10. The Viewpoint Shift Across Forms

The allocation model was developed for the novel, which is the form in which the coverage–depth trade is most acute because it is the form with enough room for the trade to be made in more than one way. It transfers to adjacent forms with adjustments that need stating, because students routinely import novel habits into forms that punish them.

10.1 The short story

The short story’s default is the single fixed vantage (§6.3, no. 1), and the default is close to a requirement.

The exceptions cluster in two places, and both are instructive. The first is the story that uses only two vantages in sequence, so that the second reframes the first — a structure that works because it is not a pattern but a reversal. The second is the story that opens or closes with a widened vantage: Joyce’s “The Dead” moves, in its final pages, from Gabriel Conroy’s mortified interiority outward to a snow falling over the whole of Ireland, and the widening is the story’s meaning. That move is the omniscient slip (§8.6) used deliberately and once, and it is unavailable to a story that has not first been rigorously confined.

Note what the form makes visible: in a story, allocation is nearly all character compression. Woloch’s model applies with unusual clarity, because the reader can see the whole character-system at once.

10.2 The novella

The novella’s characteristic viewpoint problem is that it is written by novelists. It has a novel’s ambition and a story’s word count, and multi-viewpoint structures that would be modest in a four-hundred-page book consume it. Three vantages in thirty thousand words is a page share of ten thousand words each — the length of a long short story, which is not enough to build the familiarity that makes alternation cheap.

The forms that work are the single fixed vantage and the frame (§6.3, no. 7), which is why so many canonical novellas are frames: Heart of Darkness, The Turn of the Screw, Season of Migration to the North. The frame gives a novella the effect of plurality — two positions, two moral registers, a listener — at the cost of only one additional vantage, and it supplies a reason for the story’s brevity: it is being told, by someone, to someone, in an evening.

10.3 The long novel, the series, and serial publication

Length raises the ceiling (§5.4), but not in the way writers expect. What length actually buys is not more vantages but longer stays, and a long novel that spends its extra room on additional viewpoint characters rather than on longer visits has bought nothing.

Serial publication complicates this and is worth a note, because the nineteenth-century multi-plot novel is the product of a distribution format. A monthly part had to contain variety, which encouraged multiple threads; a nineteen-month run had to sustain memory across long gaps, which encouraged strongly typed characters and repeated identifying tags — the very character compression Woloch analyses. Dickens’s number plans, which survive, show him distributing his narrations and his plots across each part deliberately.

Series fiction faces a specific and under-discussed pressure: viewpoint casts grow. A series that begins with two vantages often has more by its fourth volume (an observation, not a count), because the world has expanded and coverage is required, and because a returning reader’s affection for a secondary character is a commercial argument for promoting them. The consequences are predictable and are the diluted ensemble in slow motion. Three principles:

1. A viewpoint granted in volume three is a later grant with respect to the series, not a first grant with respect to the volume. The reader’s investment is cumulative across books.
2. Terminating a grant is harder in a series than in a novel, because readers experience a vanished vantage as a promise broken across years.
3. The bounded grant scales badly. Readers who have been trained by a series to expect continuity will read a single-appearance vantage as a setup rather than as a closed structure.

10.4 The screen comparison, admitted and corrected

Contemporary craft advice imports heavily from film and television, and viewpoint is where the import does most damage, because the two media have opposite defaults.

Film’s default is external focalisation. A camera shows behaviour. Interiority must be constructed — by voice-over, by point-of-view shots, by performance, by score — and every instrument is expensive and conspicuous. Prose’s default is the opposite: the moment a novel reports what someone thought, it has granted interiority, and it costs nothing.

Two consequences follow, and both correct common advice.

• “Show, don’t tell” has been reinforced by screen logic in a medium that has the opposite economy. The phrase descends from Lubbock’s preference for showing over telling (1921), which is older than the screenwriting manuals that now repeat it. In film, interiority is expensive and behaviour is free. In prose, interiority is free and behaviour must be described. A novel that renders everything behaviourally has voluntarily adopted film’s constraint without acquiring film’s compensations — the face, the pause, the actor.
• The “ensemble show” model does not transfer. Television can cut among eight characters cheaply because the viewer re-identifies each in a quarter of a second, by face. Prose re-identification requires a name and a paragraph. An eight-vantage novel is not the equivalent of an eight-hander drama; it is considerably more expensive.

What does transfer is David Bordwell’s analysis of narration in Narration in the Fiction Film (1985), whose three variables — knowledgeability (how much the narration knows), communicativeness (how much of what it knows it discloses) and self-consciousness (how far it acknowledges addressing an audience) — are a cleaner instrument than “point of view” for describing what a novel’s narration is doing. Genette’s zero focalisation is high knowledgeability; Stoker’s Dracula is high communicativeness within severely limited knowledge; Eliot’s narrator is high on all three.

A note on restricted narration. Hitchcock’s much-quoted account of suspense — the audience shown the bomb under the table that the characters cannot see — is dramatic irony (§4.2) produced by an allocation of knowledge, and it is transferable. What is not transferable is the cheapness of the allocation: a film supplies the bomb in one shot; a novel must grant a vantage or contrive a report.

10.5 Translated fiction

Viewpoint is one of the areas where this book’s standing distinction, stated in its Introduction — structural claims survive translation; texture claims do not — is most consequential, because the chief carrier of internal focalisation in the modern novel is free indirect style, and free indirect style is grammatically dependent.

Four specific hazards.

(1) Free indirect style depends on tense and pronoun systems that differ across languages. English marks it by backshifted tense and third-person pronouns; languages with different tense architectures mark it differently or leave it ambiguous. A passage that is unmistakably free indirect in Russian or French may become, in English, either plain psycho-narration or quoted thought, depending on the translator’s decision. The architecture of the novel is unaffected; the distance at any given moment is a translator’s choice.

(2) Languages that do not require subject pronouns leave focalisation less explicitly marked. In Japanese, for instance, an unmarked sentence may not specify whose perception it reports, and the reader infers from honorifics, verb forms of giving and receiving, and context. Translation into English forces a choice that the original did not make. Structural claims about a novel’s allocation survive; claims about the precise point at which the reader knows whose head they are in do not.

(3) Naming systems carry viewpoint information that translation flattens. Chapter 5 §9.4 makes the point about Russian patronymics and diminutives: which form of a name the narration uses signals the social frame the narration has currently adopted. In English the signal is largely lost, and with it a quantity of focalisation information the original reader receives free.

(4) Animal and non-human focalisation is especially translation-sensitive. §9.4’s Laska passage differs measurably across English versions in how much articulate thought the dog is permitted. The structural fact — Tolstoy focalises a dog — is robust. Everything a student might want to say about how far inside the dog the narration goes is a fact about the translation.

The teaching rule. A student writing about viewpoint in a translated novel should name the translation in the first footnote, restrict distance claims to that translation, and confine cross-linguistic claims to architecture: who has a vantage, how many, in what proportion, and when each begins. Those survive.

10.6 Concordance: concept to novel

A study aid mapping each concept to the novel that best displays it.