Literature as Speech-Act

I have decided not to apologize for this post as I did for the previous one. So what if I want to write abstrusely about epistemology and ontology? The discussion may be difficult but I believe what I have to say about fiction and truth is both notable and progressive. The theory that I am proposing is, I believe, quite a radical departure from previous theories of literature and I want to hope that some reader will stumble over it accidentally, understand it, enjoy it and recommend it to their friends. This does not seem, I hope, too immodest an ambition.

The last few essays, beginning with the post “Applying Predicate Calculus to Literature”, have been concerned with approaching literature from the position of Formal Logic. In today’s post I want to take this discussion a little further and tie up some loose ends. In particular, I would like to relate my project to John Searle’s theory of Speech-Acts. In passing, I will be saying something about the real world as well as about fiction.

Any literary text can be decomposed into a broad set of simple statements or propositions, propositions that the text makes either explicitly or implicitly. For example, the novel Ulysses can be broken down into ‘facts’ such as the statement “Leopold Bloom is married to Molly Bloom” or the statement “Bloom carries a potato in his pocket wherever he goes”. The number of statements asserted by a literary work can be enormous, particularly in the case of a novel as sprawling as Ulysses. Literary interpretation is tasked with the job of reproducing these statements in a different format, particularly those statements that are made implicitly rather than explicitly. (Analysis also, of course, is engaged with the task of finding relationships between these statements, a process we can describe as structural analysis.) One can make an important distinction between two different types of proposition. Some propositions refer to our shared reality and some refer only to the fictional world represented by the story. For example, on the 16^th of June 1904, the horse Throwaway won the Dublin Gold Cup, an event to which Ulysses indirectly refers. Consequently, one proposition that Ulysses asserts is “Throwaway won the Gold Cup”. This statement is open to historical scrutiny with respect to its truth or falsity. The statement “Bantam Lyons won a lot of money by betting on Throwaway” is, however, not a statement susceptible to independent verification; Bantam Lyons is a fictional character who exists only in Joyce’s Bloom-world and not in the real one. Our only source for information about him is the novel itself.

 The most important non-fictional propositions a story makes are the ‘thesis’ and the ‘antithesis’. I have spent quite a bit of time talking about these concepts in earlier posts, perhaps most clearly in the post “Applying Predicate Calculus to Literature”. In today’s post, though, I want to talk some more about the fictional propositions a text makes rather than its real-world relevant ones. How do we assess the truth or falsity of a statement drawn from Star Wars like “Luke Skywalker is Princess Leia’s brother”? In the previous post “Meongianism and the Phenomenology of Knowledge” I argued that these statements could be usefully understood as true statements concerning fictional worlds. Tolkien, I argued, had privileged access to a fictional world called Middle-Earth and faithfully in his stories transcribed facts about it, such as “Gimli is a dwarf” and “Frodo is a hobbit”. From this perspective, all the ideas in The Lord of the Rings are true statements concerning the fictional world to which he had access. We believe these ‘facts’, not because we can objectively, empirically and independently test them, but because we put our faith in Tolkien as an absolute authority on the history and inhabitants of Middle-Earth.

This construal has its advantages. It is helpful with respect to a number of problems concerning the status of fictional statements. Moreover it sheds light on an interesting aspect of literary criticism: stories are usually written in the past perfect but criticism tends to be written in the present tense. This is an indication that critics tacitly subscribe to the idea that fictions are windows on other worlds. However, one can wonder if this proposal is really feasible. Surely Tolkien invented Middle-Earth rather than discovering it? The rest of this post will be concerned with an alternative perspective on this issue, which employs ideas attributable to the philosopher John Searle.

I don’t have Searle’s works on hand and so I am working from memory. Searle, as I remember, argued that many types of language-use are not representational or referential propositions but are, rather, speech-acts that effect changes on the world. For example, the speech-act “By the power invested in me, I now pronounce you man and wife” is a statement that, when said by a cleric or justice of the peace, alters the world. The statement is, itself, neither true or false but (when said by someone with the right kind of authority) it creates new conditions of future verifiability. Thereafter, if I say “Sam is Mary’s husband” or “Mary is Sam’s wife” I am making true statements. The speech-act has changed the world. Another such speech-act is the phrase “You are under arrest” ­– an order that only has effective power if said by a member of the police.

Although I don’t think Searle talks about this as much, there is a class of speech-act that is highly salient to the discussion– definitions. Suppose I say, “I define the term ‘psueoJedi’ as designating that class of people who identify themselves as Jedis on census reports but do so ironically or facetiously” and supposing that this term takes off, people may, in the future, be capable of making true statements such as “I call myself a Jedi but really I’m a pseudoJedi”. My original definition of the term, though, is not itself either true or false. A real world example of a definition type speech-act is furnished by the ‘discovery’, in the late nineteenth century by Arnold Pick and Emil Kraeplin, of a new disease that Pick called dementia praecox and which later came to be called schizophrenia. In coining a label for a collection of symptoms that he had observed, Pick redefined reality. What had formerly been loosely considered a type of madness would now be categorized as a physical illness. By coining this term, they changed the world.

A literary work can be viewed as an extended speech-act – and the type of speech-act it most closely resembles is a definition. The original definition is, from this perspective, neither true nor false, but it creates the conditions for future verifiability. The meaning of the word ‘hobbit’ as established by Tolkien includes the idea that hobbits walk bare-footed and have hairy feet. If I say “Frodo has hairy feet”, I am not only making a true statement, I am making a statement that is analytically true. (I touched on this idea in the post “Ascribing Attributes to Imaginary Entities”). According to Kant, analytic truths are known a priori. I think this is wrong. The mistake Kant and very many other philosophers make is that they forget that language is acquired. When I was very young and was basing my understanding of the world largely on The Tale of Mr Jeremy Fisher, I thought all fish were called ‘minnows’. Words are learned. Most garden-variety words are learned via ostensible definition: for example, we learn what the word ‘apple’ means by being shown one and being told the word. Later, we extend the term to all other objects that resemble this original example. Other words, though, obtain their meanings from some authority that has, at some point, defined the word (using other words) and has had his or her definition accepted by the relevant linguistic community. The meanings of words always rest on some authority and, critically, can change if the community decides to put its faith in a different authority.

I hope I am expressing myself clearly. I may have to write about these ideas again some time. I should say though that I have thought about these issues for years and, as proof of this, thought I would finish this essay with an anecdote. In around 2004, I wrote a number of Research essays on the poet John Ashbery; I decided to write about him quite arbitrarily, mainly because he was very influential on the post-modern poetic scene. I set myself the task of reading and interpreting his incredible long and difficult autobiographical poem Flow Chart. Having read it twice, I was forced to conclude that Ashbery must be gay and that this poem was his ‘coming-out’. Coming out as gay is a pre-eminent example of a speech-act, a verbal utterance that changes the world, and this was the gist of my essay about Ashbery’s poem, that it was a ‘presentation’ rather than a ‘representation’. This was years before I had read John Searle. Even then, though, I believed in speech-acts. These days I have wondered if I made a mistake in the essay I wrote about Ashbery. Perhaps he was attempting that most bizarre of actions, coming out as straight. The world is a strange place and such things are not impossible.