Penultimate Draft
To appear in: Rivista di Estetica 70 (1/2019)
Engaging with Works of Fiction1
Abstract: The contemporary debate in the philosophy of literature is strongly shaped by the anticognitivist challenge, according to which works of literary fiction (that contain propositions that are
neither literally true nor affirmed by the author) cannot impart (relevant) knowledge to the readers
or enrich their worldly understanding. Anti-cognitivists appreciate works of literary fiction for their
aesthetic values and so risk to reduce them to mere ornaments that are entertaining, but eventually
useless. Many philosophers have reacted to this challenge by pointing at ways in which works of
literary fiction can be informative even though they lack worldly reference: it has been argued, for
example, that works of fictions are thought experiments; that they add not to our theoretical
knowledge, but to our know-how or to our phenomenal knowledge; or that that they help readers to
understand the perspectives of others. A stubborn defense of literary cognitivism, however, risks to
collapse into an instrumental understanding of literature.
In my paper I suggest that both sides in the debate focus too narrowly on semantic features
of the works in question that is tied to what I will call the “referential picture” of language. A shift
perspective is needed: for one, we ought to fully appreciate that the term “literature” does not refer
to a homogeneous phenomenon, but rather to a very heterogeneous and multifarious set of works
that are read by many different readers for many different reasons in many different ways. Second,
we need to understand that these works have in common much more than the semantic peculiarity
of lacking worldly reference: they are a unique means of communication between authors and
readers – and in particular the role of the latter is often neglected in contemporary debate. These
two points should help us to get a more comprehensive understanding of the practice of literature
and the vast range of values we can find works of literary fiction – and the interplay between them.
Keywords: Philosophy of fiction; Cognitive value of literature; Truth in fiction
The paradox disappears only if we make a radical break with the idea that language always
functions in one way, always serves the same purpose: to convey thoughts – which may be about
houses, pains, good and evil, or whatever.
(L. Wittgenstein Philosophical Investigations: § 304)
The “desert of the real” – an unattainable ideal?
The film Matrix depicts a dystopian scenario in which sentient machines subdue the human
population by making them victims of a shared illusion: while keeping their bodies captive in
tanks, they feed their sensory apparatus with stimuli that create in them the collective illusion that
they live and interact with one another in Los Angeles in 1999. Neo, the film’s protagonist,
succeeds in breaking free from this simulated reality and finds himself in a hostile future
environment that is marked by serious ecological disasters. He is welcomed with the words
1
Earlier versions of this paper were presented at the XIIth Workshop on Aesthetic Experience (Universidad de Murcia),
the Philosophische Gesellschaft krino (Universität Bern) and the Seminari di Filosofia della Percezione, Mente e
Linguaggio (Università di Milano). I have profited a lot from the discussions there and want to thank for suggestions
and critique. In particular, I would like to thank Francisca Pérez-Carreño, Georg Brun, and John Gibson for their
comments.
«Welcome to the desert of the real»2. This quote, I think expresses very well the ambivalent
attitude towards fiction we find in contemporary philosophy of literature: while many
philosophers regard fiction as a colourful game of imagination that can be entertaining, but need
not be taken seriously, others warn of its deceptive and manipulative power. Both seem to accept
the underlying idea that the main function of fiction is to invite the audience to indulge in escapist
pleasures, to leave the grey desert of everyday life behind and immerse in colourful parallelworlds that are rich of spectacular scenarios, where unrealistic features like witchcraft,
teletransportation, or happy, enduring relationships count as most ordinary, and which are
opulently furnished with spies, monsters, superheroes, space aliens, and other illustrious
creatures.
And while this picture seems to suggest that reality is an unpleasant place from which most
people want to escape, a grey and unspectacular desert that is void of meaning – especially when
compared to the colourful worlds of fiction – many philosophers call for sobriety. They remind us
that reality, even though it sometimes might seem grey and insignificant, remains beyond the
reach of works of fiction, like an unattainable ideal. The latter present nothing but pale, empty, or
even frivolous representations that might bear some vague resemblance to reality, but often
produce only a contorted caricature of it. Works of fiction, thus, are often regarded as selfreferential games that can be entertaining but, at the end of the day, are nothing but a frictionless
spinning in the void. According to this line of reasoning, it is not the real world that lacks
splendour, it is rather the worlds of fiction that lack a decisive moment: existence. To express it
metaphorically: an average fictional dream palace is more imposingly, splendidly, lavishly, and
elegantly furnished than the most luxurious of all existing castles – all this splendour cannot make
good for the decisive flaw, however, that it cannot provide shelter from wind and rain which, I
suspect, constitutes a serious flaw if you have an only marginal interest in the commodities of
everyday life.
This tension has a long history in the philosophical perspective on fiction: already Plato was
sceptical that works of fiction could contribute to our understanding of reality, but warned at the
same time of their seductive powers to manipulate the audience. In the third book of Republic we
can find the suggestion that passages that describe the underworld and its terrors be deleted from
2
This quote, incidentally, is only one in a series of references to Jean Baudrillard’s Simulacra and Simulation, from
which it is taken.
2
Homer’s works «since their descriptions are untrue, and will do harm to our future warriors.»
Plato considers these passages dangerous
«not because they are unpoetical, or unattractive to the popular ear, but because the
greater the poetical charm of them, the less are they meet for the ears of boys and men
who are meant to be free, and who should fear slavery more than death.»3
And though Plato suggests in other places that all works of representational arts are inapt tools to
deepen our understanding of the real world4 he suggests in this passage that the higher the
aesthetic quality of the work, the greater is its power to change the readers’ views or beliefs.
We find this tension over and over again in the history of philosophy: while at times works
of fiction are considered defective because they lack any connection with reality – they are taken
as ornaments that can give aesthetic pleasure but are, eventually, useless – at other times they
are celebrated (or discredited) for their power to stimulate our imagination, to enrich our
understanding, or to change our cognitive perspective.
Ornament vs Instrument
This ambivalent attitude becomes particularly manifest in the debate concerning the cognitive
value of literature. While most people would agree with the idea that reading works of (high)
literature contributes decisively to a person’s culture and refinement, several philosophers have
taken a more sceptical stance and denied that works of literature could add to our (worldly)
knowledge5. A widespread form of literary anti-cognitivism builds on a line of reasoning that I
propose to call the “Simple Argument”, and which suggests that works of literary fiction do not
contain faithful descriptions of real persons or scenarios and do not, in consequence, contain
relevant true propositions (but at best banalities or commonplaces). Moreover, some anti-
3
Plato, Republic III: 387b.
Cf. for example, Republic X: 598 ff., 607 f.
5
From the viewpoint of the sociology of science it is interesting to note that the anti-cognitivist view is rarely explicitly
defended from philosophers who specialize in philosophy of literature, but very often taken for granted from those
who work in other philosophical disciplines. Neither Hume’s famous dictum that «Poets themselves, though liars by
profession, always endeavour an air of truth to their fictions» (Hume 1978: 121), nor Bertrand Russell’s comment on
Shakespeare’s Hamlet that «… the propositions in the play are false because there was no such man.» (Bertrand
Russell 1940: 294), or Carnap’s remark that «[l]yrical poets … do not try to refute in their poem the statements in a
poem by some other lyrical poet; for they know they are in the domain of art and not in the domain of theory»
(Carnap 1966: 79f) were intended as theoretical reflections on literature or poetry, but served to illustrate some other
philosophical point. The short quotes are intended as exemplary for a widespread attitude; needless to say that we
can find important exceptions on both sides: there are eminent philosophers who take the cognitive value of
literature for granted, and some of the central contributions to the philosophy of literature develop an anti-cognitivist
position.
4
3
cognitivist philosophers remind us that literary works of art do not in general provide evidence or
arguments in favour of a more general hypothesis that might be contained in or emerge from the
text6. There is some plausibility to this view: we would not hesitate to regard a person, who takes
a work of fiction as a faithful representation of the world, a victim of a serious misunderstanding.
If someone, who reads the first sentence in Kafka’s novel The Trial: «Someone must have
slandered Joseph K., for one morning, without having done anything wrong, he was arrested.»
would seriously start to worry about Joseph K., search for his email-address on the internet, or
consider to initiate an online petition for his liberation, we would hardly appreciate her for her
talent to extract profound insights from the novel, but rather consider her behaviour pathological.
Readers typically know that what they read in literary works of art is not to be taken at face value.
There is, thus, some plausibility to the “Simple Argument”. But can it really show that literature
does not contribute to our knowledge?
Readers who care for literature and actively engage with literary works of art on a regular
basis will likely have the impression that the “Simple Argument” misses the crucial point. To
question the cognitive value of literature and reduce it to a mere aesthetic ornament not only
seems to manifest a dismissive attitude towards literature that could seriously undermine its
relevance7, it also seems at odds with our everyday reading experience. It should, therefore, not
come as a surprise that many philosophers, who have contributed to the philosophy of literature,
have thought it necessary to react to the anti-cognitivist line of reasoning. Many of them have not
only insisted that, but also tried to show how literature can impart relevant worldly knowledge to
the reader. In what follows I will briefly sketch three cognitivist arguments.
One strategy that immediately seems to offer itself is to argue that we learn from works of
literary fiction not in virtue of the propositions that are asserted in the text, but rather in virtue of
propositions that emerge from the text. Literary works of art, one might be tempted to argue,
contain a special or more profound kind of truths, call it artistic or literary truth – which, unlike
scientific truths, are not explicitly stated on the “surface” of the text, as it were, but contained in
the work on a more profound level. It arguably takes a deep engagement with the work and some
interpretive effort from the reader or critic to extract this level of meaning. To the extent in which
the latter can be explicitly stated – or paraphrased (in a sufficiently appropriate manner) – by the
6
I am following here Noël Carroll’s distinction between three anti-cognitivist arguments: the «banality argument», the
«no-evidence argument» and the «no-argument argument» (cf. Carroll 2002: 4ff).
7
With this I do not want to suggest that anti-cognitivists do, in fact, have a dismissive attitude towards literature. I
merely want to recall that their position has often been received in this way.
4
critic, one could argue that the resulting statements express what readers of a work are invited to
mull over and consider. Peter Kivy, for example, has argued that a statement of this sort can be
treated as a «live hypotheses», i.e. as one «that appears to the person who contemplates it as at
least a viable candidate for belief, even though he or she might not believe it»8.
This strategy is not unproblematic, though, for it is not at all clear whether the theme of a
work can be expressed or paraphrased by an explicit statement in the first place9. But even if
readers or critics could come up with (a set of?) thematic statements that can be extracted from
the text, these statements risk at being either too specific to be relevant or so general that they
turn trivial, as Jerome Stolnitz has argued in his widely received article On the Cognitive Triviality
of Art10. Rather than acquiring relevant artistic truths that can only be communicated by literary
works of art – or are communicated by them in a particular manner – we would end up with a list
of most general platitudes. Moreover, even if we could come up with an interesting and relevant
statement, the mere fact that it is expressed or implied by a work of art is not a good reason to
believe it, as Peter Lamarque has pointed out. «The novel does not prove the claim, at best it
illustrates it, and if we were to try to prove it we would need arguments from philosophy or
psychology»11 or other disciplines.
Another attempt to defend the cognitive value of literature against the anti-cognitivist
critique focuses on the analogies between literary works of narrative fiction and thought
experiments12. It is often pointed out that thought experiments are fictional narratives, as they
invite the reader to imagine «a series of temporally distinct events»13 without asserting that they
have necessarily taken place in exactly the way described. Like real experiments, it is argued, both
thought experiments and many works of literary fiction are set in carefully controlled settings,
where potentially confounding factor are excluded.
The basic idea of this line of reasoning strikes as very plausible. Some qualifying
observations need to be made, though. First, we can note that there is notorious disagreement on
how exactly thought experiments are to be conceived – some argue that they are arguments
«disguised in a vivid pictorial or narrative form»14 while others suggest that they are autonomous
8
Kivy 2006: 101f.
For a discussion of this point, cf. John (2016).
10
Stolnitz (1992).
11
Lamarque 2009: 237.
12
Cf. Carroll 2002; Elgin 2007; Davies 2007b.
13
Egan 2016: 140.
14
Norton 2004: 45.
9
5
sources of knowledge, means of grasping abstract principles15. In philosophical contexts, they
often serve to defend or motivate a claim, or to render it vivid or plausible. There is wide
agreement, however, on the fact that thought experiments serve to justify or illustrate (or show
wrong) a clear-cut claim or hypothesis. As a consequence, thought experiments are focused and
avoid unnecessary details. Typically, the claim that is to be illustrated or argued for is explicitly
stated in the text, ambiguities are avoided as good as possible. Moreover, scientific thought
experiments are clearly not designed to arouse pleasure in the reader – if they do so, it is at best a
collateral benefit. Rather, they are successful if they get their point across. Works of literary
fiction, on the other hand, are typically rich of (seemingly) «unnecessary details»16, they make use
of a rich and colourful language that should arouse aesthetic pleasure in the reader. Often, they
play with ambiguities. In fact, we appreciate literary works of art for their openness that allows for
competing interpretations. If an author tries too hard to get a certain point across, she risks to be
criticized for being didactic. The very fact that occasionally we re-read literary works of art shows
that they mean more to us than a good illustration of an abstract point. Thus, the analogy
between literature and thought experiments carries only up to a certain point.
Moreover, the analogy with thought experiments seems to work particularly well for some
works of literature, but not for others. An example where it does work particularly well is Jorge
Luis Borges’ short story Pierre Menard, Author of the Quixote, which depends on specific
characteristics of the text that distinguish it from many other literary works of art. The work is
quite short and does not develop a complex and many-layered narrative. Most importantly, it
advocates a clear-cut thesis, suggesting that properties that are not intrinsic to the work (but are
related to the context in which it was produced) are relevant for the work’s identity criteria. This
thesis is illustrated with a fictional example that is well explained by the narrator, who even
presents arguments to support it17. It is sometimes argued that works of literary fiction that
develop a more complex narrative and the “message” of which cannot easily be boiled down to a
15
Cf. for example, Brown 2011.
Cf. Rorty 1991: 80f.
17
The short story is about a French writer, Pierre Menard, who has created a literary work that is identical, word for
word, with (parts of) Cervantes’ Qixote. The narrator suggests that Menard’s is not a mere copy of Cervantes’ work,
but a self-standing artwork, arguing, for example, that «Menard’s fragmentary Quixote is more subtle that
Cervantes’», it is «infinitely richer» and «more ambiguous» (Borges 2007: 42) and even though the texts are identical
word by word, the narrator suggests that «[t]he contrast in style is also vivid. The archaic style of Menard—who is, in
addition, not a native speaker of the language in which he writes—is somewhat affected. Not so the style of his
precursor, who employs the Spanish of his time with complete naturalness.» (Borges 2007: 43). These points show
that the identity criteria of a work are not restricted to intrinsic properties of the text, but may include extrinsic
properties, such as historical properties related to the context in which the work was created.
16
6
clear-cut thesis might still be considered thought experiments insofar as they situate their
protagonists in specific settings to “see what happens”, as it were. This could be the motive why
Jane Austen liked to work on «three or four families in a country village»18. Catherine Elgin, who
brought up this example, explains:
Because the relations among the members of three or four suitably characterized families are
sufficiently complicated, and the demands of village life sufficiently mundane, her stories can
exemplify something worth noting about ordinary life and the development of moral personality19.
There are literary works of art that can hardly be interpreted as thought experiments, though, and
in some cases this strategy even seems to do violence to the work. Many works of poetry, for
example, are written to express a feeling or an inner sentiment and not to prove or illustrate a
hypothesis. Many novels, and in particular historical novels, aim at giving a detailed sketch of what
it was like to live in a certain situation or in a given historical period, without making the minor
effort to develop a controlled setting in which to perform a thought experiment. In short, the
strategy proposed is suited well for some works, but cannot be applied to literature tout court20.
The analogy between literature and thought experiments is interesting as long as it is not forced
on all works of literature.
A third cognitivist strategy that I want to consider is to suggest that the kind of knowledge
we can gain from works of literature is not propositional knowledge, the content of which can be
explicitly stated and communicated to others by a proposition, but rather a form of phenomenal
knowledge, i.e., a knowledge of what-it-is-like to have a certain experience, be in a given situation,
or be a certain character. It is important to note, however, that this form of knowledge can be
conceived in two different ways21. On the one hand it can be argued, with Frank Jackson and
Thomas Nagel22, that phenomenal knowledge is irreducible to propositional knowledge and can be
acquired only on the basis of first-person experience. Frank Jackson presented a famous thought
experiment concerning Mary, a brilliant scientist who (for the sake of argument) has never seen a
colour in her life, but has learnt all there is to know about the neurophysiology of colour-vision.
18
Jane Austen in a letter to her niece in September 1814, quoted in Elgin (2014: 232f).
Elgin 2014: 233.
20
This point is sometimes acknowledged explicitly, but often omitted. There are good reasons to believe that no one
really wants to suggest that all (cognitively valuable) works of literature are thought experiments and that sometimes
the thesis is only formulated in strong terms just to make it more efficient. It is important for the point I want to
develop in this paper, though, that literature is a very multifaceted phenomenon. Thus, we should not expect to find
one single mechanism that could explain how literature tout court can enrich our cognitive perspective on the world.
21
Cf. Brendel 2013.
22
Jackson 1982, Jackson 1986, Nagel 1974.
19
7
One day she succeeds in breaking free from her black-and-white environment and gets to see a
ripe tomato. In this moment she learns, according to Jackson, something she did not know
beforehand: she learns what it is like to see something that is red.
Even if we grant the controversial assumption that phenomenal knowledge in this sense is
a form of genuine knowledge that is distinct from propositional knowledge, it seems quite obvious
that we cannot acquire this form of knowledge from works of literature. It does not suffice to read
about it, one has to live the experience first-hand to get to know the distinctive qualitative
character one experiences when tasting a pineapple or seeing a red-and-orange Rothko painting,
for example. The suggestion that a well-crafted sonnet, a rousing short-story, or a perceptive
novel could communicate this kind of knowledge would undermine our motives to postulate this
form of knowledge in the first place: phenomenal knowledge (in Jackson’s sense) cannot possibly
be expressed on a linguistic level, not even in its highest artistic forms.
One could, on the other hand, conceive of phenomenal knowledge as a complex form of
propositional knowledge and suggest that works of literature can communicate what it is like to
be in a situation of some complexity, like being a single parent, growing up in a small town or a
living as a foreigner in an Easter-European city. In fact, works of literature – and, in particular,
works of narrative fiction – often contain detailed descriptions of aspects the reader might have
neglected so far, but appear salient to the protagonist or the (implied) author. Moreover, they
give insight to the protagonist’s stream of thought or inner monologue or share (general)
reflections with the reader. In this way, a distinct perspective emerges. The cognitive value of
literature, according to this line of reasoning, lies in broadening the readers’ horizon by presenting
perspectives or subjective points of view23 that were up to that moment unfamiliar to the reader.
This is not the place to discuss the particular merits and problems of this proposal in depth;
I rather want to focus on two aspects that are relevant to the point I want to develop. First, what
strikes me as most interesting about this proposal is that it works best if it does not try to reduce
the cognitive value of a literary work to specific bits of (propositional or non-propositional)
knowledge that are imparted to the reader. It seems to me that the proposal is most promising if
it admits that it is not a specific (set of) proposition(s), but rather to a quite complex picture that is
not transmitted, but presented to the reader. While the “Simple Argument” suggests, as we have
seen above, that the cognitive value of literature lies in the communication of true propositions
that are absorbed by the reader who (is supposed to) integrate them in his system of beliefs, in
23
Cf., for example, Walsh 1943; Burri 2007; Vendrell Ferran 2018.
8
the present proposal the reader is confronted with a complex picture she is not expected to
appropriate.
Second, it is important to note that not all works of literature, and not even all works that
have cognitive value, inform the reader by allowing her to acquire knowledge of this kind. Only a
sub-set of all works of literature can make it manifest what it is like to be in a certain situation or
to be a certain character; and even those that try to do so often do it in very different ways. Thus,
even with this third strategy we should not expect to have unveiled a hidden mechanism that can
explain the value of works of literature tout court, but only one of many ways in which some
works of literature can help the open-minded readers to make cognitive progress.
So far I have discussed three cognitivist proposals. We could take into considerations
further proposals. John Gibson, for example, has made an interesting point arguing that literature
enriches us not on the level of propositional knowledge, but on a deeper level of understanding,
that of acknowledgement24. Martha Nussbaum has shown how some works can enhance our
sensitivity when it comes to moral perception25 and I have argued that formal, stylistic aspects of
literary works of art train the readers’ capacities to acquire amplify their (propositional)
knowledge26. More than giving a comprehensive overview of literary cognitivism, my aim here is
to underline an aspect they all have in common: all (interesting) cognitivist proposals I am aware
of elaborate on a very plausible idea that can, in fact, explain the cognitive effectiveness of some
works of literature, but cannot uncover an aspect all works of (good) literature have in common
and in virtue of which we can consider them as cognitively valuable. We should, thus, opt for a
pluralistic position and consider the conclusion that each literary work of art that has cognitive
value adds to the readers’ knowledge in its own, particular way.
This conclusion should not come as a big surprise. After all, literature is a very multifarious
phenomenon and most writers do their best to distinguish themselves from the others.
Nonetheless, it seems to me, the conclusion is not really appreciated in the contemporary debate
of philosophy of literature. As a consequence, we find the discipline at an impasse: anticognitivists, who defend the general thesis that no work of fictional literature can have cognitive
value, reduce them to mere aesthetic ornaments that are entertaining, but eventually useless. The
cognitivist replies, on the other hand, often make the mistake of generalizing on aspects that are
24
Cf. Gibson 2003, Gibson 2007.
Cf. Nussbaum 1990.
26
Cf. Huemer 2007.
25
9
present in some works of literature, but not in others27. In this way, they risk instrumentalizing
literature, suggesting that the communication of knowledge is a central and substantial value of all
serious works of fictional literature.
The discussion so far has shown, I hope, that if we want to get a clearer idea on the
cognitive value of literature, we have to appreciate the fact that it is one value among many that
literary works of art can have. Being able to widen the readers’ cognitive horizon is not a sufficient
condition for the literary value of a work; it is not even a necessary one. Most importantly, we
should not expect that all works who are of cognitive value share a common trait; we should
rather enjoy the idea that each literary work of art is unique and that authors often try to find new
and original ways to broaden the readers’ cognitive horizons.
«A picture held us captive»28
If what I have said so far is plausible, it shows that the philosophical reflection on literature should
start afresh by adopting a new perspective on the phenomenon in question. In the last section I
have suggested that we should not overlook the differences between works of literature. We will
get a more profound understanding of literature and its cognitive value, I have argued, if we
restrain our urge to come up with coarse generalizations and instead draw our attention to the
distinguishing features that render interesting literary works of art unique. In what follows I will
suggest that it will also be important to get a more comprehensive perspective on the full
phenomenon of literature: rather than focusing exclusively on the text or on its (real or implied)
author, we should understand literature as a social phenomenon that includes, next to texts and
authors, also readers who, in many cases, are invited to confront their own perspectives with the
one that emerges from the text.
In the debate concerning the cognitive value of literature, especially in the last four or five
decades, this point has often been overlooked. The reasons for this, it seems to me, are
sociological rather than philosophical: many philosophers who have taken part in this debate have
taken for granted an understanding of language that could be dubbed referential picture29 and
according to which the main function of language consists in describing persons and events that
27
There are exceptions, though. Íngrid Vendrell Ferran, for example, has recently advocated a pluralistic conception of
the cognitive value of literary fiction; cf. Vendrell Ferran (2018).
28
Wittgenstein Philosophical Investigations, § 115.
29
Cf. Huemer 2004.
10
belong to an extra linguistic reality30. Consequently, the concepts of truth and reference play a
central role: they are taken to be the basic concepts on the basis of which a comprehensive theory
of language is to be developed. Within this picture, the “Simple Argument” gains plausibility and
the question of whether propositions contained in texts of fictional literature can be true has
become central – a focus, which is, in my point of view, a form of déformation professionelle and
which should puzzle us: the notions of truth and reference prima facie do not seem to be relevant
when it comes to account for the language of fictional literature. Moreover, readers who engage
with literature typically focus primarily on works, and not on single propositions contained in it.
More importantly, we can note that in the debate a very narrow conception of cognitive
value is taken for granted – a conception of that is inadequate not only in the context of fictional
literature, but also beyond. This conception, which presupposes that cognitive progress consists
(mainly) in the absorption of true information or the fixation of belief, was criticized by Goodman
and Catherine Elgin31, who has elegantly described and criticized it with the following words:
Human beings seem to gather information in the way that squirrels gather nuts. Bit by bit, we
amass data and store it away against future needs. Many epistemologists and laymen take
cognitive progress to consist in data gathering.32
According to Elgin, this conception has clear limits because there are cases where cognitive
progress does not consist in the absorption of new bits of information, but in categorical
understanding, in a re-organization of our conceptual schemes that allows us to see hitherto
ignored aspects that have always been there before our eyes. Literature and, more generally, art
enriches us on a cognitive level not by adding new propositions to our knowledge data-base, but
rather by the fact that it seeks «to challenge, to disorient, to disrupt, to explore, and thereby to
reveal what more regimented approaches lack the resources to attempt»33.
The importance of Elgin’s point lies in the fact that she invites us to a shift of perspective. It
does seem to me, however, that she grants a lot, probably too much, when she grants that very
often cognitive progress does consist in taking in true information. With this, in combination with
30
This picture of language has been particularly successful among early analytic philosophers, who took propositions
as starting points of their analyses of language. The current debate in philosophy of literature is strongly shaped by
the contributions of analytic philosophers who have, from the mid 1970ies on (cf. Huemer 2017), focused more and
more on fictional literature and who have considerably enriched the debate. Their influence has also led, however, to
a dominance of the referential picture of language in contemporary philosophy of literature.
31
Cf. Goodman 1978: 21f, Elgin 2002.
32
Elgin 2002: 1.
33
Elgin 2002: 12.
11
her suggestion that «works of fiction are thought experiments»34, she runs the risk of
instrumentalizing literary works of art. It seems to me that the absorption of true proposition is of
little importance not only when we re-organize our conceptual schemes; it hardly counts in most
cases, in which we engage with forms of written communication. Very rarely we read a text with
the immediate goal to take in true propositions. We do so when we open a telephone book or
check the birth-date of a historical person on Wikipedia. When it comes to more complex texts,
however, we typically read to get to know a perspective on a specific aspect of the world,
independently of whether this perspective is “true” or “false” – a perspective that we can critically
examine or adopt and make our own. Hardly ever we will take it for the last word on the topic.
This holds even in the cases of scientific textbooks or newspaper articles – which are often quoted
as paradigmatic examples of texts that do have cognitive value. When one studies a textbook, one
typically knows that the theory presented might be state of the art today, but dated in a couple of
years and falsified some years later. Even more so with newspaper articles: we have been trained
from early schooldays on not to blindly trust what we read: every article gives not only factual
information on the world, but also communicates the perspective of its (real or implied) author
and might, in consequence, be tendentious.
With this I do not mean to question the idea that we can learn from textbooks or
newspaper articles; I rather wish to note that when we engage with these texts, our attitude is not
(or better: should not be) that of a blind absorption of the information contained in it. We rather
(should) take the chance to confront our own perspective with the one that emerges from the text
– with open results: we might revise old beliefs that are in contrast with the new perspective that
emerges from the text or insist in what we have taken for true so far and dismiss the perspective
that was offered. Whether or not a reader learns from a text, in other words, does not depend on
whether the text offers true propositions that the reader could add to her knowledge data-base,
as it were. Readers of texts typically learn by engaging in critical reflections that allow them to
draw their own, personal conclusions that are the result of a (critical) contrast of their own
perspective – a perspective they had already before reading the text – with the one that emerges
from the text.
This process is guided, but not determined by the text: readers are autonomous and free
to accept what they think should be accepted. Moreover, it seems relevant to me that texts not
only invite to reflect a certain set of assertions; they also show how a given topic can be
34
Elgin 2007: 47.
12
approached: the considerations, but also the theoretical moves that can be made in order to come
to the position where one can assert a given proposition. By studying a scientific textbook or a
philosophical treatise, for example, one is offered not only propositions that are (ideally) new to
the reader and purport to be true, one also can observe the techniques required to affirm them (in
a substantial sense). By studying a textbook, in short, one can acquire the techniques to formulate
a line of reasoning or a theory – which might even stand in contrast to ones presented in the text.
When it comes to the cognitive value of works of fictional literature, I want to suggest that
the same process is decisive. The cognitive value does not depend primarily on the information
that is contained in the text and can be communicated to the reader. It rather depends on the
ways in which authors and readers do engage with the text. To get a more adequate
understanding of the cognitive value of literature we should, thus, not primarily ask for the
message an author might have intended to communicate or for the propositions that are
contained in or emerge from the text; nor should we ask whether there is a particular hypothesis
that the text illustrates or demonstrates in form of a thought experiment. These questions are not
helpful if they are not embedded in a wider context that takes the social practice of literature into
account. We should, thus, not focus on the work, but on the social practice in which the work is
produced and received35. Only in this way we will get to understand why, in a given community,
literature matters and why people invest their time and energy to produce and to engage with
works of literary fiction.
The social practice of literature
In the last section I have invited to approach questions concerning the cognitive value of literature
from a new perspective that does not focus exclusively on the text, but takes into consideration a
wider context that includes the social practices in which literature plays a role. I have not,
however, differentiated between works that qualify as works of literary fiction and other sorts of
texts, such as scientific textbooks or newspaper articles. This raises the question of whether
35
I am thinking here of social practices and communities broadly along Wittgensteinian lines. It will become clearer in
the next pages that I am not inviting for an empirical sociological approach to literature – which does have its merits,
but treats the phenomenon of literature at a completely different level. Rather, I would like to underline that
literature is a form of practice, something we engage in, and that we do so together not privately, but in a praxis that
is shared with others and constituted by (mainly implicit) rules.
13
literature has a cognitive value sui generis, i.e., whether it can be a source of knowledge in virtue
of its distinctively literary features36.
I think we should be very cautious with claims along these lines. Literature is a multifarious
phenomenon, and it is not clear at all that we will be able to discern features that are distinctive
for all or most literary works of art. We could try, however, to individuate features of works of
literature in our relevant social practices which – though not shared by all or most works of
literature – can explain why we consider some individual works valuable sources of knowledge.
With a look at the current debate in philosophy of literature and without any claim to
completeness or universality, we could list the following points: works of literature typically aim at
arousing the readers’ imagination and focusing it on a specific topic that is of psychological, social,
historical, or political relevance. Often they have aesthetic properties or manifest stylistic
elements that catch and keep the readers’ attention. Many works play with the readers’
expectations and create, in this way, surprise and suspense. Longer works of narrative literature
often manage to make very detailed perspectives emerge. Finally, with their attention on literary
form and the careful choice of language, many works of literature invite for a critical engagement
with the text.
This list is, of course, incomplete and none of the features listed are shared by all or even
most works of literature. More importantly, these features are not distinctive for works of
literature. Many texts that are clearly not works of literature are designed to arouse the readers’
imagination – think of scientific thought experiments or commercials. The latter also try to catch
the reader’s attention with aesthetic or stylistic means and play with their expectations.
Biographies can make complex perspectives emerge and invite the reader to confront it with their
own. Philosophical essays often manifest a careful choice of language and typically invite the
reader for a critical engagement with the text.
This indicates that literature does not have a cognitive value sui generis; the knowledge
one can obtain from a good book could also be communicated by different means. If we perform
the change of perspective I am proposing, however, a different aspect emerges that sheds a new
light on the uniqueness of literature.
36
Sometimes it is suggested that the distinctive features of the work that count in this context are the ones that make
it a work of fiction (cf. Green 2017: 48). This seem reductive to me: there are literary works of art that can be a source
of knowledge in a way that depends crucially not on their being fictional, but on their distinctively literary features or
on their narrative structure, or on a combination of two or all three of these aspects.
14
Our social community, the culture we live in, basically depends on our succeeding in
coordinating our behaviour with that of others. Moreover, among all the rule-governed practices
we engage in, language is definitely the most complex and the most sophisticated. In the early 20th
century philosophers were convinced that the main function of language consists in the
transmission of true information. Thanks to the contributions of philosophers like Wittgenstein,
Sellars, Austin, and others, many have come to appreciate that language does much more than
that. Among other things, it makes it possible for us to exchange with other members of the
community and confront their perspectives on a shared environment with ours. This allows us not
only to fine-tine our own perspective, but also to calibrate it to that of the others. Unlike
textbooks or newspaper articles, works of literary fiction are not bound by the «fidelity
constraint»37, according to which an author is expected to include to the text «only events she
believes to have occurred, narrated as occurring in the order in which she believes them to have
occurred»38. We have, in consequence, a very different, playful use of language that allows for a
free play of imagination – an aspect that can be helpful to make a certain perspective on our
shared environment emerge in a more clear-cut manner – a perspective, though, that could be
expressed as well in a different manner.
What seems most noteworthy to me is that the reader is not a tabula rasa, but a person
who already has a perspective. The cognitive gain the work can provide, thus, will essentially
depend on how the reader engages with the work. Is she willing or able to pay attention to the
distinctively literary features of the text, or to the particular way in which the narrative unfolds? Is
she willing or able to confront the perspective that emerges from the text with her own? How will
she resolve a potential conflict between these two perspectives? These are open questions that
each reader has to address in her own way. The way she relates with the text will essentially
determine, however, whether and how the work can enrich her cognitive perspective.
In conclusion we can say that the traditional debate on the cognitive value of literature
risks instrumentalizing literature or reducing it to an ornament – because it aims at individuating a
single mechanism that can explain how texts can enrich the readers on a cognitive level. Against
this tendency I have argued that the cognitive value of literature depends not only on what is
contained in the text, but also on the importance that is attributed to literary texts in a social
community and on how readers engage with the text. If we want to understand how literary works
37
38
Cf. Davies 2007a, Davies 2015.
Davies 2007a: 46.
15
of art can be of cognitive value, we need to take the practice, in which it was created, and the one
in which it was read, into account. This means that we will not get a simple answer to the question
of whether we can learn from literature. We can, however, attune our sensitivities to the
phenomenon in question: literature and its place in our social community.
We must never forget that the practice of literature is very rich and has many different
faces and works of works of literature (and, more generally, works of fiction) are valued for many
different reasons. Sometimes we look for distraction or like to indulge in escapist fantasies, other
times we look for the aesthetic pleasures it can provide. Often we hope that a good read can
widen our intellectual horizons. What seems central to me is that our literary practice leaves space
for all these different uses. Moreover, it is a shared practice that constitutes or at least shapes
relevant aspects of our social community. At the beginning of the 21st century, we can (still) take it
for granted that the interesting works of literature are the product of the phantasy, the intellect,
and the artisanship of real human beings, i.e., persons who take part in this (or a similar) social
community (and not, say, the product of an anonymous algorithm computed by a form of artificial
intelligence). Engaging with a work of literature, thus, also means engaging with the expression of
another person or group of persons; it is always also an act of communication which, at least
sometimes, can serve to confront the others’ perspective on our shared environment with our
own.
References to works cited
BORGES, JORGE LUIS
– 2007, Pierre Menard, Author of the Quixote, in Labyrinths: Selected Stories & Other Writings,
translated by Donald A. Yates and James East Irby, New York, New Directions, pp. 36–44.
BRENDEL, ELKE
– 2013, Wissen, Berlin, De Gruyter.
BROWN, JAMES ROBERT
– 2011, The Laboratory of the Mind: Thought Experiments in the Natural Sciences. 2nd ed, New
York, Routledge.
16
BURRI, ALEX
– 2007, Art and the View from Nowhere, in John Gibson, Wolfgang Huemer, and Luca Pocci (eds.),
A Sense of the World: Essays on Fiction, Narrative, and Knowledge, New York and London,
Routledge: 308–17.
CARNAP, RUDOLF
– 1966, The Elimination of Metaphysics through Logical Analysis of Language, in Alfred J Ayer (ed.),
Logical Positivism», translated by Arthur Pap, New York, Simon and Schuster, pp. 60–81.
CARROLL, NOËL
– 2002, The Wheel of Virtue: Art, Literature, and Moral Knowledge. «The Journal of Aesthetics and
Art Criticism» 60 (1): 3–26.
DAVIES, DAVID
– 2007a, Aesthetics and Literature, London, Continuum.
– 2007b, Thought Experiments and Fictional Narratives, «Croatian Journal of Philosophy» 7 (1):
29–45.
– 2015, Fictive Utterance and the Fictionality of Narratives and Works, «The British Journal of
Aesthetics» 55 (1): 39–55.
EGAN, DAVID
– 2016, Literature and Thought Experiments, «The Journal of Aesthetics and Art Criticism» 74 (2):
139–150.
ELGIN, CATHERINE Z.
– 2002, Art in the Advancement of Understanding, «American Philosophical Quarterly» 39 (1): 1–
12.
– 2007, The Laboratory of the Mind, in John Gibson, Wolfgang Huemer, and Luca Pocci (eds.) A
Sense of the World: Essays on Fiction, Narrative, and Knowledge, New York, London:
Routledge: 43–54.
– 2014, Fiction as Thought Experiment, «Perspectives on Science» 22 (2): 221–41.
GIBSON, JOHN
– 2003, Between Truth and Triviality, «The British Journal of Aesthetics» 43 (3): 224–37.
17
– 2007, Fiction and the Weave of Life, Oxford, Oxford University Press.
GOODMAN, NELSON
– 1978, Ways of Worldmaking, Indianapolis, Hackett.
GREEN, MITCHELL
– 2017, Narrative Fiction as a Source of Knowledge, in Paula Olmos (ed.), Narration as Argument,
Cham, Springer International Publishing: 47–61.
HUEMER, WOLFGANG
– 2004, Introduction: Wittgenstein, Language, Philosophy of Literature, in John Gibson and
Wolfgang Huemer (eds.) The Literary Wittgenstein, London, Routledge: 1–13.
– 2007, Why Read Literature? The Cognitive Function of Form, in John Gibson, Wolfgang Huemer,
and Luca Pocci (eds.), A Sense of the World: Essays on Fiction, Narrative, and Knowledge,
New York and London, Routledge: 233–45.
– 2017, La Filosofia della letteratura: un pregiudizio a favore della finzione, in Guido Ferraro and
Antonio Santangelo (eds.), Finzione e realtà. Il senso degli eventi, Roma, Aracne: 137–54.
HUME, DAVID
– 1978, A Treatise of Human Nature. 2d ed. Oxford, Clarendon Press.
JACKSON, FRANK
– 1982, Epiphenomenal Qualia, «The Philosophical Quarterly» 32 (127): 127–36.
– 1986, What Mary Didn’t Know, «The Journal of Philosophy» 83 (5): 291–95.
JOHN, EILEEN
– 2016, Theme, in Noël Carroll and John Gibson (eds.) The Routledge Companion to Philosophy of
Literature, New York, Routledge: 205–16.
KIVY, PETER
– 2006, The Performance of Reading: An Essay in the Philosophy of Literature. Malden, MA,
Blackwell.
LAMARQUE, PETER
18
– 2009, The Philosophy of Literature, Malden, MA, Blackwell.
NAGEL, THOMAS
– 1974, What Is It Like to Be a Bat? «The Philosophical Review» 83 (4): 435–50.
NORTON, JOHN D
– 2004, Why Thought Experiments do not Transcend Empiricism, in Christopher Hitchcock (ed.)
Contemporary Debates in the Philosophy of Science, Malden, MA, Blackwell: 44–66.
NUSSBAUM, MARTHA CRAVEN
– 1990, Love’s Knowledge: Essays on Philosophy and Literature, New York: Oxford University Press.
PLATO
– Republic, Trans. by Benjamin Jowett. http://classics.mit.edu/Plato/republic.html (April 3, 2018).
RORTY, RICHARD
– 1991, Heidegger, Kundera, and Dickens, In Essays on Heidegger and Others: Philosophical Papers,
Cambridge, Cambridge University Press: 66–82.
RUSSELL, BERTRAND
– 1940, An Inquiry into Meaning and Truth, London, George Allen & Unwin.
STOLNITZ, JEROME
– 1992, On the Cognitive Triviality of Art, «The British Journal of Aesthetics» 32 (3): 191–200.
VENDRELL FERRAN, ÍNGRID
– 2018, Die Vielfalt der Erkenntnis. Eine Analyse des kognitiven Werts der Literatur, Paderborn:
mentis.
WALSH, DOROTHY
– 1943, The Cognitive Content of Art, «The Philosophical Review» 52 (5): 433–51.
WITTGENSTEIN, LUDWIG
– 2009, Philosophical Investigations, edited by P. M. S. Hacker and Joachim Schulte, translated by
G.E.M. Anscombe, P. M. S. Hacker, and Joachim Schulte. Rev. 4th ed. Malden, MA:
Blackwell.
19