Robert Louis Stevenson, in his spirited essay entitled "A Humble Remonstrance," has given very valuable advice to the writer of narrative. In concluding his remarks he says, "And as the root of the whole matter, let him bear in mind that his novel is not a transcript of life, to be judged by its exactitude; but a simplification of some side or point of life, to stand or fall by its significant simplicity. For although, in great men, working upon great motives, what we observe and admire is often their complexity, yet underneath appearances the truth remains unchanged: that simplification was their method, and that simplicity is their excellence." Indeed, as we have already noted in passing, simplification is the method of every art. Every artist, in his own way, simplifies life: first by selecting essentials from the helter-skelter of details that life presents to him, and then by arranging these essentials in accordance with a pattern. And we have noted also that the method of the artist in narrative is to select events which bear an essential logical relation to each other and then to arrange them along the lines of a pattern of causation.

Of course the prime structural necessity in narrative, as indeed in every method of discourse, is unity. Unity in any work of art can be attained only by a definite decision of the artist as to what he is trying to accomplish, and by a rigorous focus of attention on his purpose to accomplish it,—a focus of attention so rigorous as to exclude consideration of any matter which does not contribute, directly or indirectly, to the furtherance of his aim. The purpose of the artist in narrative is to represent a series of events—wherein each event stands in a causal relation, direct or indirect, to its logical predecessor and its logical successor in the series. Obviously the only way to attain unity of narrative is to exclude consideration of any event which does not, directly or indirectly, contribute to the progress of the series. For this reason, Stevenson states in his advice to the young writer, from which we have already quoted: "Let him choose a motive, whether of character or passion: carefully construct his plot so that every incident is an illustration of the motive, and every property employed shall bear to it a near relation of congruity or contrast; ... and allow neither himself in the narrative, nor any character in the course of the dialogue, to utter one sentence that is not part and parcel of the business of the story or the discussion of the problem involved. Let him not regret if this shortens his book; it will be better so; for to add irrelevant matter is not to lengthen but to bury. Let him not mind if he miss a thousand qualities, so that he keeps unflaggingly in pursuit of the one he has chosen." And earlier in the same essay, he says of the novel: "For the welter of impressions, all forcible but all discreet, which life presents, it substitutes a certain artificial series of impressions, all indeed most feebly represented, but all aiming at the same effect, all eloquent of the same idea, all chiming together like consonant notes in music or like the graduated tints in a good picture. From all its chapters, from all its pages, from all its sentences, the well-written novel echoes and re-echoes its one creative and controlling thought; to this must every incident and character contribute; the style must have been pitched in unison with this; and if there is anywhere a word that looks another way, the book would be stronger, clearer, and (I had almost said) fuller without it."

The only way in which the writer of narrative may attain the unity that Stevenson has so eloquently pleaded for is to decide upon a definite objective point, to bear in mind constantly the culmination of his series of events, and to value the successive details of his material only in so far as they contribute, directly or indirectly, to the progress of the series toward that culmination. To say the thing more simply, he must see the end of his story from the beginning and must give the reader always a sense of rigorous movement toward that end. His narrative, as a matter of construction, must be finished, before, as a matter of writing, it is begun. He must know as definitely as possible all that is to happen and all that is not to happen in his story before he ventures to represent in words the very first of his events. He must not, as some beginners try to do, attempt to make his story up as he goes along; for unless he holds the culmination of his series constantly in mind, he will not be able to decide whether any event that suggests itself during the progress of his composition does or does not form a logical factor in the series.

The preliminary process of construction may be accomplished in either of two ways. Authors with synthetic minds will more naturally reason from causes to effects; and authors with analytic minds will more naturally reason from effects to causes. The former will construct forward through time, the latter backward. Standing at the outset of a narrative, it is possible to imagine forward along a series of events until the logical culmination is divined; or standing at the culmination, it is possible to imagine backward along the series to its far-away beginnings. Thackeray apparently constructed in the former manner; Guy de Maupassant apparently constructed in the latter. The latter method—the method of building backward from the culmination—is perhaps more efficacious toward the conservation of the strictest unity. It seems on the whole a little easier to exclude the extraneous in thinking from effects to causes than in thinking from causes to effects, because analysis is a stricter and more focused mood of mind than synthesis.

But in whichever way the process of construction be accomplished, the best stories are always built before they are written; and that is the reason why, in reading them, we feel at every point that we are getting somewhere, and that the author is leading us step by step toward a definite culmination. Although, as is usually the case, we cannot, even midway through the story, foresee what the culmination is to be, we feel a certain reassurance in the knowledge that the author has foreseen it from the start. This feeling is one of the main sources of interest in reading narrative. In looking on at life itself, we are baffled by a muddle of events leading everywhither; their succession is chaotic and lacking in design; they are not marshaled and processional; and we have an uncomfortable feeling that no mind but that of God can foresee their veiled and hidden culminations. But in reading a narrative arrangement of life, we have a comfortable sense of order, which comes of our knowledge that the author knows beforehand whither the events are tending and can make us understand the sequence of causation through which they are moving to their ultimate result. He makes life more interesting by making it more intelligible; and he does this mainly by his power of construction.

The simplest of all structures for a narrative is a straightway arrangement of events along a single strand of causation. In such a narrative, the first event is the direct cause of the second, the second of the third, the third of the fourth, and so on to the culmination of the series. This very simple structure is exhibited in many of the tales which have come down to us from early centuries. It is frequently employed in the "Gesta Romanorum," and scarcely less frequently in the "Decameron" of Boccaccio. It has the advantage of being completely logical and entirely direct. But we feel, in reading stories so constructed, that the method of simplification has been carried too far, and that simplicity has therefore ceased to be an excellence. Such a story is in this way misrepresentative of life:—it fails utterly to suggest "the welter of impressions which life presents," the sudden kaleidoscopic shifts of actual life from one series of events to another, and the consequent intricacy and apparent chaos of life's successive happenings. The structure is too straightforward, too direct, too unwavering and unhesitant.

The simplest way to introduce the element of hesitance and wavering, and thereby make the story more truly suggestive of the intricate variety of life, is to interrupt the series by the introduction of events whose apparent tendency is to hinder its progress, and in this way emphasize the ultimate triumph of the series in attaining its predestined culmination. Such events are not extraneous; because, although they tend directly to dispute the progress of the series, they tend also indirectly to further it through their failure to arrest it. The events in any skilfully selected narrative may, therefore, be divided into two classes: events direct or positive, and events indirect or negative. By a direct, or positive, event is meant one whose immediate tendency is to aid the progress of the series toward its predetermined objective point; and by an indirect, or negative, event is meant one whose immediate tendency is to thwart this predetermined outcome. It would be an easy matter, for example, in examining "Pilgrim's Progress," to class as positive those events which directly further the advance of Christian toward the Celestial City, and to class as negative those events whose immediate tendency is to turn him aside from the straight and narrow path. And yet both classes of events, positive and negative, make up really only a single series; because the negative events are conquered one by one by the preponderant power of the positive events, and contribute therefore indirectly, through their failure, to the ultimate attainment of the culmination.

When a straightway arrangement of positive events along a single strand of causation is varied and emphasized in this way by the admission of negative events, whose tendency is to thwart the progress of the series, the structure may be made very suggestive of that conflict of forces which we feel to be ever present in actual life. This structure is exhibited, for example, in Hawthorne's little tale of "David Swan." The point of the story is that nothing happens to David; the interest of the story lies in the events that almost happen to him. The young man falls asleep at noon-time under the shade of a clump of maples which cluster around a spring beside the highroad. Three people, or sets of people, observe him in his sleep. The first would confer upon him Wealth, the second Love, the third Death, if he should waken at the moment. But David Swan sleeps deeply; the people pass on; and all that almost happened to him subsides forever to the region of the might-have-been.

A simple series of this sort, wherein the events proceed, now directly, now indirectly, along a single logical line, may be succeeded by another simple series of the same sort, which in turn may be succeeded by a third, and so on indefinitely. In this way is constructed the type of story known as picaresque, because in Spain, where the type was first developed, the hero was usually a picaro, or rogue. The narrative expedient in such stories is merely to select a hero capable of adventure, to fling him loose into the roaring and tremendous world, and to let things happen to him one after another. The most widely known example of the type is not a Spanish story, but a French,—the "Gil Blas" of Alain RenÉ Le Sage. As soon as Gil Blas arrives at the culmination of one series of adventures, the author starts him on another. Each series is complete in itself and distinct from all the rest; and the structure of the whole book may be likened, in a homely figure, to a string of sausages. The relation between the different sections of the story is not organic; they are merely tied together by the continuance of the same central character from one to another. Any one of the sections might be discarded without detriment to the others; and the order of them might be rearranged. Plays, as well as novels, have been constructed in this inorganic way,—for example, MoliÈre's "L'Etourdi" and "Les Facheux." If the actors, in performing either of these plays, should omit one or two units of the sausage-string of incidents, the audience would not become aware of any gap in structure. Yet a story built in this straightforward and successive way may give a vast impression of the shifting maze of life. Mr. Kipling's "Kim," which is picaresque in structure, shows us nearly every aspect of the labyrinthine life of India. He selects a healthy and normal, but not a clever, boy, and allows all India to happen to him. The book is without beginning and without end; but its very lack of neatness and compactness of plan contributes to the general impression it gives of India's immensity.

But a simple series of events arranged along a single strand of causation, or a succession of several series of this kind strung along one after the other, may not properly be called a plot. The word plot signifies a weaving together; and a weaving together presupposes the co-existence of more than one strand. The simplest form of plot, properly so called, is a weaving together of two distinct series of events; and the simplest way of weaving them together is by so devising them that, though they may be widely separate at their beginnings, they progress, each in its own way, toward a common culmination,—a single momentous event which stands therefore at the apex of each series. This event is the knot which ties together the two strands of causation. Thus, in "Silas Marner," the culminating event, which is the redemption of Marner from a misanthropic aloofness from life, through the influence of Eppie, a child in need of love, is led up to by two distinct series of events, of which it forms the knot. The one series, which concerns itself with Marner, may be traced back to the unmerited wrong which he suffered in his youth; and the other series, which concerns itself with Eppie, may be traced back to the clandestine marriage of Eppie's father, Godfrey Cass. The initial event of one series has no immediate logical relation to the initial event of the other; but each series, as it progresses, approaches nearer and nearer to the other, until they meet and blend.

A type of plot more elaborate than this may be devised by leading up to the culmination along three or more distinct lines of causation, instead of merely two. In the "Tale of Two Cities," Sydney Carton's voluntary death upon the scaffold stands at the apex of several series of events. And a plot may be still further complicated by tying the strands together at other points beside the culmination. In "The Merchant of Venice," the two chief series of events are firmly knotted in the trial scene, when Shylock is circumvented by Portia; but they are also tied together, though less firmly, at the very outset of the play, when Antonio borrows from Shylock the money which makes it possible for Bassanio to woo and win the Lady of Belmont. Furthermore, any event in one of the main strands of causation may stand at the culmination of a minor strand, and thus may form a little knot in the general network of the plot. In the same play, the minor strand of the elopement of Lorenzo and Jessica attains its culmination in a scene which stands only midway along the progress of the two main strands, that of the bond and that of the caskets, toward their common result in the defeat of Shylock.

But however intricately woven a plot may be, and however many minor knots may tie together the various strands which enter into it, there is almost always one point of greatest complication, one big knot which ties together all the strands at once, and stands as the common culmination of all the series, major and minor. The story concerns itself chiefly with telling the reader how the major knot came to be tied; but in a plot of any complexity, the reader naturally desires to be told how the knot became untied again. Therefore this point of greatest complication, this culmination of all the strands of causation which are woven in the plot, this objective point of the entire narrative, is seldom set at the very end of a story, but usually at a point about three quarters of the way from the beginning to the end. The first three quarters of the story, speaking roughly, exhibit the antecedent causes of the major knot; and the last quarter of the story exhibits its subsequent effects. A plot, therefore, in its general aspects, may be figured as a complication followed by an explication, a tying followed by an untying, or (to say the same thing in French words which are perhaps more connotative) a nouement followed by a dÉnouement. The events in the dÉnouement bear a closer logical relation to each other than the events in the nouement, because all of them have a common cause in the major knot, whereas the major knot is the ultimate effect of several distinct series of causes which were quite separate one from another at the time when the nouement was begun. For this reason the dÉnouement shows usually a more hurried movement than the nouement,—one event treading on another's heels.

Undoubtedly it was this threefold aspect of a plot—1. The Complication; 2. The Major Knot; 3. The Explication—which Aristotle had in mind when he stated that every story must have a beginning, a middle, and an end. These words were not intended to connote a quantitative equality. What Aristotle called the "middle" may, in a modern novel, be stated in a single page, and is much more likely to stand near the close of the book than at the center. But everything that comes after it, in what Aristotle called the "end," should be an effect of which it is the cause; and everything that comes before it, in what Aristotle called the "beginning," should be, directly or indirectly, a cause of which it is the effect. Only under these conditions will the plot be, as Aristotle said it should be, an organic whole. Only in this way can it conform to the principle of unity, which is the first principle of all artistic endeavor.

Bearing the principle of unity ever in his mind, Stevenson, in a phrase omitted for the moment in one of the quotations from "A Humble Remonstrance" set forth at the beginning of this chapter, advised the fiction-writer to "avoid a sub-plot, unless, as sometimes in Shakespeare, the sub-plot be a reversion or complement of the main intrigue." It seems safe to state that a sub-plot is of use in a novel only for the purpose of tying minor knots in the leading strands of causation, and should be discarded unless it serves that purpose. There is no reason, however, why a novel should not tell at once several stories of equal importance, provided that these stories be deftly interlinked, as in that masterpiece of plotting, "Our Mutual Friend." In this novel, the chief expedient which Dickens has employed to bind his different stories together is to make the same person an actor in more than one of them, so that a particular event that happens to him may be at the same time a factor in both one and the other series of events. Through the skilful use of this expedient, Dickens has contrived to give his novel unity of plot, in spite of the diversity of its narrative elements. But on the other hand, in "Middlemarch," George Eliot has told three stories instead of one. She has failed to make her plot an organic whole by deftly interweaving the three strands which she has spun. And therefore this monumental novel, so great in other ways, is faulty in structure, because it violates the principle of unity.

According to the extent of complication in the plot, novels may be grouped into two classes,—the discursive and the compacted. Thackeray wrote novels of the former type, Hawthorne of the latter. In "Vanity Fair" there are over half a hundred characters; in "The Scarlet Letter" there are three, or possibly four. The discursive novel gives a more extensive, and the compacted novel a more intensive, view of life. English authors for the most part have tended toward the discursive type, and Continental authors toward the compacted. The latter type demands a finer and a firmer art, the former a broader and more catholic outlook on the world.

The distinction between the two types depends chiefly upon how much or how little of his entire story the author chooses to tell. In actual life, as was stated in a former chapter, there are no very ends; and it may now be added that also there are no absolute beginnings. Any event that happens is, in Whitman's words, "an acme of things accomplished" and "an encloser of things to be"; and in thinking back along its causes or forward along its effects, we may continue the series until our thought loses itself in an eternity. In any narrative, therefore, we are doomed to begin and end in mid-career; and the question is merely how extended a section of the entire imaginable and unimaginable series we shall choose to represent to the reader. For instance, it would be a very simple matter to trace the composition of Rossetti's "House of Life" back along a causal series to the birth of a boy in Arezzo in 1304; for it is hardly likely that Rossetti would have written a cycle of love sonnets if many other poets, such as Shakespeare and Ronsard, had not done so before him; and Shakespeare and Ronsard, as Mr. Sidney Lee has proved, were literary legatees of Petrarch, the aforesaid native of Arezzo. And yet, if we were to tell the story of how Rossetti's sonnets came to be composed, it is doubtful if we should go further back in time than the occasion when his friend Deverell introduced him to the beautiful daughter of a Sheffield cutler who became the immediate inspiration of his poetry of love.

Dickens, in many novels, of which "David Copperfield" may be taken as an example, has chosen to tell the entire life-story of his hero from birth up to maturity. But other novelists, like Mr. Meredith in "The Egoist," have chosen to represent events that pass, for the most part, in one place, and in an exceedingly short stretch of time. It is by no means certain that Mr. Meredith does not know as much about the boyhood and youth of Sir Willoughby Patterne as Dickens knew about the early years of David Copperfield; but he has chosen to compact his novel by presenting only a brief series of events which exhibit his hero at maturity. Surely TurgÉnieff, after writing out that dossier of each of his characters to which Mr. Henry James referred, must have known a great many events in their lives which he chose to omit from his finished novel. It is interesting to imagine the sort of plot that George Eliot would have built out of the materials of "The Scarlet Letter." Probably she would have begun the narrative in England at the time when Hester was a young girl. She would have set forth the meeting of Hester and Chillingworth and would have analyzed the causes culminating in their marriage. Then she would have taken the couple overseas to the colony of Massachusetts. Here Hester would have met Arthur Dimmesdale; and George Eliot would have expended all her powers as an analyist of life in tracing the sweet thoughts and imperious desires that led the lovers to the dolorous pass. The fall of Hester would have been the major knot in George Eliot's entire narrative. It would have stood at the culmination of the nouement of her plot: the subsequent events would have been merely steps in the dÉnouement. Yet the fall of Hester was already a thing of the past at the outset of the story that Hawthorne chose to represent. He was interested only in the after-effects of Hester's sin upon herself and her lover and her husband. The major knot, or culmination, of his plot was therefore the revelation of the scarlet letter,—a scene which would have been only an incident in George Eliot's dÉnouement. It will be seen from this that any story which is extended in its implications may offer a novelist materials for any one of several plot-structures, according to whichever section of the entire story happens most to interest his mind.

It will be seen, also, that much of the entire story must, in any case, remain unwritten. A plot is not only, as Stevenson stated, a simplification of life; it is also a further simplification of the train of events which, in simplifying life, the novelist has first imagined. The entire story, with all its implications, is selected from life; and the plot is then selected from the entire story. Often a novelist may suggest as much through deliberately omitting from his plot certain events in his imagined story as he could suggest by representing them. Perhaps the most powerful character in Mr. Meredith's "Evan Harrington" is the great Mel, whose death is announced in the very first sentence of the novel. Hawthorne, in "The Marble Faun," never clears away the mystery of Miriam's shadowy pursuer, nor tells us what became of Hilda when she disappeared for a time from the sight and knowledge of her friends.

After the novelist has selected from his entire story the materials he means to represent, and has patterned these materials into a plot, he enjoys considerable liberty in regard to the point at which he may commence his narrative. He may begin at the beginning of one or another of his main strands of causation, as Scott usually does; or he may adopt the Homeric device, commended by Horace, of plunging into the midst of his plot and working his way back only afterward to its beginning. In the first chapter of "Pendennis," the hero is seventeen years old; the second chapter narrates the marriage of his father and mother, and his own birth and boyhood; and at the outset of the third chapter he is only sixteen years of age.

It is obvious that, so long as the novelist represents his events in logical sequence, it is not at all necessary that he should present them in chronological succession. Stories may be told backward through time as well as forward. Thackeray often begins a chapter with an event that happened one day, and ends it with an event that happened several days before; he works his way backward from effects to causes, instead of forward from causes to effects. In carrying on a plot which is woven out of several strands, it is hardly ever possible to represent events in uninterrupted chronological succession, even when the author consistently works forward from causes to effects; for after he has pursued one strand of his plot to a certain point in time, he is obliged to turn backward several days or weeks, or possibly a longer period, to pick up another strand and carry it forward to the same point in time at which he left the first. Retrogression in time, therefore, is frequently not only permissible but necessary. But it is only common-sensible to state that chronological sequence should be sacrificed merely for the sake of making clear the logical relation of events; and whenever juggling with chronology tends to obscure instead of clarify that logical relation, it is evidence of an error of judgment on the part of the narrator. TurgÉnieff is often guilty of this error of judgment. He has a disconcerting habit of bringing a new character into the scene which stands for the moment before the eye of the reader, and then turning the narrative backward several years in order to recount the past life of the newcomer. Frequently, before this parenthetic recital is completed, the reader has forgotten the scene from which the author turned to the digression.

In most plots, as has been stated, the nouement is more significant than the dÉnouement, and the causes leading to the tying of the major knot are more interesting than the effects traced during the process of untying it. This is the reason why the culmination is usually set well along toward the conclusion of the story. Sometimes even, when the major knot has been tied with a Gordian intricacy, the author sets it at the very end of his narrative, and suddenly cuts it instead of carefully untying it. But there is no absolutely necessary reason why it should stand at the end, or, as is more frequently the case, at a point about three quarters through the story. It may even be set at the very beginning; and the narrative may concern itself entirely with an elaborate dÉnouement. This is the case, for example, in the detective story, where a very intricate knot is assumed at the outset, and the narrative proceeds to exhibit the prowess of the detective-hero in untying it.

A well-constructed plot, like any other sort of well-articulated pattern, is interesting in itself; and certain novels and short-stories, like Wilkie Collins' "Moonstone" and Poe's "Murders in the Rue Morgue," maintain their interest almost through the element of plot alone. But since the purpose of fiction is to represent reality, a story will fail of the highest effect unless the people acting in its pattern of events produce upon the reader the illusion of living human beings. We must therefore turn our attention next to a study of the element of character.