The writing of fiction

by Edith Wharton

One of the chief obligations, in a short story, is to give the reader an immediate sense of security. Every phrase should be a sign-post, and never (unless intentionally) a misleading one: the reader must feel that he can trust to their guidance. His confidence once gained, he may be lured on to the most incredible adventures—as the Arabian Nights are there to show. A wise critic once said: “You may ask your reader to believe anything you can make him believe.” It is never the genii who are unreal, but only their unconvinced historian’s description of them. The least touch of irrelevance, the least chill of inattention, will instantly undo the spell, and it will take as long to weave again as to get Humpty Dumpty back on his wall. The moment the reader loses faith in the author’s sureness of foot the chasm of improbability gapes.

Improbability in itself, then, is never a danger, but the appearance of improbability is; unless, indeed, the tale be based on what, in my first chapter, I called pathological conditions—conditions of body or mind outside the field of normal experience. But this term, of course, does not apply to states of mind inherited from an earlier phase of race-culture, such as the belief in ghosts. No one with a spark of imagination ever objected to a good ghost story as “improbable”—though Mrs. Barbauld, who doubtless lacked the spark, is said to have condemned “The Ancient Mariner” on this ground. Most of us retain the more or less shadowy memory of ancestral terrors, and airy tongues that syllable men’s names. We cannot believe a priori in the probability of the actions of madmen, or neurasthenics, because their reasoning processes escape most of us, or can at best be imagined only as belonging to abnormal and exceptional people; but everybody knows a good ghost when he reads about him.

When the reader’s confidence is gained the next rule of the game is to avoid distracting and splintering up his attention. Many a would-be tale of horror becomes innocuous through the very multiplication and variety of its horrors. Above all, if they are multiplied they should be cumulative and not dispersed. But the fewer the better: once the preliminary horror posited, it is the harping on the same string—the same nerve—that does the trick. Quiet iteration is far more racking than diversified assaults; the expected is more frightful than the unforeseen. The play of “Emperor Jones” is a striking instance of the power of simplification and repetition to excite in an audience a corresponding state of tension. By sheer voodoo-practice it shows how voodoo acts.

In “The Turn of the Screw”—which stands alone among tales of the supernatural in maintaining the ghostliness of its ghosts not only through a dozen pages but through close on two hundred—the economy of horror is carried to its last degree. What is the reader made to expect? Always—all through the book—that somewhere in that hushed house of doom the poor little governess will come on one of the two figures of evil with whom she is fighting for the souls of her charges. It will be either Peter Quint or the “horror of horrors,” Miss Jessel; no diversion from this one dread is ever attempted or expected. It is true that the tale is strongly held together by its profound, its appalling moral significance; but most readers will admit that, long before they are conscious of this, fear, simple shivering animal fear, has them by the throat; which, after all, is what writers of ghost stories are after.

III

It is sometimes said that a “good subject” for a short story should always be capable of being expanded into a novel.

The principle may be defendable in special cases; but it is certainly a misleading one on which to build any general theory. Every “subject” (in the novelist’s sense of the term) must necessarily contain within itself its own dimensions; and one of the fiction-writer’s essential gifts is that of discerning whether the subject which presents itself to him, asking for incarnation, is suited to the proportions of a short story or of a novel. If it appears to be adapted to both the chances are that it is inadequate to either.

It would be as great a mistake, however, to try to base a hard-and-fast theory on the denial of the rule as on its assertion. Instances of short stories made out of subjects that could have been expanded into a novel, and that are yet typical short stories and not mere stunted novels, will occur to every one. General rules in art are useful chiefly as a lamp in a mine, or a hand-rail down a black stairway; they are necessary for the sake of the guidance they give, but it is a mistake, once they are formulated, to be too much in awe of them.

There are at least two reasons why a subject should find expression in novel-form rather than as a tale; but neither is based on the number of what may be conveniently called incidents, or external happenings, which the narrative contains. There are novels of action which might be condensed into short stories without the loss of their distinguishing qualities. The marks of the subject requiring a longer development are, first, the gradual unfolding of the inner life of its characters, and secondly the need of producing in the reader’s mind the sense of the lapse of time. Outward events of the most varied and exciting nature may without loss of probability be crowded into a few hours, but moral dramas usually have their roots deep in the soul, their rise far back in time; and the suddenest-seeming clash in which they culminate should be led up to step by step if it is to explain and justify itself.

There are cases, indeed, when the short story may make use of the moral drama at its culmination. If the incident dealt with be one which a single retrospective flash sufficiently lights up, it is qualified for use as a short story; but if the subject be so complex, and its successive phases so interesting, as to justify elaboration, the lapse of time must necessarily be suggested, and the novel-form becomes appropriate.