10. The Descriptive Detail Across Forms

The model developed above was built for the novel. It transfers to adjacent forms with adjustments, stated here because writers routinely import novel-descriptive habits into forms that punish them.

10.1 The short story

The story’s descriptive economy differs from the novel’s not in degree but in kind, because a story has no time to establish a baseline of the ordinary. In a novel, a detail can be planted flat on page twenty and promoted on page two hundred; the reader’s sense of what is normal in this world has been built by then, and a deviation registers. A story has no such reserve.

The practical inference is counter-intuitive: a short story tolerates less unmotivated detail than a novel, not more, and the airlessness §16 warns against is much harder to avoid. The classic remedy in the form is the single obtrusively concrete object given no explanation — the strategy Chekhov uses constantly — which supplies air by being specific rather than by being useless.

10.2 The novella

The novella’s characteristic descriptive problem is that it is usually written by novelists. It has a novel’s tonal amplitude and a story’s word count, and a page of description costs a fortieth of the book. Its natural solution is the single dominating object or place: one setting described more thoroughly than a novel would allow, and everything else left bare. The concentration is a strength of the form rather than a compromise, and it is why so many novellas are remembered by a place.

10.3 The long novel and the series

Two effects operate at scale, and both are about recurrence.

• Motifs get longer intervals and need stronger variation. A detail returning after two hundred pages must be recognisable without being identical, which usually means the object stays and the register of naming changes.
• Series volumes must re-establish objects without recapping them. The reliable technique is to reintroduce the object in use rather than in description: a returning reader recognises it, and a new reader receives it as ordinary world detail. Recapping its significance serves neither.

Long novels also enjoy a licence short ones do not: they can afford whole descriptive chapters, because the reader’s accumulated investment underwrites them. Melville, Hugo and Tolstoy all spend it. The licence is not available on page ten.

10.4 Screen comparison

Contemporary craft advice imports heavily from film, and description is where the import does the most damage.

Two inferences. First, prose has no establishing shot, and writers who begin scenes with one have borrowed a free device from a medium that could afford it. Second — and this is the finding that most repays attention — film cannot help producing the reality effect, and prose must manufacture it. Every frame contains hundreds of objects nobody chose to signify. This is one reason screen adaptation so often makes a sparse novel feel cluttered, and one reason a novel written to a screenplay’s descriptive discipline can feel airless on the page.

What does transfer is the principle of describing through action rather than through survey: a camera reveals a room while somebody crosses it, and prose is better for doing the same (§3.5, on motion).

10.5 Translated fiction

The series’ standing principle applies with unusual force here, because description is the part of a novel most damaged in translation.

• Structural claims survive. That Flaubert describes an object where another writer would describe a feeling; that Satis House is built from negatives; that Achebe’s world is delivered through procedure — all of this is visible in any competent translation, and every analysis in §9 that concerns selection is translation-independent.
• Connotative claims do not. The precise social charge of a barometer, the register of a proverb, the difference between two words for a garment, the sound of a list: these are the first casualties. A student analysing translated description should confine confident claims to what is chosen, and hedge claims about how it sounds.
• A specific hazard: the translator’s normalisation of the odd detail. Translators, reasonably, tend to smooth a detail that reads as bizarre in the target language, and the bizarre detail is often the load-bearing one. Comparing two translations of a single descriptive paragraph is the fastest way to see which of its effects were structural.

10.6 Speculative and historical fiction: the descriptive tariff

A point that follows from §3.6 and deserves its own statement. The contemporary two-detail norm depends on cueing: a writer names a tenement, a courtroom or a suburban kitchen and the reader supplies the rest from an enormous stock of pre-supplied images. Fiction set in an unfamiliar world has no such stock to draw on, and must pay a higher descriptive tariff — more detail, and more of it load-bearing, because the reader is building rather than retrieving.

Two consequences. Speculative and historical fiction are structurally more vulnerable to the info-dump (§8.5), because they need more explanation and have the same aversion to it. And the unexplained proper noun is their most valuable instrument, since it delivers world detail while generating curiosity rather than spending it. Chapter 9 (The Opening Pages) makes the same argument about openings; the generalisation is that a cue is cheap where a stock exists and worthless where it does not.

10.7 Concordance: concept to novel

A study aid: the novel that best displays each concept, with the section that discusses it.