Lewis' "Truth in Fiction" and Literary Possibilities

In his 1978 work, "Truth in Fiction," David Lewis addresses the problem of determining the truth value of statements in a work of fiction by using a system of possible worlds. This may seem a strange step, but it is a logical result of Lewis' prior work with modal logic, where he established his own extensive and coherent theory of how possible worlds operate. Upon consideration of the details of his possible worlds theory, his resulting work with fiction seems only natural: as Lewis believes that there are an infinite number of possible worlds, it follows that the events described in a story - in every story but logically impossible ones - must happen somewhere in the plurality of worlds. Lewis defines a possible world, as do others, as a maximal set of consistent properties. Lewis differs from most other theorists in that he holds that every possible physically, or concretely, exists just as our own does; the only reason we call our world the actual world is that we happen to live in it. Most other possible world theorists believe that possible worlds are abstract, and Lewis' belief in the physicality of worlds will have some impact on the application of his "Truth in Fiction" theories to literary criticism.

The question Lewis begins with is, how can we hold that certain things are true about fictional characters - say, to imitate Lewis, that Sherlock Holmes is a cocaine user - when these characters don't even exist? The easy answer, and the one he addresses first, is that we automatically assume that statements about fictions are implicitly prefixed by the phrase, "In the fiction f . . ." e.g., "In the fictions written by Sir Arthur Conan Doyle about Sherlock Holmes, Sherlock Holmes is a cocaine user." Lewis uses his theory of possible worlds to modify this statement, saying there are an infinite number worlds where "the fiction f" occurs, and that one way to judge truth value is by "Analysis 0: A sentence of the form 'In fiction f, is true iff is true at every world where f is told as known fact rather than fiction"(268). This use of the theory gives us justification for asserting the truth of everything explicitly stated in a story, but nothing else. That is, we cannot assume the truth or falsity of anything not exlicitly mentioned, even something so apparently innocuous as assuming that Sherlock Holmes has a navel or wears underwear.

Lewis finds this to be an unsatisfying conclusion - largely for personal rather than theoretical reasons - and so proceeds to Analysis 1, which states instead that "A sentence of the form 'In the fiction f, is non-vacuously true iff some world where f is told as known fact and is true differs less from our actual world, on balance, than does any other world where f is told as known fact and is not true. It is vacuously true iff there are no possible worlds where f is told as known fact"(270). We need only concern ourselves with the "non-vacuously true" section of this analysis. Analysis 1 allows us to make the kind of assumptions about fiction that most of us intuitively wish to make, ones that let us posit fully formed worlds as close as possible to our own, which contain the events detailed in the story. Under this analysis, we can not only assume that Holmes has a navel, but that his London is similar to our own in important respects, such as street and building locations, social attitudes, etc.

Before it becomes Lewis' final judgment, this analysis undergoes one final adjustment, one which takes into account our presumed reasons for wanting to be able to include information from our actual world in a story. The "problem" with Analysis 1 is that it allows authors to be wrong in what Lewis judges to be inappropriate ways. The example Lewis uses is from an article by Carl Gans about the movement of snakes:

In "The Adventure of the Speckled Band," Sherlock Holmes solves a murder mystery by showing that the victim has been killed by a Russell's viper that has climbed up a bell-rope. What Holmes did not realize is that Russell's viper is not a constrictor. The snake is therefore incapable of concertina movement and could not have climbed the rope. Either the snake reached its victim some other way or the case remains open. (271)

In this case, making the world of the story as close as possible to our own means that an important part of the Sherlock Holmes canon has been violated: namely, that Holmes has solved all of his cases.

Unsatisfied with this effect, Lewis' final analysis disallows this sort of error. Analysis 2 reads: "A sentence of the form 'In the fiction f, ' is non-vacuously true iff, whenever w is one of the collective belief worlds of the community of origin f, then some world where f is told as known fact and is true differs less from the world w, on balance, than does any world where f is told as known fact and is not true. It is vacuously true iff there are no possible worlds where f is told as known fact"(273). This sounds much more complicated than it is. It means that where we run into difficulties by describing the non-specified world of the story as that which most closely matches our own, the implicit truths of a story are defined instead as how its author and the belief community surrounding the author overtly believe the actual world to be. In this way Lewis ensures, as he wishes, that the truth of a story cannot change: it is set when its author first tells it to his or her intended audience.

This may seem like a lot of work to accomplish what many would presume to be common-sense material. Why do we claim that people in novels have feet, even if their feet are not specifically mentioned? Well, because that's what the author intended - and it's just silly to assume anything else; like Ockham's fabled razor pronounces, the simplest course is probably the correct one. Lewis' theory does something special, however. Because he believes that possible worlds are concrete, they must be maximal. There can be no physical world where Holmes' feet do not exist simply because he never removes his shoes. Because these are the places where stories are occuring, we are justified in stating that there is a world outside the story. In the Sherlock Holmes stories, how many Baker Street Irregulars are there? We don't know, exactly - but we can confidently state that we do know there must be an exact number of them, because they are physical people in a maximal world. Ultimately, they must be countable. Lewis covers new ground not necessarily by making claims about how we treat fictions, but why we have valid philosophical grounds for doing so.

In "Truth in Fiction," Lewis does not seem to consciously claim any connection to a specifically literary theory, and it is somewhat difficult to place him definitively in a category because his article refrains from commenting on the symbolic meaning or aesthetic value of works of fiction. However, his assumptions about the literal meaning of the work clearly imply that he supports an author-centered approach. All of the "intuitions" he references in support of his theory - such as his assertion that we desire to claim more than is explicitly stated or that authors cannot be wrong - seem to stem from some sort of pre-theoretical tendencies towards an author-centered, historically conscious theory. His final analysis is one that would seem to lend itself well to use by critics in this field. The two may not, however, be entirely compatible.

Called, variously, "historical-biographical," "historical author-centered," and "historical genetic," as well as other labels, this seems to be the kind of criticism that Lewis seems to be "intuitively" based in, although the article in question is too early for the New Historicism focus of literary studies of the early eighties. In this historical-biographical - to pick a term - field of thought, ultimate meaning comes from the author. It is not always possible to directly question an author as to the meaning of their work, however, so historical research as to social beliefs and practices is a major part of this criticism. Based as it often is on a sort of exaltation of the author as a superior genius, the foundations of historical-biographical criticism have been repeatedly attacked. Two notable instances are Wimsatt and Beardsley's 1954 New Criticism work, "The Intentional Fallacy," and Barthes' poststructuralist "The Death of the Author" in 1977, published one year before Lewis' "Truth in Fiction." It should not seem odd, however, that Lewis' essay does not seem to recognize the ideologies which claimed to preclude historical-biographical criticism, or the literary pressures which influenced Barthes and others. Lewis was far more concerned with the philosophical ramifications of his theory, as his "concretist" view of possible worlds was then and continues to be a difficult sticking point for most modal logicians.

Lewis' tacit assumption of the validity of historical-biographical criticism is not without validity. Although numerous challenges have been brought against it, it remains a common, basic form of literary criticism, especially in the form of New Historicism. His Analysis 2, because of his ideological approach, serves as a theory that many historical-biographical critics could use as justification for their own assumptions about what is true within a fiction. Both turn to the same source for their information - the historical community surrounding an author at the time of that author's writings. Where critics like H. A. Taine and E. D. Hirsch use their own theoretical basis for drawing symbolic meaning and social implications from the historical community, Lewis has drawn tacitly on this same theory to provide an avenue for drawing literal truth from this same community. Critics of other schools could certainly be successful in incorporating Lewis' Analysis 0 or Analysis 1 in their own fields of work, should they find the implications of either to be appropriate - Lewis himself suggests the use of Analysis 1 by psychoanalytic critics, and offers his Analysis 2 as an alternative only because he prefers it, not because he can provide better-supported philosophical ground beneath it. However, the essay as a whole, and its final analysis, are so well fitted to a historical-biographical approach that it seems well suited for appropriation by that school of thought.

There are, however, some philosophical and literary consequences of adopting Lewis' theory that would need to be resolved by individual critics first. Lewis' theory of possible worlds is one that demands certain ideological sacrifices many may not be willing to make. Some of these have been commented on at length by other modal logicians, and are better left to those discussions. Far more problematic for historical-biographical critics are the literary implications. As mentioned before, Lewis is emphatically not writing literary theory, but a philosophical justification for speaking of truth in fiction. As such, he does not present any theory of how to read texts for meaning or worth. Rather than two similar literary theories, we have instead two systems designed for two different types of evaluation.

The most troublesome conflict between Lewis' system and historical-biographical criticism has to do with the repercussions of applying Analysis 2 to texts. To accept Analysis 2, critics must give up privileged insights into an author's meaning. To partially appropriate a Norman Holland and Stanley Fish example, let us imagine that secret notes of William Faulkner's are found, declaring that all the characters in "A Rose for Emily" are secretly Eskimos. Faulkner was so embarrassed about his fascination with Eskimos that he intentionally left out any clues as to the characters' true natures, but he intended the story to be, at heart, one of transplanted Eskimos in the American South. Lewis' theory of truth in fiction as presented in Analysis 2, however, states that the truth of a story must depend on an overt belief of an author and his or her intended audience. Because of this, an author's private intentions are no longer an important consideration when determining truth. We cannot claim that everyone in the story is really an Eskimo, because Faulkner did not intend his audience to realize this on their own, carefully covering his tracks, and because his audience did not share a common belief about large groups of Eskimos living in the American South. For some critics, this may be an unacceptable conclusion. For others, it may be a valid departure from standard historical-biographical practice - if they are more concerned about the work as something which reflects and has an impact on society, rather than a personal revelation of the author, it is not unlikely that they could seize Lewis' work as an opportunity to focus their criticism appropriately.

David Lewis' "Truth in Fiction," as part of his highly consistent philosophical stance, brings much with it as a complement of literary criticism. Much of it, most importantly philosophical justification for assuming implied facts in fiction, can be utilized by several theories, and its final analysis of its subject seems, at first, particularly beneficial to historical-biographical criticism. The restrictions it places on privileged knowledge of an author's intention are indeed problematic for this approach, but do not necessarily preclude the two from being used together. Perhaps, while arriving too early to be in the mainstream of New Historicism, Lewis' paper unwittingly presents a theory that, if successfully brought to the attention of that field of criticism, could provide a new link between philosophical theory and the theory of literature.

Works Cited

Lewis, David. "Truth in Fiction,"Philosophical Papers, Volume 1, Oxford University Press, 1983, pp. 261-275. Originally published in American Philosophical Quarterly 15 (1978) 37-46.

HOME Back to English essay directory

Nobody comes here at all. You're only number Site 
Meter to discover this hidden vale of wonder.