Bortolussi, Marisa, and Peter Dixon. Psychonarratology: Foundations for the Empirical Study of Literary Response. Cambridge: Cambridge University Press, 2003.

Chapter 3: The Narrator

Narrator as participant Presence of a narrator, distinct from author, a central feature of literary narratives; and reader treats narrator as participant in communicative situation (60). The presence of a narrator in (almost) every narrative, unless strongly indicated otherwise.
Status of narrator: still a "person" Types of narrator: diegetic (part of story world), extradiegetic (outside story world), etc. (63-4); whether readers are responsive to the differences has not been studied empirically. Textual signs of the narrator, whether recognizable as a person or not; even when not, principles of selection, summary, etc., imply a person (64-5).
Multiple entities in narrative Problems with different levels of narration, from real author, implied author, narrator, to implied reader, real reader. In all narratology the real reader remains unconsidered; or an assumed "ideal" reader who will respond like a narratologist (69).
Reader creates narrator: shares values, is cooperative Their approach: to see the narrator as a construction of the reader (based on actual text features), similar in certain respects to a conversational partner. Implications: 1) readers will assume narrator shares their language and cultural framework (72). 2) Readers assume narrator is cooperating (Grice maxims), and when not will make inferences to repair the problem, hence role of "narratorial implicatures," i.e., inferences about narrator (73). Reader assumes narrator is addressing them for some purpose, and will be motivated to discover what this is (73) [not, perhaps, to arrive at an interpretation, which would be like treating a conversation as a therapeutic interview to establish a diagnosis of the narrator's situation].
Narrator and author relationship; model of narrator* That for the reader the narrator and the author (represented author) may be distinct or overlap, or be identified, depending on the narrative in question (75). Advantages of their scheme in reducing the communicative situation to the reality of the reading experience (76) -- diegetic levels, etc., may be of little influence. That as in conversation the reader maintains a mental "representation of the narrator [that] consists of information concerning the narrator's knowledge, goals, and perspective" (77).
Narrator and narratee Discussion of such terms as narratee and implied reader (78-9) argues for simplifying the number of supposed entities: the most salient being the narrator and the narratee (the reader's conception of who the narrator assumes the reader to be). Version of reader ("implied reader"; "narratee") is encoded in text, reader likely to endow this reader with his own qualities (79) [ask: what is reader supposed to know?]
Attributions or inferences about narrator; or associated with a character Processing the narrator: explicit attributions that elaborate character attributes of narrator, etc.; or inference invitations, where reader must infer from signs about narrator. (80-81). Another method: narrator has close association with one character, so assumed to share character's attributes, through limited omniscience or free indirect discourse (82).
Inference invitations Inference invitations: textual basis for drawing conclusions, inferring narrator's attitude, such as irony being intended (81). Or reader may assume that a given character is a model for the narrator: "narrator-character association"; example: use of free indirect speech (82).
Unreliable narrators Questions about unreliable narrators, how to distinguish. (82-3). One may doubt the information provided by the narrator but still find the narrator cooperative (83). First person narrators may seem more authentic (84). Important challenge to empirical study, what clues make a reader conclude that a narrator is unreliable; might just be on ideological grounds (84).
Implicatures and identification with narrator That readers may identify with the narrator on the basis of implicatures drawn from the text. Discussion of some empirical work that shows this: Larsen and Seilman (1988) on personal resonance of remindings, literary texts eliciting more first person memories; Cupchik, Oatley, and Vorderer (1998) who oriented readers to be sympathetic or self-oriented, and found the latter "rated the experience-loaded stories higher in personal meaningfulness" (87) [note implications for immersion models -- these readers self-aware while reading]. See also discussion of Gerrig, Cupchik, Tan (87-8) on strategies for eliciting identification and role of reader's familiarity, knowledge, etc. Readers might be thought to vary if dependent on matching personal qualities of reader and narrator; but this not the case, hence text features likely to determine identification (89).
Transparency of character* That identification involves "transparency," belief of reader that he understands a character and his feelings and thoughts (89); especially likely to occur when reader brings attitudes and beliefs to the reading that he believes to be shared by the narrator.
Reducing implicatures = lower transparency Empirical study of identification and transparency. Propose that transparency is produced when readers attribute their own knowledge to narrator. Opening preamble of a Munro first-person narrator story altered to spell out implicatures that it was expected most readers would draw. The altered preamble was rated lower on transparency, said to be because readers needed to generate few implicatures (91-94). Results explained: that with the implicit preamble "readers have a greater opportunity to attribute their own experience to the narrator" (94).
* [Theory of Mind: attributing traits, feelings, intentions, etc., to narrator.]

return to course page

Document prepared March 13th 2007 / Updated December 11th 2011