Every prologue opens a gap between itself and the story that follows. What separates a working prologue from an orphaned one is whether the main narrative does something with the gap.
Distance and its four axes
Distance is how far a prologue sits from Chapter One, measured along four axes: an earlier or later time, a different place, a viewpoint the reader will not hold again soon, and a register the main narrative does not share. The axes open different gaps and are closed by different means, so the table below is the one place in the chapter where they are set side by side.
| Axis | What it buys | How it is normally closed | Characteristic failure |
|---|---|---|---|
| Time | Compressed antecedent, dramatic irony, a framed ending | The narrative reaches the prologue’s moment, or explains it | A rendezvous the reader waits for and never gets |
| Place | Scale, an off-page threat, a second front | A character reaches the place, or its consequences arrive | A decorative location that never matters |
| Viewpoint | The absent vantage, access to a dead or opposing mind | The viewpoint returns, or its traces are recognised in what others find | A consciousness the reader attached to and lost |
| Register | Protocol, lore, documentary authority | The register recurs, or its claims are cashed in the ordinary voice | A book that reads as a falling-off from its opening |
Most prologues use more than one axis, and one of them carries the load. The household prologue is distant in time (three days before the father stopped answering the phone), in viewpoint (his) and slightly in register (a retrospective formality). The load-bearing axis is viewpoint: any flashback could cross the time gap, but nothing in the main narrative can recover his vantage. Naming that axis tells the writer what the closing must deliver. The axes also differ in what it costs to leave them open. An open register gap is often tolerable, since a reader accepts that a book had an overture; an open viewpoint gap is the most expensive, because the reader built a mind into a contextual frame and had it withdrawn.
Closing the distance
Closing the distance is the point or points at which the main narrative catches up with, recontextualises, completes or deliberately refuses to close what the prologue opened. There are four modes, and they are not ranked; each suits a different book. In total closure the narrative reaches the prologue’s moment and completes it: Frankenstein returns to Walton’s ship at the end, and the pilot of Breaking Bad (AMC, 2008), which opens on Walt White in a gas mask in an RV in the desert, recording a message to his family as sirens approach, goes back to his fiftieth birthday and catches up with the desert by the episode’s end. In fragmentary closure elements of the prologue recur singly across the book and are recognised each time, as the Chorus of Henry V returns before every act. In recontextualisation the narrative never reaches the prologue but changes what it meant, so that a second reading differs from the first. In deliberate non-closure, which the Reference calls the held note, the gap is held open for good, as Henry James never returns to the fireside that frames The Turn of the Screw (1898).
Deliberate and negligent non-closure look the same on the page, and three signals tell them apart: the book withholds other closures too, so that open ends are its policy; the non-closure follows from its subject, as the unresolvable governess’s tale would be falsified by a return to the confident fireside; and the prologue is about something that cannot be got back. Where none of the three applies, the honest diagnosis is a distance the book forgot.
Genette’s Narrative Discourse (1972; translated by Jane E. Lewin, 1980) supplies the vocabulary for the two ways a prologue can plant what it will close. Advance notice (annonce) announces what is to come and is closed by arrival: the event occurs. The seed (amorce) is a detail that passes as insignificant and acquires significance only in retrospect, and it is closed by recognition: the reader meets it again and understands. The prologue of Stieg Larsson’s The Girl with the Dragon Tattoo (2005; translated by Reg Keeland, 2008) is built on a seed. An old man receives a pressed flower on his eighty-second birthday, as he has every year since his niece disappeared, and he takes it for a taunt from her killer. The flowers return when Henrik Vanger tells the journalist Mikael Blomkvist their history, and they are closed when the investigation finds the sender: Harriet herself, alive, telling him so in the only way she dared. A reader who meets the flower in the prologue carries a question the whole investigation answers.
The argument for fragmentary closure
Craft advice, where it addresses the closing at all, tends to assume a single late reveal: the prologue’s meaning is held back and delivered in one place near the end. That is usually the weaker structure, and Emmott’s frame model, in How It Works, explains why. A single late reveal asks the reader to reinstate the prologue’s thin frame after the longest interval, at the point where the reinstatement costs most. Fragmentary closure inverts the schedule. A series of small returns re-primes the frame at the cost of a clause each time, confirms that the prologue is being used, makes each later recognition cheaper, and spreads the risk, since a reader who misses one fragment can catch the next. Three rules follow. Plant three to five recoverable specifics, objects and phrases rather than moods, since a reader can recognise a fountain pen and cannot recognise an atmosphere. Bring the first fragment back within the first quarter of the book, where its job is reassurance, not payoff. And vary the weight, from an unremarked mention to a noticed one to a scene that turns on the object.
The single reveal is right in two cases: when the prologue’s meaning is the book’s central secret, which fragments would give away, and when the effect depends on the reader having forgotten. The conditional form is the useful one. A prologue meant to be carried, as a dread, an irony or a knowledge the characters lack, should be closed in fragments, because its effect is continuous. A prologue meant to be sprung should be closed once, and the writer should shorten the interval instead of lengthening the suspense. Most prologues, examined, turn out to be of the first kind and to have been built as if they were of the second.
Types, and the test of substitution
The Reference chapter tabulates fourteen recognisable types of prologue. A few recur in revision. The proleptic flash-forward is strongest when it shows an aftermath, not the climax: a narrator in exile remembering reveals that something happened without saying what, and buys concord at little cost in suspense. The victim’s last hour, a character followed until a death, works when the victim is a person and fails when the victim is a function. The frame narrator’s reception and the editorial note serve the authority domain. The false prologue relies on a misreading the reader makes, such as taking a scene for the beginning when it is the end. The chapter proposes a fourteenth type, the last ordinary day: the routine that immediately precedes a rupture the main narrative will investigate, in which nothing happens and the dread comes wholly from the reader’s knowledge that the ordinariness is ending. It contains no event to be arrived at, only objects to be recognised, so it is the type that most needs fragmentary closure. The household prologue is one.
The taxonomy is for substitution, not classification. A prologue that is not working is often the right material in the wrong shape, and rewriting it as another type that keeps its load-bearing axis and its domains shows what each shape buys. Substituting the last ordinary day for the victim’s last hour suits a book about aftermath, not pursuit; substituting an editorial note for a deep-historical overture suits material that is information, not experience. Substitution often ends in absorption, which is a success of the method, not a failure of the draft.
Multiplicity as evidence
Multiplicity, in The Purposes of a Scene’s sense (Chapter 1), is a unit’s doing several jobs at once. In an opening it is an economy; in a prologue it is evidence. One domain can always be explained away, since tone can be set inside Chapter One and a fact delivered later, but three domains braided into the same three hundred words are hard to relocate without scattering them across three places, and the writer can then say what would be lost. As a rule of thumb, a prologue serving one domain is suspect, two is defensible if one requires exteriority, and three is the normal shape of one that survives revision. The multiplying work lives in four places: the choice of moment (the household prologue’s three days before, not the day itself), the objects handled, the axis chosen, and the last sentence. That sentence is the final and most heavily weighted position of a bounded unit, and a prologue that spends it on a portentous half-line has wasted its strongest place.
See also: the Reference treatment of distance and its closing, with all fourteen types, a table of substitutions and the historical drift of the axes; the Reference multiplicity, with Hurston’s first sentence annotated and the domain audit.