducible to one another.  Nor does it deny that mental activity is governed by different principles of mental functioning.  The further we enter the realm of the imaginary, the more likely are we to see stark and irreconcilable differences between imaginative activity and belief-desire cognition.

Our engagement with literature likewise may involve more than one type of mental activity or “code.”  I agree with Nichols that our engagement with fictions can be viewed as an example of pretend play in adults, although adult fictive engagement tends to be more passive, a difference that may result in different emotional responses.[17]  I agree, too, that genre plays a vital role in the way we engage with fictions; different kinds of fictions will require different kinds of engagement.  It is a mistake to slice representational content from aesthetic form as their function in shaping emotional responses is crucial.  Our emotional responses are directed at and governed by more than the plot or story alone.  “[F]orm and content are inseparable,” noted novelist and critic David Lodge, and “style is not a decorative embellishment upon subject matter, but the very medium in which the subject is turned into art.”[18]  The philosopher Richard Moran agrees:

If people can respond with emotion to some nonfigurative work of painting or music, which provides no scenario for them to make-believe is true, then perhaps what makes the difference between works that are emotionally engaging and those that aren’t is not a matter of aptness for make-believe or the generation of additional fictional truths, but rather a difference in certain other features.  In the theater, for instance, we might think of the various nonmimetic effects of things such as music (including song), lighting, figurative language, pacing, and compression of time, and other effects that provide emotional punctuation and tone, but do not necessarily represent anything themselves, and indeed may be quite mimetically out of place in the scene presented.  They may thus directly impair the aptness for make-believe, without contributing any additional fictional truths of their own, while yet being directly responsible for the emotional involvement of the audience.[19]

Poetry is a clear example of a genre[20] where the emotional resonance is typically located not only (or even mainly) in the subject-matter of the poem but rather in what I’ll term the 'aesthetic elements,' such as association, alliteration, rhyme, meter, metaphor, imagery, and so on.  A case study is Rupert Brooke’s World War I sonnet, “The Soldier,”[21] which continues to enjoy wide-spread popularity as shown by its inclusion in various anthologies.  Brooke’s poem arouses the emotions in ways other than its patriotism.  It engages the imagination in lively and surprising ways, and it is this feature, I suspect, rather than its pro-English sentiment, that assures its longevity.  The overt sentiment of “The Soldier” espouses patriotism, in that the speaker urges the reader not to mourn his death in battle because it serves the noble and worthy cause of fighting and dying in the service of one’s country.  But the emotional pull on the reader is achieved through other means.

The imagery of the sonnet, a form  traditionally used for a love poem, traces a movement from death to life.  It begins with an emphasis on death – “If I should die” – and ends with a vision of eternity and the afterlife – “Under an English heaven.”  In the poem the speaker is resurrected, altered from a piece of earth – “some corner of a foreign field” – to a breathing, speaking, being.  The line that begins, “A dust whom England bore” echoes the creation of humans by the God of Genesis, who breathed life into inert clay.  The octave focuses on the reawakening of the physical being of the soldier by emphasizing the body and physical activity:  He roamed the paths of England, breathed her air, and so on.

In the sestet, the imagery shows activities less physical and more intellectual and emotional:  the soldier is transformed from a piece of earth into a “pulse in the eternal mind.”  Note the emphasis on “thoughts,” “sights and sounds,” “dreams,” and “laughter” in the activities of thinking and learning.  Not only does the poem describe these activities, it also induces the reader to adopt the perspective of the soldier:  we too are urged to sense those “sights and sounds,” to dream along with him.  The reader moves from a position of thinking about the soldier to one of thinking and feeling with him.  The dead soldier comes alive by being thought about by the reader, and he begins to live through the reader as we come to view the world through his eyes.  This completes the resurrection of the dead soldier.  Through the act of reading, he gains new life, and the reader becomes the vehicle of this resurrection.

It is this kind of creative engagement, one that depends not so much upon the overt sentiment or theme of the poem, whereby the work achieves its greatest effect.  This effect depends upon the way the formal elements of the poem, that is, its imagery, rhythm, rhyme scheme, and structure, work together.  The literary elements influence the reader’s engagement with the poem in unconscious ways that require some detailed analysis in order to be discovered. 

It would be difficult to explain our emotional involvement with “The Soldier” if pretend-beliefs and ordinary beliefs were functionally isomorphic.  If I am right that the soldier comes to life emotionally through  the reader’s engagement with the poem, this cannot be fully explained by thinking of our literary imagination as akin to ordinary belief.   For one thing, we have no way of knowing how we would react were we to believe that the dead can be resurrected.  It is crucial that “The Soldier” asks us to contemplate a state of affairs, a resurrection, that could not happen in reality, because the poem invites the reader to engage in an act of reverie very different from ordinary thinking.  It is part of the function of the poem to make us think impossible thoughts and to lead us away from ordinary mundane life and not towards it. 

The poem uses a variety of literary techniques, including allusion, rhythm, and meter, and thereby draws from the reader a variety of emotional responses that have no clear parallel in ordinary psychological experiences.  Certainly not all poetry draws us away from ordinary life; but even where the poetic focus is on the everyday, poetry, and especially great poetry, brings an extraordinary attention that rises above the commonplace and allows us to see things anew.[22]  Peter Lamarque makes the point that our cognitive and emotional responses to poetry are often very different from our ordinary responses.  In response to Ben Jonson’s “On My First Son,” which speaks of the death of the poet’s seven-year old child, Lamarque writes that the poem “is an expression of grief, not the grief expressed.”[23]  What he seems to mean by this is that the poetic expression of a psychological state is not to be confused with the psychological state itself, that is, poetic expression is a crafted construction:  “To understand the emotions expressed in lyric poetry we need to know not about psychology but about poetry, its conventions and powers.”[24]

Conversely, engagement by readers with works of poetry will evoke and invoke imaginative responses different in important respects from our everyday cognitive and emotional responses.   We respond not only with pity to Ben Jonson’s poem but also with a joyful delight at the beauty of the poem and an admiration for its measured tone and lack of sentimentality.  Imaginative engagement not only deepens our emotional engagement in many ways but it also broadens it:  it opens our minds to ways of responding that ordinary cognitive behavior cannot.

Poetry, because it is highly stylized, is perhaps one of the clearest examples of the differences between imaginative and ordinary “cognitive” thinking.[25]  More “realistic” genres, such as the novel, often attempt to elide the distinction between their fictional worlds and the real world, and encourage their readers to feel emotional responses more akin to ordinary real-life responses.  But even the clear-sighted realism of Jane Austen’s writing is achieved precisely by the use of figurative elements in language.[26]  In discussing a passage from Persuasion, the literary critic Stephen Booth notes that “the special quality of Jane Austen’s vision of experience” is “communicated to us through a special kind of language, language which is more than the transparent container of Ideas.”[27]  The famous opening sentences of Pride and Prejudice invite the reader into a world where property and possessions are important in defining status: 

It is a truth universally acknowledged, that a single man in possession of a good fortune must be in want of a wife.  However little known the feelings or views of such a man may be on his first entering a neighborhood, this truth is so well fixed in the minds of the surrounding families, that he is considered as the rightful property of someone or other of their daughters.

The irony of Austen’s opening sentences allows us to adopt multiple viewpoints simultaneously:  we are made aware of the ambitions of the families to which we will soon be introduced, and the importance of status is highlighted.  The ironic tone alerts the reader to the fact that the narrator or author may not share the values of her character; the irony is subtle, though, mixing criticism with sympathy.  Note also the way in which the second sentence functions as a kind of literary camera lens, moving from the “neighborhood” to the “surrounding families” to “someone or other of their daughters,” and finally to Mrs. Bennet, who speaks in the next sentence.  These superbly constructed sentences allow the reader to engage with the events as they are being narrated while simultaneously wondering what the author thinks about her characters.

Another way of putting this is to suggest that Austen’s writing leads us to think about how we read.  Even as we engage with the fictional events, we adopt several viewpoints simultaneously.  This kind of multiple-perspective taking does not typically happen in ordinary cognition.  I do not suggest that it cannot, but the delicate literary constructions of a great writer like Austen open our imaginations in ways different from ordinary thinking, and this is part of the pleasure that reading gives us.  Literary style shapes our imagination in ways that ordinary cognition does not, and perhaps cannot. 

6. Conclusion

A full defense of my claim that imagination and belief are not governed by a “single code” would require a much longer and more detailed defense than the limits of this paper provide, and it would require nuanced discussion of genre and other aesthetic considerations.  There are important differences between various literary genres that only detailed discussion can foreground.  For instance, the ways in which the different elements of classical tragedy, such as form, content, dramatic structure, language, and so on, give rise to the  “singular phenomenon” of tragic pleasure (as noted by Hume), require a different kind of psychological engagement and, hence, explanation from the ways in which black comedies move us to laughter.  Such explorations must be left for another time.  

However, the general point holds, namely that works of literature engage us imaginatively in ways that very often differ fundamentally from ordinary cognition.   The psychologist Donald Winnicott noted that play provides children a relief from “the strain of relating inner and outer reality.”[28]  Likewise, literature provides adults with ways in which the imagination allows us, albeit temporarily, to avoid the strains of dealing with, in the words of T. S. Eliot, “very much reality.”[29]


The Soldier, by Rupert Brooke

If I should die, think only this of me:

That there's some corner of a foreign field

That is forever England. There shall be

In that rich earth a richer dust concealed;

A dust whom England bore, shaped, made aware,

Gave, once, her flowers to love, her ways to roam,

A body of England's, breathing English air,

Washed by the rivers, blest by suns of home.


And think, this heart, all evil shed away,

A pulse in the eternal mind, no less

Gives somewhere back the thoughts by England given;

Her sights and sounds; dreams happy as her day;

And laughter, learnt of friends; and gentleness,

In hearts at peace, under an English heaven.



Elisa Galgut

Elisa Galgut teaches in the Department of Philosophy at the University of Cape Town.  Her research interests include issues in aesthetics; emotional responses to fiction; the philosophy of psychology, with a particular interest in psychoanalysis; and animal ethics.

Published February 10, 2014.



[1] See, for instance, Kendall Walton, “Fearing Fictions,” The Journal of Philosophy, 75 (1978), 5-27; Mimesis as Make-Believe: On the Foundations of the Representational Arts (Cambridge, Massachusetts:  Harvard University Press, 1990); “Spelunking, Simulation, and Slime: On Being Moved by Fiction” in Emotion and the Arts, eds. Mette Hjort and Sue Laver, (New York:  Oxford University Press, 1997) pp. 37-49; Gregory Currie, “The Paradox of Caring” in Hjort and Laver, pp. 63-77;  The Nature of Fiction (Port Chester:  Cambridge University Press, 1990); “Imagination and Simulation: Aesthetics Meets Cognitive Science” in Mental Simulation, eds, Martin Davies and Tony Stone, (Oxford: Blackwell, 1995), pp. 151-169; Susan Feagin, Reading with Feeling: The Aesthetics of Appreciation (Ithaca: Cornell University Press, 1996).

[2] Shaun Nichols, Stephen Stich, Alan Leslie and David Klein, "Varieties of off-line simulation" in Theories of Theories of Mind, eds. Peter Carruthers and Peter K. Smith, (Cambridge: Cambridge University Press: 1996), pp. 39-74; ref. on p. 41.

[3] Shaun Nichols, “Imagining and Believing: The Promise of a Single Code,” The Journal of Aesthetics and Art Criticism, 62, 2 (2004), 129-139; ref. on 129.

[4] It is important to note that although Nichols argues that the truth of Simulation Theory makes a case for the single-code hypothesis, the converse does not hold; (a version of a) single-code hypothesis may be true even if Simulation Theory is false.  However, for the purposes of this paper I shall take Nichols to be arguing for a single code simulation hypothesis.

[5] Nichols, “Imagining and Believing,” p. 133.

[6] Ibid., p. 135.

[7] Shaun Nichols, “Just the Imagination: Why Imagining Doesn’t Behave Like Believing,” Mind and Language, 21 (2006), 459-474.

[8] See Tamar Gendler, “The Puzzle of Imaginative Resistance,” The Journal of Philosophy, 97, 2 (2000), 55-81; Kendall Walton, “On the (So-Called) Puzzle of Imaginative Resistance,” in Marvellous Images: On Values and the Arts (New York: Oxford University Press, 2008), pp. 47-59

[9] Nichols, “Just the Imagination,” p. 472.

[10] Ibid., p. 460.

[11] Loc. cit.

[12] Loc. cit.

[13] Susan Isaacs, “The Nature and Function of Phantasy,” in Unconscious Phantasy, ed. Riccardo Steiner (London and New York: Karnac Books, 2003), pp. 145-198; ref. on pp. 188-9.

[14] Hanna Segal, “Imagination, Play and Art” in Steiner, pp. 211-221; ref. on p. 212.

[15] Peter Fonagy, György Gergely, Elliot Jurist and Mary Target, Affect Regulation, Mentalization, and the Development of the Self (New York: Other Press, 2004), pp. 261-262.  Reference to Lev S. Vygotsky, Mind in Society: The Development of Higher Psychological Processes (Cambridge, MA: Harvard University Press, 1978), p. 102.

[16] Nichols, loc. cit.

[17] Alex Neill makes this point in “Fear, Fiction and Make-Believe,” The Journal of Aesthetics and Art Criticism (1991), 47-56.

[18] David Lodge, Language of Fiction (London; Routledge, 2002), p. 31.

[19] Richard Moran, “The Expression of Feeling in Imagination,” The Philosophical Review, 103, 1 (1994), 75-106; ref. on 81-82.

[20] I use the term ‘genre’ rather loosely to refer to both genus and species of literature; thus poetry and the novel are distinct genres, but so are lyrical and dramatic poems, comic and tragic drama. 

[21] Please see the Appendix at the end of the paper for the complete poem.

[22] I would like to thank a reviewer for bringing this point to my attention, and for other helpful comments.

[23] Peter Lamarque, “On Keeping Psychology Out of Literary Criticism,” in The Aesthetic Mind: Philosophy and Psychology, eds. Elizabeth  Schellekens and Peter Goldie, (Oxford University Press; Oxford and New York, 2011), pp. 299-312; ref. on p. 307.

[24] Ibid., p. 306.

[25] One could arguably judge Joyce’s Finnegans Wake as the best example in all of English literature.

[26] “The general effect is pervasive in Jane Austen:  an ironic detachment combined with a carefully discriminating sympathy and understanding.  It is indeed a logical passage; but such logic applied to human experience in fiction is not normative.  It constitutes the special quality of Jane Austen’s vision of experience, and is communicated to us through a special kind of language, language which is more than the transparent container of Ideas.”  David Lodge discussing a passage of Jane Austen’s Persuasion in Lodge, Language of Fiction, p. 16.

[28] Donald Winnicott, Playing and Reality (Harmondsworth, Middlesex; Penguin Books, 1988), p. 15.

[29] T. S. Eliot. “Four Quartets: Burnt Norton”(1935) in Collected Poems:  1909-1962 (London and Boston; Faber and Faber, 1963), p. 190.