The theory below is limited to what the method uses. Each strand opens with the writer’s question, then gives the source that answers it. Most of the sources describe a particular tradition, the classical tragedy, the golden-age detective story, the modernist unreliable narrator, and their rules are period conventions describing contracts that readers of those forms accepted, not laws of narrative.
What does a reader do at a reveal?
A writer who has a secret wants to know what the disclosure will do to the reader, and the oldest answer is Aristotle’s. In the Poetics (chapter 11, in S. H. Butcher’s translation) a recognition is “a change from ignorance to knowledge”, and the finest recognition is one that coincides with a reversal, a change of fortune, as when Oedipus’s search for the murderer of Laius discovers himself. In chapter 16 Aristotle ranks the kinds of recognition: lowest are those made by tokens and scars, or contrived by the poet; higher are those made by memory and by reasoning; best of all is the recognition that “arises from the incidents themselves”, where the discovery is produced by the chain of events the play has already set moving. The ranking is the earliest statement of two conditions this chapter uses: a reveal should be received by a character in an act that changes something, and it should come from inside the book’s causal system rather than from a token supplied for the purpose.
What Aristotle describes from the stage, the reader performs in her head. The disclosure sends her back over what she has read, and the pleasure of a good reveal is the pleasure of finding the earlier pages rearranged and still whole. The double reading in Definitions names that movement.
What is the reader feeling before it?
Before a reveal the writer has to know what kind of interest the withheld fact is generating, because that decides how the reveal should land. Meir Sternberg, in Expositional Modes and Temporal Ordering in Fiction (1978) and later essays, distinguishes three interests produced by the order of telling. Suspense is uncertainty about what will happen. Curiosity is uncertainty about something in the past that the reader knows she lacks, as when a book opens on a body and the reader wants to know who killed it. Surprise is the discovery of a gap the reader did not know was there, which sends her back to revise what she thought she understood. A reveal answers curiosity or produces surprise, and the chapter calls the two kinds the curiosity reveal and the surprise reveal. Roland Barthes’s S/Z (1970; translated by Richard Miller, 1974) describes the same field from the side of the text: a hermeneutic code that poses an enigma and delays its answer by snares, equivocations, partial answers, suspended answers and jamming, until a disclosure. The terms are useful as a list of what a writer can do between posing a question and answering it.
Alfred Hitchcock put the trade between surprise and suspense in an image that writers still use. In his interviews with François Truffaut (Hitchcock, 1966; English 1967) he described a bomb under a table: if it goes off unannounced the audience has fifteen seconds of surprise, but if it has been shown to the audience beforehand, the conversation above it gives fifteen minutes of suspense. The image is about film and about bombs, but its lesson is general: a fact disclosed to the reader early, before the characters learn it, converts surprise into suspense. Where to disclose is a choice between the two, and The Mechanism treats it as alignment.
What makes a reveal fair?
A writer worried that a reveal will feel like a trick is asking about fair play, a term the detective story made explicit. Ronald Knox’s “Decalogue” (1929) and S. S. Van Dine’s “Twenty Rules for Writing Detective Stories” (1928) set out what a puzzle writer owed the reader. Knox’s first rule says that the criminal must be mentioned early but must not be anyone whose thoughts the reader has been allowed to follow; his ninth that the detective’s companion, the Watson, must not conceal any thoughts which pass through his mind. Three years earlier Agatha Christie’s The Murder of Roger Ackroyd (1926) had been narrated by its murderer, and the arguments over whether it cheated have run ever since. The rules are a period convention, and few writers outside the puzzle mystery keep them literally. What survives is their principle: the reader must have been able, in principle, to read the passages correctly, and no thought of a mind the reader shares may be hidden without a reason the book can show. The principle is the double reading and the withholder of The Mechanism.
How far can the reader trust the teller?
The question underneath the largest reveals is whether the narration itself can be mistaken or lying. Wayne C. Booth’s The Rhetoric of Fiction (1961) named the unreliable narrator: a narrator who does not speak or act in accordance with the norms of the work, so that the reader learns to read past what the narrator says. Gérard Genette’s Narrative Discourse (1972; translated by Jane E. Lewin, 1980) supplies the finer instrument, paralipsis, the omission of something the narration’s vantage ought to have included. Between them they mark the difference this chapter calls the teller rule: a narrator, or a character who writes or tells within the book, can be wrong or lie, and a reveal can expose it; the narration of a book that has declared no such teller cannot, and a reveal that requires it to have lied voids what it reveals.
What does the evidence say?
The evidence is thin and experimental, using short texts rather than novels. Hans Hoeken and Mario van Vliet (Poetics, 2000) gave readers versions of one story with its discourse rearranged and found that adding an unexpected event raised both readers’ appreciation and their recall of the story, which the authors attribute to readers reassessing the preceding events: the double reading, measured. Jonathan Leavitt and Nicholas Christenfeld (Psychological Science, 2011) found that readers given the outcome of short stories in advance, including stories with ironic twists and mysteries, rated them as more enjoyable, not less. Benjamin Johnson and Judith Rosenbaum (Communication Research, 2015) found the opposite with longer stories: unspoiled readers reported more suspense and more enjoyment. A conjecture consistent with both is that a reveal’s value depends less on the surprise than on what the reader does with it, which is the claim this chapter makes on other grounds.
See also: the full treatment, section 2, with Aristotle’s six kinds of recognition, Barthes’s terms, the fair-play codes and the evidence at greater length.