The Epistemology of Fiction: Knowing v. “Knowing”

Abstract

Alan Palmer reports that in his experience understand fictional minds much better when we apply to them some of the work done on real minds by psychologists, philosophers, and cognitive scientists (Cognitive Approaches'). Undoubtedly his experience is shared by many readers, as is its corollary, that understanding fictional minds helps us understand real minds. But what we can about fictional minds is different from what we can about real minds. Palmer calls it cliche of literary studies that ... novels can give us direct access [whereas] we can never really what other people are thinking. suggest that he rejects too rapidly the epistemological difference between fiction and nonfiction and thereby loses some of the pleasure that readers of fiction experience. To my knowledge, Lubomir Dolezel is the first to adapt J. L. Austin's analysis of what we can with words to explain the epistemological status of fiction. Austin, of course, distinguishes between constative statements, which may be true or false, and performative statements (such as the I do of the wedding ceremony), which, when said by an authorized person under appropriate circumstances, make something happen. A performative statement brings about a state of affairs we can know: the person becomes married. response to a constative statement, in contrast, at best we may think that we know whether it is true or false. For Dolezel, the speech act of the anonymous third-person (heterodiegetic) narrator of fiction carries an authentication authority ... that is analogous to the force of performative speech acts described by Austin (11). Like the spoken I do that makes the speaker married, a heterodiegetic narrator's performative statements make the fictional world by describing it. The fictional world is what the heterodiegetic narrator says it is. Even earlier, Kate Hamburger, in The Logic of Literature (initially published in German in 1957), draws attention to the necessary correlation between the epistemological status of third-person fiction and the representation of characters' minds. (1) Developing Hamburger's ideas, Dorrit Cohn recognizes fiction's unique potential for presenting characters' minds: In fiction cast in the third person, this presentation involves a distinctive epistemology that allows a narrator to what cannot be known in the real world ... the inner life of his figures (16). For Gerard Genette, the relation between the representation of characters' minds and the performative nature of fiction is necessary because causal. Noting that the features of fiction that Hamburger describes refer to ... direct access to the character's subjectivity, Genette says that features are only effects produced by the fictional nature of the narrative ... If narrative fiction alone gives us direct access to the subjectivity of another person, this is not by virtue of some miraculous privilege: it is because that other person is a fictitious being ... That person's thoughts are imagined by the author while he is pretending to report them: one's guesses are unerring only in the case of something that one is in the process of inventing (65). Although these theorists' vocabulary varies (Genette chooses inventing; Cohn says distinctive epistemology; Dolezel speaks of the performative), they express the same idea: that what a heterodiegetic narrator says about the narrative world is a knowable fact because the narrator says it. Palmer calls the idea a cliche; see it as the fundamental fact of fiction. To ignore it, argue, is to reduce all statements in fiction to the same epistemological level. Shlomith Rimmon-Kenan makes the distinction have in mind in her reading of a passage from Henry James's Portrait of a Lady in which the narrator states that the protagonist, Isabel Archer, is young person of many theories; her imagination was remarkably active: Such naming of a character's qualities counts as direct characterization only if it proceeds from the most authoritative voice in the text . …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 31 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.