Since he practically always does so, readily enough, it may seem unnecessary to insist upon the matter. Not often have we seen a novelist pushing his self-denial beyond reason, rejecting the easy way for the difficult without good cause. But in order to make sure of breaking a sound rule at the right point, and not before—to take advantage of laxity when strictness becomes unrewarding, and only then—it is as well to work both ways, from the easy extreme to the difficult and back again. The difficult extreme, in fiction, is the dramatic rule absolute and unmitigated; having reached it from the other end, having begun with the pictorial summary and proceeded from thence to drama, we face the same stages reversed. And it is now, I think, that we best appreciate the liberties taken with the resources of the novelist by Balzac. His is a case that should be approached indirectly. If one plunges straight into Balzac, at the beginning of criticism, it is hard to find the right line through the abundance of good and bad in his books; there is so much of it, and all so strong and staring. It looks at first sight as though his good and his bad alike were entirely conspicuous and unmistakable. His devouring passion for life, his grotesque romance, his truth and his falsity, these cover the whole space of the ComÉdie between them, and nobody could fail to recognize the full force of either. He is tremendous, his taste is abominable—what more is there to say of Balzac? And that much has been said so often, in varied words, that there can be no need to say it again for the ten-thousandth time.

Such is the aspect that Balzac presents, I could feel, when a critic tries to face him immediately; his obviousness seems to hide everything else. But if one passes him by, following the track of the novelist's art elsewhere, and then returns to him with certain definite conclusions, his aspect is remarkable in quite a new way. His badness is perhaps as obvious as before; there is nothing fresh to discover about that. His greatness, however, wears a different look; it is no longer the plain and open surface that it was. It has depths and recesses that did not appear till now, enticing to criticism, promising plentiful illustration of the ideas that have been gathered by the way. One after another, the rarer, obscurer effects of fiction are all found in Balzac, behind his blatant front. He illustrates everything, and the only difficulty is to know where to begin.

The effect of the generalized picture, for example, supporting the play of action, is one in which Balzac particularly delights. He constantly uses it, he makes it serve his purpose with a very high hand. It becomes more than a support, it becomes a kind of propulsive force applied to the action at the start. Its value is seen at its greatest in such books as Le CurÉ de Village, PÈre Goriot, La Recherche de l'Absolu, EugÉnie Grandet—most of all, perhaps, in this last. Wherever, indeed, his subject requires to be lodged securely in its surroundings, wherever the background is a main condition of the story, Balzac is in no hurry to precipitate the action; that can always wait, while he allows himself the leisure he needs for massing the force which is presently to drive the drama on its way. Nobody gives such attention as Balzac does in many of his books, and on the whole in his best, to the setting of the scene; he clearly considers these preparatory pictures quite as important as the events which they are to enclose.

And so, in PÈre Goriot, all the potent life of the Maison Vauquer is deliberately collected and hoarded up to the point where it is enough, when it is let loose, to carry the story forward with a strong sweep. By the time the story itself is reached the Maison Vauquer is a fully created impression, prepared to the last stroke for the drama to come. Anything that may take place there will have the whole benefit of its setting, without more ado; all the rank reality of the house and its inmates is immediately bestowed on the action. When the tale of Goriot comes to the front it is already more than the tale of a certain old man and his woes. Goriot, on the spot, is one of Maman Vauquer's boarders, and the mere fact is enough, by now, to differentiate him, to single him out among miserable old men. Whatever he does he carries with him the daily experience of the dingy house and the clattering meals and the frowzy company, with Maman Vauquer, hard and hungry and harassed—Mrs. Todgers would have met her sympathetically, they would have understood each other—at the head of it. Into Goriot's yearnings over his fashionable daughters the sounds and sights and smells of his horrible home have all been gathered; they deepen and strengthen his poor story throughout. Balzac's care in creating the scene, therefore, is truly economical; it is not merely a manner of setting the stage for the drama, it is a provision of character and energy for the drama when it begins.

His pictures of country towns, too, Saumur, Limoges, AngoulÊme, have the same kind of part to play in the ScÈnes de la vie de province. When Balzac takes in hand the description of a town or a house or a workshop, he may always be suspected, at first, of abandoning himself entirely to his simple, disinterested craving for facts. There are times when it seems that his inexhaustible knowledge of facts is carrying him where it will, till his only conscious purpose is to set down on paper everything that he knows. He is possessed by the lust of description for its own sake, an insatiable desire to put every detail in its place, whether it is needed or no. So it seems, and so it is occasionally, no doubt; there is nothing more tiresome in Balzac than his zest, his delight, his triumph, when he has apparently succeeded in forgetting altogether that he is a novelist. He takes a proper pride in Grandet or Goriot or Lucien, of course; but his heart never leaps quite so high, it might be thought, as when he sees a chance for a discourse upon money or commerce or Italian art. And yet the result is always the same in the end; when he has finished his lengthy research among the furniture of the lives that are to be evoked, he has created a scene in which action will move as rapidly as he chooses, without losing any of its due emphasis. He has illustrated, in short, the way in which a pictorial impression, wrought to the right pitch, will speed the work of drama—will become an effective agent in the book, instead of remaining the mere decorative introduction that it may seem to be.

Thus it is that Balzac was able to pack into a short book—he never wrote a long one—such an effect of crowds and events, above all such an effect of time. Nobody knows how to compress so much experience into two or three hundred pages as Balzac did unfailingly. I cannot think that this is due in the least to the laborious interweaving of his books into a single scheme; I could believe that in general a book of Balzac's suffers, rather than gains, by the recurrence of the old names that he has used already elsewhere. It is an amusing trick, but exactly what is its object? I do not speak of the ordinary "sequel," where the fortunes of somebody are followed for another stage, and where the second part is simply the continuation of the first in a direct line. But what of the famous idea of making book after book overlap and encroach and entangle itself with the rest, by the device of setting the hero of one story to figure more or less obscurely in a dozen others? The theory is, I suppose, that the characters in the background and at the corners of the action, if they are Rastignac and Camusot and Nucingen, retain the life they have acquired elsewhere, and thereby swell the life of the story in which they reappear. We are occupied for the moment with some one else, and we discover among his acquaintances a number of people whom we already know; that fact, it is implied, will add weight and authority to the story of the man in the foreground—who is himself, very likely, a man we have met casually in another book. It ought to make, it must make, his situation peculiarly real and intelligible that we find him surrounded by familiar friends of our own; and that is the artistic reason of the amazing ingenuity with which Balzac keeps them all in play.

Less artistic and more mechanical, I take it, his ingenuity seems than it did of old. I forget how few are the mistakes and contradictions of which Balzac has been convicted, in the shuffling and re-shuffling of his characters; but when his accuracy has been proved there still remains the question of its bearing upon his art. I only touch upon the question from a single point of view, when I consider whether the density of life in so many of his short pieces can really owe anything to the perpetual flitting of the men and women from book to book. Suppose that for the moment Balzac is evoking the figure and fortunes of Lucien de RubemprÉ, and that a woman who appears incidentally in his story turns out to be our well-remembered Delphine, Goriot's daughter. We know a great deal about the past of Delphine, as it happens; but at this present juncture, in Lucien's story, her past is entirely irrelevant. It belongs to another adventure, where it mattered exceedingly, an adventure that took place before Lucien was heard of at all. As for his story, and for the reality with which it may be endowed, this depends solely upon our understanding of his world, his experience; and if Delphine's old affairs are no part of it, our previous knowledge of her cannot help us with Lucien. It detracts, rather, from the force of his effect; it sets up a relation that has nothing to do with him, a relation between Delphine and the reader, which only obstructs our view of the world as Lucien sees it. Of the characters in the remoter planes of the action (and that is Delphine's position in his story) no more is expected than their value for the purpose of the action in the foreground. That is all that can be used in the book; whatever more they may bring will lie idle, will contribute nothing, and may even become an embarrassment. The numberless people in the ComÉdie who carry their lengthening train of old associations from book to book may give the ComÉdie, as a whole, the look of unity that Balzac desired; that is another point. But in any single story, such of these people as appear by the way, incidentally, must for the time being shed their irrelevant life; if they fail to do so, they disturb the unity of the story and confuse its truth.

Balzac's unrivalled power of placing a figure in its surroundings is not to be explained, then, by his skill in working his separate pieces together into one great web; the design of the Human Comedy, so largely artificial, forced upon it as his purpose widened, is no enhancement of the best of his books. The fullness of experience which is rendered in these is exactly the same—is more expressive, if anything—when they are taken out of their context; it is all to be attributed to their own art. I come back, therefore, to the way in which Balzac handled his vast store of facts, when he set out to tell a story, and made them count in the action which he brought to the fore. He seldom, I think, regards them as material to be disguised, to be given by implication in the drama itself. He is quite content to offer his own impression of the general landscape of the story, a leisurely display which brings us finally to the point of action. Then the action starts forward with a reserve of vigour that helps it in various ways. The more important of these, as I see them, will be dealt with in the next chapter; but meanwhile I may pick out another, one that is often to be seen in Balzac's work and that he needed only too often. It was not the best of his work that needed it; but the effect I mean is an interesting one in itself, and it appeals to a critic where it occurs. It shows how a novelist, while in general seeking to raise the power of his picture by means of drama, will sometimes reverse the process, deliberately, in order to rescue the power of his drama from becoming violence. If fiction always aims at the appearance of truth, there are times when the dramatic method is too much for it, too searching and too betraying. It leaves the story to speak for itself, but perhaps the story may then say too much to be reasonably credible. It must be restrained, qualified, toned down, in order to make its best effect. Where the action, in short, is likely to seem harsh, overcharged, romantic, it is made to look less so, less hazardous and more real, by recourse to the art of the picture-maker.

Balzac, it cannot be denied, had frequent cause to look about him for whatever means there might be of extenuating, and so of confirming, an incredible story. His passion for truth was often in conflict with his lust for marvels, and the manner in which they were mixed is the chief interest, I dare say, of some of his books. See him, for example, in the Splendeurs et MisÈres des Courtisanes, trying with one hand to write a novel of Parisian manners, with the other a romance of mystery, and to do full justice to both. Trompe-la-Mort, the Napoleon of crime, and Esther, the inspired courtesan, represent the romance, and Balzac sets himself to absorb the extravagant tale into a study of actual life. If he can get the tale firmly embedded in a background of truth, its falsity may be disguised, the whole book may even pass for a scene of the human comedy; it may be accepted as a piece of reality, on the same level, say, as EugÉnie Grandet or Les Parents Pauvres. That is evidently his aim, and if only his romance were a little less gaudy, or his truth not quite so true, he would have no difficulty in attaining it; the action would be subdued and kept in its place by the pictorial setting. The trouble is that Balzac's idea of a satisfying crime is as wild as his hold upon facts is sober, so that an impossible strain is thrown upon his method of reconciling the two. Do what he will, his romance remains staringly false in its contrast with his reality; there is an open gap between the wonderful pictures of the town in Illusions Perdues and the theatrical drama of the old convict which they introduce. Yet his method was a right one, though it was perverse of Balzac to be occupied at all with such devices, when he might have rejected his falsity altogether. In another man's work, where there is never this sharp distinction between true and false, where both are merged into something different from either—in Dickens's work—the method I refer to is much more successfully followed; and there, in any of Dickens's later books, we find the clearest example of it.

I have already been reminded of Stevenson's word upon this matter; Stevenson noted how Dickens's way of dealing with his romantic intrigues was to lead gradually into them, through well-populated scenes of character and humour; so that his world is actual, its air familiar, by the time that his plot begins to thicken. He gives himself an ample margin in which to make the impression of the kind of truth he needs, before beginning to concentrate upon the fabulous action of the climax. Bleak House is a very good case; the highly coloured climax in that book is approached with great skill and caution, all in his most masterly style. A broad stream of diversified life moves slowly in a certain direction, so deliberately at first that its scope, its spread, is much more evident than its movement. The book is a big survey of a quantity of odd and amusing people, and it is only by degrees that the discursive method is abandoned and the narrative brought to a point. Presently we are in the thick of the story, hurrying to the catastrophe, without having noticed at all, it may be, that our novel of manners has turned into a romantic drama, with a mysterious crime to crown it. Dickens manages it far more artfully than Balzac, because his imagination is not, like Balzac's, divided against itself. The world which he peopled with Skimpole and Guppy and the Bayham Badgers was a world that could easily include Lady Dedlock, for though she is perhaps of the theatre, they are certainly not of the common earth. They and she alike are at the same angle to literal fact, they diverging one way, she another; they accordingly make a kind of reality which can assimilate her romance. Dickens was saved from trying to write two books at once by the fact that one completely satisfied him. It expressed the exciting, amazing, exhilarating world he lived in himself, with its consistent transmutation of all values, and he knew no other.

The method which he finally worked out for himself was exactly what he required. There might be much to say of it, for it is by no means simple, but I am only concerned with one or two points in it. The chief characteristic I take to be this careful introduction of violent drama into a scene already prepared to vouch for it—a scene so alive that it compels belief, so queer that almost anything might happen there naturally. The effect which Dickens gets from the picture in his novels, as opposed to the action, is used as a sort of attestation of the action; and it surely fulfils its mission very strikingly in the best of his work—the best from this point of view—Bleak House, Dombey and Son, Our Mutual Friend. His incurable love of labyrinthine mystification, when it really ran away with him, certainly defeated all precautions; not even old Dorrit's Marshalsea, not even Flora and Mr. F.'s Aunt, can do anything to carry off the story of the Clennams. But so long as he was content with a fairly straightforward romance, all went well; the magnificent life that he projected was prepared to receive and to speed it. Blimber and Mrs. Pipchin and Miss Tox, the Podsnaps and Twemlow and the Veneerings, all contribute out of their overflow of energy to the force of a drama—a drama in which they may take no specific part, but which depends on them for the furnishing of an appropriate scene, a favouring background, a world attuned. This and so much more they do that it may seem like insulting them even to think for a moment of their subordination to the general design, which is indeed a great deal less interesting than they. But Dickens's method is sound and good, and not the less so because he used it for comparatively trivial purposes. It is strange that he should have known how to invent such a scene, and then have found no better drama to enact on it—strange and always stranger, with every re-reading. That does not affect his handling of a subject, which is all that I deal with here.

The life which he creates and distributes right and left, in such a book as Bleak House, before bending to his story—this I call his picture, for picture it is in effect, not dramatic action. It exhibits the world in which Lady Dedlock is to meditate murder, the fog of the suit in Chancery out of which the intrigue of the book is to emerge. It is the summary of a situation, with its elements spreading widely and touching many lives; it gathers them in and gives an impression of them all. It is pictorial as a whole, and quite as much so as any of Thackeray's broad visions. But I have noted before how inevitably Dickens's picture, unlike Thackeray's, is presented in the form of scenic action, and here is a case in point. All this impression of life, stretching from the fog-bound law courts to the marshes of Chesney Wold, from Krook and Miss Flite to Sir Leicester and Volumnia, is rendered as incident, as a succession of particular occasions—never, or very seldom, as general and far-seeing narrative, after Thackeray's manner. Dickens continually holds to the immediate scene, even when his object is undramatic; he is always readier to work in action and dialogue than to describe at large; he is happier in placing a character there before us, as the man or woman talked and behaved in a certain hour, on a certain spot, than in reflecting a long impression of their manner of living. In Thackeray's hands the life of Miss Flite, for instance, would have become a legend, recalled and lingered over, illustrated by passing glimpses of her ways and oddities. With Dickens she is always a little human being who figures upon a scene, in a group, a visible creature acting her small part; she is always dramatic.

And Dickens, using this method everywhere, even in such a case as hers—even where his purpose, that is to say, is pictorial, to give the sense of a various and vivacious background—is forced to crystallize and formulate his characters very sharply, if they are to make their effect; it is why he is so often reduced to the expedient of labelling his people with a trick or a phrase, which they have to bring with them every time they appear. Their opportunities are strictly limited; the author does not help them out by glancing freely into their lives and sketching them broadly. Flite, Snagsby, Chadband and the rest of them—whatever they are, they must be all of it within narrow bounds, within the few scenes that can be allotted to them; and if one of them fails now and then it is not surprising, the wonder is that most of them succeed so brilliantly. In thus translating his picture into action Dickens chose the most exigent way, but it was always the right way for him. He was curiously incapable in the other; when occasionally he tries his hand at picture-making, in Thackeray's manner—attempting to summarize an impression of social life among the Veneerings, of official life among the Barnacles—his touch is wild indeed. Away from a definite episode in an hour prescribed he is seldom at ease.

But though the actual presentation is thus dramatic, his books are in fact examples of the pictured scene that opens and spreads very gradually, in order to make a valid world for a drama that could not be precipitated forthwith, a drama that would be naked romance if it stood by itself. Stevenson happened upon this point, with regard to Dickens, in devising the same method for a story of his own, The Wrecker, a book in which he too proposed to insinuate an abrupt and violent intrigue into credible, continuous life. He, of course, knew precisely what he was doing—where Dickens followed, as I suppose, an uncritical instinct; the purpose of The Wrecker is clearly written upon it, and very ingeniously carried out. But I doubt whether Stevenson himself noticed that in all his work, or nearly, he was using an artifice of the same kind. He spoke of his habitual inclination towards the story told in the first person as though it were a chance preference, and he may not have perceived how logically it followed from the subjects that mostly attracted him. They were strongly romantic, vividly dramatic; he never had occasion to use the first person for the effect I considered a while ago, its enhancement of a plain narrative. I called it the first step towards the dramatization of a story, and so it is in a book like Esmond, a broadly pictured novel of manners. But it is more than this in a book like The Master of Ballantrae, where the subject is a piece of forcible, closely knit action. The value of rendering it as somebody's narrative, of placing it in the mouth of a man who was there on the spot, is in this book the value of working the drama into a picture, of passing it through a man's thought and catching his reflection of it. As the picture in Esmond is enhanced, so the drama in Ballantrae is toned and qualified by the method of presentation. The same method has a different effect, according to the subject upon which it is used; as a splash of the same grey might darken white surface and lighten a black. In Esmond the use of the first person raises the book in the direction of drama, in Ballantrae it thrusts the book in the other direction, towards the pictured impression. So it would seem; but perhaps it is a fine distinction that criticism can afford to pass by.