What follows is a guest post by Catharine Abell. Catharine is Senior Lecturer in Philosophy at The University of Manchester. She has published work on various topics in aesthetics, including pictorial representation, the nature of art, expression in art, and genre. She is currently trying to develop solutions to a range of philosophical problems posed by fiction.
In this post, I want to discuss how we should understand the content of works of fiction. Their content is philosophically puzzling because, to understand a work of fiction, one must usually do more than just grasp its literal content. A fiction may implicitly convey, rather than explicitly represent, some aspects of its content. Let us call the literal content of a work of fiction its explicit representational content. Let us call the content of the story told by a work of fiction “what is true according to the fiction”. What is true according to a fiction can outstrip its explicit representational content and can also exclude certain aspects of its explicit representational content. For example, in The High Window, Raymond Chandler writes, “there are ratty hotels where nobody except people named Smith and Jones sign the register and where the night clerk is half watchdog and half pander.” Here, what’s true according to the fiction outstrips explicit representational content, because it is true according to the fiction that the hotel is used to conduct illicit affairs, although this is not part of its explicit representational content. What’s true in a fiction also excludes aspects of its explicit representational content: it is part of the novel’s explicit representational content that the only people who sign the hotel register just happen to be named Smith or Jones, although this isn’t true according to the fiction. The philosophical challenge is to provide an account of truth in fiction that is able to explain how it differs from explicit representational content.
Philosophers generally respond to this challenge with accounts according to which truth in fiction is very broad in scope. Take, for example, the accounts proposed by David Lewis. Lewis provides two alternative characterisations of truth in fiction. On the first, what’s true according to a fiction incorporates everything that would be true, were that fiction told as known fact, rather than as fiction (Lewis cashes out what would be true in such a situation by appeal to his possible worlds account of modality, but the details of this analysis need not concern us here) (Lewis 1983a). As Lewis acknowledges, this account has the unpalatable consequence that what’s true according to a fiction may include facts about the world unknown to either author or audience. For example, it will be part of the narrative content of Jane Austen’s Emma that the invention of the telephone is less than a century away because, were the events recounted in Emma told as known fact, the telephone would be invented some time in the coming century. It is anachronistic, though, to suppose that telephones feature in the content of Regency novels.
To avert this problem, Lewis offers an alternative account, on which what’s true according to a fiction incorporates whatever would be true, according to the overt beliefs of its maker’s community, were it told as known fact (Lewis 1983a: 273). A community’s overt beliefs are those beliefs which almost all of its members hold, believe one another to hold, and believe one another to believe one another to hold, and so on. On this construal, what’s true according to Emma excludes future innovations in telecommunication, because Jane Austen’s community held no overt beliefs about them. The account still construes truth in fiction very broadly, however, because it construes things overtly believed by Jane Austen’s community, but apparently irrelevant to the story – such as the basic tenets of Christianity, or that broccoli is nutritious – as true according to Emma.
Moreover, as Kendall Walton notes, Lewis’s accounts give rise to various “silly questions” (Walton 1990: 174-83). Consider the following lines, spoken by Othello:
Had it pleased heaven
To try me with affliction; had they rain’d
All kinds of sores and shames on my bare head.
Steep’d me in poverty to the very lips,
Given to captivity me and my utmost hopes,
I should have found in some place of my soul
A drop of patience: but, alas… (Act 2, Scene 2)
If Othello uttered the words above, it would be true (both in reality and according to the overt beliefs of Shakespeare’s community) that he uttered superb verse. Walton asks: “How did Othello, a Moorish general and hardly an intellectual, manage to come up with such superb verse on the spur of the moment, and when immensely distraught? Apparently he is to be credited with an almost unbelievable natural literary flair”(Walton 1990: 175). This question is silly because it is not one audiences ask, nor one to which Othello supplies an answer. This poses a problem for Lewis’s accounts, because they cannot explain the interpretative illegitimacy of asking this question.
Many works of fiction raise silly questions: narrators in novels often remember in great detail events that occurred many years ago (why do they have such excellent memories?); characters in novels often speak in complete, grammatical sentences (why are they so articulate?); and English novels set in countries in which English is not widely spoken often represent all their characters as speaking fluently in English (where did they get their language skills?). In all these cases, were things as they are represented as being, things would be true (either in reality, or according to the overt beliefs of the author’s community) that are not true according to the works.
I think the silliness of these questions shows that truth in fiction isn’t nearly as broad in scope as philosophers commonly take it to be. J. L. Austin distinguished three different kinds of act that one may perform in making a spoken or written utterance. Locutionary acts are acts of speaking or writing certain words with certain senses and references (Austin 1976: 94). What locutionary act one performs is determined by the rules of the language in which one speaks or writes. By contrast, illocutionary acts are acts that one performs in speaking or writing words with a certain sense and reference (Austin 1976: 98-9). In performing the locutionary act of uttering the words “the cat sat on the mat”, one may also perform the illocutionary act of asserting that the cat sat on the mat. Similarly, in uttering “Would you like to come to the dance with me?”, I may perform the illocutionary act of inviting you to the dance. Finally, perlocutionary acts are acts one performs by performing a certain locutionary act (Austin 1976: 101). They concern the consequences or effects of speaking or writing. For example, by uttering the words “Beluga is the only caviar worth eating” I may either impress you with my discernment or repulse you with my pretentiousness.
I propose that what’s true according to a fiction is given by the contents of the illocutionary acts authors perform in producing works of fiction. Illocutionary acts have both force (e.g. that of assertion, invitation, or warning) and content (i.e. what one asserts, invites someone to do, or warns against). Because it restricts what’s true according to a fiction to the contents of authors’ illocutionary acts, this account does not construe narrative content as incorporating either features of the world unknown to author or audience, or facts known to author and audience, but seemingly irrelevant to the story itself.
Nevertheless, this account can accommodate the fact that truth in fiction can outstrip explicit representational content because the content of an illocutionary act can be non-literally conveyed. For example, I can invite you to the dance by saying “Come and kick up your heels”. Similarly, when Chandler writes “there are ratty hotels where nobody except people named Smith and Jones sign the register” he performs, by nonliteral means, an illocutionary act part of the content of which is that the people who sign the register are engaged in illicit activities.
Truth in fiction excludes certain aspects of explicit representational content because a work’s explicit representational content helps to determine what is true in the work only when it is used to perform an illocutionary act. Explicit representational content that is not used as a means of doing so is irrelevant to truth in fiction. Illocutionary acts are not the only acts that fiction makers perform in producing their works. Often, in addition to issuing invitations, authors seek to affect their audiences in various ways. Authors of literary works often aim to provide their audience with aesthetic pleasure, and thriller writers often aim to scare their audiences. When a work’s explicit representational content is used exclusively as a means of performing perlocutionary acts, and does not contribute to the performance of any illocutionary act, it does not contribute to what’s true according to the fiction.
Authors often represent narrators as remembering in great detail events that occurred many years ago in order to perform the perlocutionary act of providing readers with insight into their histories. Similarly, they often represent characters in novels as speaking in complete, grammatical sentences in order to perform the perlocutionary act of making their conversations intelligible to readers. English language novelists often represent characters in non-English speaking places as speaking fluent English so as to perform the perlocutionary act of enabling their readers to understand what those characters say. Because they are used in the performance of perlocutionary acts but not of illocutionary acts, these aspects of the works’ explicit representational contents do not contribute to what’s true according to those works.
It is not true in Othello that Othello is a poet of the highest order because, in representing Othello as uttering the words he does, Shakespeare’s purpose was to provide the audience with aesthetic pleasure. He did not perform an illocutionary act with the content that Othello uttered precisely the words he is explicitly represented as uttering. By explicitly representing characters as making certain utterances, authors may either fail to perform any illocutionary act at all, or instead perform illocutionary acts the content of which is that those characters make, not those specific utterances, but other utterances with roughly the same meaning. In the latter case, truth in fiction is indeterminate: it is true according to the works that the characters make utterances with certain approximate meanings, but not true according to the works that they make specific utterances with those meanings. This explains why characters in English language novels set in non-English-speaking countries may, without contradiction, be explicitly represented as saying “I don’t speak any English”: the content of the author’s illocutionary act is that they utter words, in a foreign language, with the same meaning as the English sentence by which their utterance is explicitly represented.
As the critic James Wood writes:
who really thinks that it is Leopold Bloom, in the midst of his stream-of-consciousness, who notices ‘the flabby gush of porter’ as it is poured into a drain, or appreciates ‘the buzzing prongs’ of a fork in a restaurant – and in such fine words? These exquisite perceptions and beautifully precise phrases are Joyce’s, and the reader has to make a treaty, whereby we accept that Bloom will sometimes sound like Bloom and sometimes sound more like Joyce.
This is as old as literature: Shakespeare’s characters sound like themselves and always like Shakespeare, too. It is not really Cornwall who wonderfully calls Gloucester’s eye a ‘vile jelly’ before he rips it out – though Cornwall speaks the words – but Shakespeare, who has provided the phrase (Wood 2008: 25).
Characters sound like themselves only when, by representing them as making certain utterances, authors perform illocutionary acts the content of which is that those characters made exactly those utterances. In As You Like It, Shakespeare explicitly represents Jaques as addressing to Orlando the words “Nay then, goodbye an you talk in blank verse” (4.1.29). By doing so, Shakespeare performs an illocutionary act the content of which is that Orlando actually speaks in the blank verse in which he is represented as speaking. This passage is funny precisely because Shakespeare generally explicitly represents his characters as speaking in verse, but only in this case performs an illocutionary act which ascribes this property to them.
A work’s explicit representational content can be essential to appreciating the work even when it does not figure in what’s true according to the fiction. While none of the illocutionary acts Shakespeare performed in producing Othello required him to represent Othello as uttering precisely the words he is represented as uttering, he would not have provided readers with the aesthetic pleasure he sought to elicit if he had not done so. Small changes to the work’s explicit representational content may not affect what’s true according to the work, but would adversely affect the aesthetic pleasure we derive from it.
It is an interesting question what kind of illocutionary act authors perform in producing works of fiction. Some philosophers deny that they actually perform any illocutionary acts, but claim instead that they merely pretend to do so, on the basis that works of fiction are generally comprised of sentences with the grammatical form of assertions, but it is implausible that, in writing those works, authors perform the illocutionary act of assertion because they do not generally believe what they say, although we do not take them to be guilty of lying (Searle 1979). Gregory Currie claims that authors of fictions issue invitations to make believe (Currie 1990). I agree with Currie that fiction making involves the performance of illocutionary acts, but disagree with his account of the type of act involved. But that, as they say, is a topic for another day.
Austin, J. L. (1976). How To Do Things With Words. Oxford, Oxford University Press.
Currie, G. (1990). The Nature of Fiction. Cambridge, Cambridge University Press.
Lewis, D. (1983). Truth in Fiction. Philosophical Papers. Oxford Oxford University Press. 1:261-275.
Searle, J. (1979). The Logical Status of Fictional Discourse. Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge, Cambridge University Press: 58-75.
Walton, K. (1990). Mimesis as Make-Believe. Cambridge, MA, Harvard University Press.
Wood, J. (2008). How Fiction Works. London, Jonathan Cape.
May 16, 2014 at 4:09 pm
Thanks for your wonderful post. I just wanted to clarify something (and then ask about something else). Are you contending that truth in a fiction is *just* given by the contents of its (author's) illocutionary acts?
I ask for two reasons: (1) because it seems that the author's beliefs and her selection of an audience still play an important role in your analysis (as they should, if we're to get implicit truths into the picture!)—or is it that you think we can get these (the author + community's overt beliefs) “for free” somehow?, and (2) because it actually seems like a plausible addendum to/extension of Lewis's analysis (and probably of the make-believe and fictive utterance models too), since presumably the intended community's overt beliefs will include beliefs about all kinds of narrative and linguistic norms and conventions.
In the second instance, I was hoping you might say a little more about how you think that an analysis of truth in fiction as the contents of the author's illocutionary acts would deal with some of the standard problem cases. I have in mind especially cases of accidental inconsistencies on the author's part (e.g. Watson's wandering wound, or his changing first name), and of deliberate inconsistency (e.g. square circles, time travel, Graham Priest's “Sylvan's Box”). Would the result just be that what's true in the story is indeterminate in all of those cases?
May 17, 2014 at 6:53 am
Thanks for your reply. Yes, I am claiming that truth in a fiction is *just* given by the content's of the illocutionary acts its author performs in making it (call these “fiction making illocutionary acts”).
1) Because fiction making illocutionary acts are performed in the author's (and her community's) actual context, that context provides the background beliefs that help determine the content of her fiction making illocutionary acts. That is, just as, when I say to you, “the early bird catches the worm”, you can appeal to beliefs and conventions prevalent in the context to work out that I am asserting something about people who get up early, not about worms, so too readers can appeal to beliefs and conventions prevalent in the author's community to interpret her nonliteral fiction making acts. What I deny, however, and what distinguishes my position sharply from Lewis's, is that the background beliefs to which readers are able to appeal to interpret authors' fiction making acts are actually part of what's true in a fiction. That's why, on my account, truth in fiction is much narrower in scope than on Lewis's.
2) Some standard problem cases just aren't problems for me. For example, I am perfectly happy with deliberate inconsistencies in fiction: there's nothing to prevent an author from performing fiction making acts with inconsistent contents. (Note, also, that I don't have the problems Lewis does with stories that lack a narrator, since I can readily explain such cases as cases in which it's not part of the content of the author's fiction making acts that the story has a narrator). Accidental inconsistencies similarly result in fictions that do have inconsistent contents. However, in these cases, we exercise a principle of charity and decide that the author didn't really mean to attribute to Watson a wandering wound – much as if, when I contradict myself in some fairly innocent way, charitable interlocutors won't take me up on the inconsistency.