How Fiction Works

A Brief History of Consciousness

One reason that Cervantes needs to have Don Quixote accompanied by Sancho Panza on his travels is that the knight must have someone to talk to. When Don Quixote sends Sancho off to search for Dulcinea, and is alone for the first extended period in the novel, he does not think , as we would now understand the term. He speaks out aloud, he soliloquizes.

The novel begins in the theater, and novelistic characterization begins when the soliloquy goes inward. The soliloquy, in turn, has its origins in prayer, as we can see from Greek tragedy, or from Book 5 of The Odyssey , or from the Psalms, or from David’s songs to the Lord in 1 and 2 Samuel. Shakespeare’s heroes and heroines still use soliloquy to invoke the gods, if not quite to pray to them: “Come you spirits / That tend on mortal thoughts, unsex me here,” “Blow, winds, and crack your cheeks!” and so on. The actor comes to the front of the stage and speaks his mind to an audience, who is both God above and we spectators in the seats. Nineteenth-century novelists like Charlotte Brontë and Thomas Hardy continued to describe their characters as “soliloquizing” when speaking to themselves.

The novel has changed the art of characterization partly by changing who a character is being seen by. Consider three men, each permanently affected by a chance occurrence: King David in the Old Testament; Macbeth; and Raskolnikov in Crime and Punishment. David, strolling on his roof, sees Bathsheba, naked, bathing herself, and is instantly struck with lust. His decision to take her as his lover and wife, and to kill her inconvenient husband, sets in train a series of events that will lead to his downfall and his punishment by God. Macbeth is instantly contaminated by the suggestion of the three witches that he kill the king and take his mantle. He, too, is punished—if not explicitly by God, then by “evenhanded justice,” and by “pity, like a naked newborn babe.” And Raskolnikov, in a story clearly influenced by Shakespeare’s play, is similarly polluted by an idea—that by killing a miserable pawnbroker, he can vaunt himself over ordinary morality like a Napoleon. He, too, must “accept his punishment,” as Dostoevsky puts it, and be corrected by God.

Despite the many revelations and subtleties of the Old Testament narrative—David’s political canniness, his sorrow at the way Saul treats him, his lust for Bathsheba, his grief at the death of his son, Absalom—David remains a public character. In the modern sense, he has no privacy. He hardly ever speaks his inner thoughts to himself; he speaks to God, and his soliloquies are prayers. He is external. to us because in some way he does not exist for us , but for the Lord. He is seen by the Lord, is transparent to the Lord, but remains opaque to us. This opacity allows for a lovely margin of surprise, to use E. M. Forster’s word. For instance, David is cursed by God, who tells him, through the offices of Nathan the prophet, that the house of David will be punished, beginning with his son. And indeed, David’s son dies soon after birth. David’s response is curious. While the child is ill, he fasts and weeps, but as soon as the child has died, he washes, changes his clothes, worships God, and asks his servants to put out food. When they ask him why he has acted thus, he replies: “While the child was still alive I fasted and wept, for I thought, ‘Who knows, the LORD may favor me and the child will live.’ And now that he is dead, why should I fast? Can I bring him back again? I am going to him and he will not come back to me” (2 Samuel 12:22–23). Robert Alter, whose modern English translation this is, comments: “David here acts in a way that neither his courtiers nor the audience of the story could have anticipated.”

David’s calm, massive resignation (“I am going to him and he will not come back to me”) is beautiful, as well as surprising. David is “light in spirit.” Despite God’s curse, despite the loss of this son, and of Absalom, David dies peacefully in his bed, telling his son Solomon, “I go the way of all the earth.”

David is opaque to us, one feels, precisely because he is transparent to God, who is his real audience. What matters to the Bible writer is not the state of David’s mind, but the whole story, the entirety of the arc of David’s life. And this story, this arc, is both human and not quite human—not quite, because causation is divine as well as human. David’s life is determined partly by what he does, but the rest of his life is then overdetermined, one might say, by God’s punishment of him. In a sense, the storyteller is God, who is writing fate’s script. David does not have a mind, as we understand modern subjectivity. He has no past, to speak of, and no memory, for it is God’s memory that counts, which never forgets. And when he sees Bathsheba, what happens to him is not an idea, or at least not in the way that Jesus, that cheerless psychologist, meant when he said that for a man to look lustfully upon a woman is already to commit adultery. Jesus here announces that mental states are as important as actions. But for the writer of the David story, the mental state is precisely what is occluded; action is all: “and he saw from the roof a woman bathing, and the woman was very beautiful. And David sent and inquired after the woman, and the one he sent said, ‘Why, this is Bathsheba daughter of Eliam, wife of Uriah the Hittite.’ And David sent messengers and fetched her and she came to him and he lay with her, she having just cleansed herself of her impurity, and she returned to her house. And the woman became pregnant…” David sees, and acts. As far as the narrative is concerned, he does not think.

Macbeth is not being seen by God so much as he is being seen by us, the audience. His prayers, you might say, are soliloquies, and get very close to mental thought, as he agonizes before us over the dilemma he finds himself in. One reason for the play’s power has to do with its domestic intimacy, whereby we feel we are eavesdropping on the horrid privacy of the Macbeths’ marriage, not to mention the guilt-tainted outpourings of their monologues. At certain moments, the play seems to want to pull away and develop itself into a new form, the form of the novel. At the banquet, for instance, in Act 3, Scene 4, when Macbeth sees Banquo’s ghost, Lady Macbeth twice leans over to him, and attempts to strengthen his resolve. We have to imagine the characters almost whispering to each other in the presence of their guests. “What, quite unmanned in folly?” says Lady Macbeth. “If I stand here, I saw him,” replies Macbeth. “Fie, for shame!” is the fierce spousal response. This is always awkward when acted onstage, because the attendant lords must murmur in the background—in an unconvincing, stagy way—as if they cannot hear what is being said. The privacy of the marital conversation is what poses a theatrical difficulty: Where, onstage, can it realistically happen? I think Shakespeare is essentially being a novelist at such moments. On the page, of course, such moments have as much space to exist as the novelist feels like awarding; it is a simple matter of the adjustment of point of view (“Lady Macbeth turned quickly to her blanched lord, gripped his hand with sharp nails, and hissed at him, ‘What, quite unmanned in folly?’”).

David’s story is almost entirely public; Macbeth’s is one of publicized privacy. And this private man differs from David in being the possessor of a memory. It is the memory—“the warder of the brain”—that will not leave Macbeth alone. “My dull brain was wrought / With things forgotten,” says Macbeth, pathetically, but the play in fact bodies forth De Quincey’s terrifying, pre-Freudian admonition in his Confessions of an English Opium-Eater : “There is no such thing as forgetting.” So the real curse on the Macbeths is not theological, despite the machinery of the witches, and the ghosts; the real curse is mental, “the written troubles of the brain.” Now a character’s thought can be retrospective , can move back and forth over the present and the past, to take in a whole life:

I have lived long enough. My way of life

Is fall’n into the sere, the yellow leaf,

And that which should accompany old age,

As honour, love, obedience, troops of friends,

I must not look to have …

If Macbeth’s story is one of publicized privacy, Raskolnikov’s story is one of scrutinized privacy. God still exists, but he is not watching Raskolnikov—at least, not until the end of the novel when Raskolnikov accepts Christ. Until that moment, Raskolnikov is being watched by us, the readers. The crucial difference between this and the theater is that we are invisible. In David’s story the audience is in some important way irrelevant; in Macbeth’s the audience is visible and silent, and soliloquy does indeed have the feeling not only of an address to an audience but of a conversation with an interlocutor—us—who will not respond, a blocked dialogue. In Raskolnikov’s story the audience—the reader—is invisible but all-seeing; so the reader has replaced David’s God and Macbeth’s audience.

What are the implications of this enormous shift? The obvious one is that soliloquy does not have to be voiced, and can get closer to being true mental speech. The hero is released from the tyranny of necessary eloquence; he is an ordinary man. (This is just what Raskolnikov cannot bear.) Inner soliloquy can indulge in repetition, ellipsis, hysteria, vagueness—mental stutter. If Shakespeare’s characters often seem to be overhearing themselves when they soliloquize, 1 we are now overhearing Raskolnikov. There is no facet of his soul not tilted toward us. Something else worth noticing is that while David has no mind, to speak of, and Macbeth’s mind is punished, Raskolnikov’s mind is the author of its own woe; the idea of murdering the woman was his free invention.

Under the new dispensation of the invisible audience, the novel becomes the great analyst of unconscious motive, since the character is released from having to voice his motives: the reader becomes the hermeneut, looking between the lines for the actual motive. On the other hand, the absence of a visible audience seems to make the ordinary man seek an audience, in ways that would have seemed grotesque to lordly figures like the Macbeths. Many of the characters in Crime and Punishment seem compelled to act out horrid pantomimes and melodramas, in which they stage a version of themselves, for effect. David and Macbeth were men of action—you might say they were naturally dramatic (they knew who their audiences were); Raskolnikov is unnaturally theatrical, or better still, histrionic: he seeks attention, and he is desperately unstable and inauthentic, hiding at one moment, confessing at another, proud in one scene, self-abasing in the next. In the novel, we can see the self better than any literary form has yet allowed; but it is not going too far to say that the self is driven mad by being so invisibly scrutinized.

The novel has shown a stunning technical progression in its ability to render plot, and in making us attend to psychological motivation. In his essay “The End of the Novel,” Osip Mandelstam claimed that “the novel was perfected and strengthened over an extremely long period of time as the art form to interest the reader in the fate of the individual,” singling out two technical refinements:

1. The transformation of biography (the saint’s life, the moralizing Theophrastan sketch, and so on) into a meaningful narrative or plot.

2. “Psychological motivation.”

Adam Smith, in his Lectures on Rhetoric and Belles Lettres , complains of the relatively young form of the novel that “as newness is the only merit in a Novel and curiosity the only motive which induces us to read them, the writers are necessitated to make use of this method [i.e., suspense] to keep it up.” This is an early, mid-eighteenth- century attack on the mindlessness of suspense—the kind of complaint nowadays made routinely about thrillers and pulp fiction.

But the novel soon showed itself willing to surrender the essential juvenility of suspense, in favor of what Viktor Shklovsky calls “unconsummated” stories with “false endings” (he was referring to Flaubert and Chekhov, respectively). 2 To return to the case of Iris Murdoch, who so wanted to create free characters and so often failed, her failure is not one of psychological attention or metaphysical shallowness—quite the opposite—but a Fielding-like devotion to excessive plot-making. Her improbable, melodramatic, feeble stories, still highly indebted to eighteenth- and nineteenth-century theatrics, are not adult enough to take the strain of her complex moral analysis. 3

As Mandelstam suggested, the novel probably has its origins in a secular response to the religious lives and biographies of saints and holy men, and in the tradition inaugurated by the Greek writer Theophrastus, who offered a series of sketches of types—the miser, the hypocrite, the fond and foolish lover, and so on. (Don Quixote belongs to the modern novel in part because Cervantes is so determined to discredit the “holy,” chivalric tales of Arthur and Amadis of Gaul.) Since these were discrete portraits, you could not contrast one with the other.

The Theophrastan and religious tendency remained strong in the novel throughout the eighteenth and nineteenth centuries, and is still visible in cinema and in various kinds of pulp fiction: villains are villains, heroes are heroes, and the good and the bad are clearly separated and delineated—think of Fielding, Goldsmith, Scott, Dickens, Waugh. Character is essentially stable, has fixed attributes, in such writers.

But at the same time, another kind of novel was developing, in which good and bad wars within a single character, and the self refuses to stay still. What the novel powerfully began to do was to explore characterological relativity. This tradition in turn would influence the English and American novel in the early decades of the twentieth century, particularly as Dostoevsky began to be translated into English (Lawrence, Conrad, Ford, and Woolf were the chief beneficiaries). And all this can be traced pretty much to one extraordinary novel, Rameau’s Nephew , written by Denis Diderot in the 1760s, though not published until 1784. In this furious dialogue (it is laid out on the page like a play) an obscure nephew of the famous composer Jean-Philippe Rameau engages in a fictional encounter with an interlocutor named “Diderot.” Rameau’s nephew seems, at first, a recognizable enough French type—a sophisticated cynic, a man who sees through society, a Juvenal of the Luxembourg Gardens. But Diderot’s brilliant addition is to complicate this figure by making him resentfully dependent on his famous uncle, the composer. Rameau’s nephew plays party tricks like mimicking his uncle’s music, which he says he finds boring, by sitting at a fake piano, playing fake music, all the while grimacing and sweating and crooning. He is very unstable; he is described by Diderot as someone who changes from month to month. There is a hollowness, too, because he wants to be famous: “I would like to be somebody else, at the risk of being a man of genius, a great man … Yes, yes, I am mediocre and angry.” He says that he has never listened to his uncle’s music without ruefully thinking, “That’s what you will never do.” He dilates on his envy: “I who have composed keyboard pieces which nobody plays, but which may well be the only ones to go down to future generations who will.” He admires the criminal’s daring alienation from society, as Raskolnikov will do.

Where his interlocutor—the Diderot figure—sees reason and order in society, Rameau sees only hypocrisy. The Diderot figure says that he constantly reads “La Bruyè re, Moliè re, and Theophrastus”—the didactic creators of stable, moralizing, satiric characterization. He says that such writers teach knowledge of one’s duties, love of virtue, hatred of vice, which is exactly what we would expect him pompously to say. Rameau’s nephew replies that all he has learned from these writers is the value of fraudulence and hypocrisy: “When I read Tartuffe I tell myself: ‘be a hypocrite, by all means, but don’t talk like one. Keep the vices that come in useful to you, but don’t have either the tone or the appearance, which would expose you to ridicule.’” (Through this conversation, Diderot makes his comment on the kind of simpler characterization that his own book has outstripped.) Rameau’s nephew is a jester, a fool, but the richness of the book lies in its subtle suggestions that he may be a kind of frustrated genius, possibly more talented than his uncle.

From this character flows much of the psychological flamboyance and acuity of Stendhal, Dostoevsky, Hamsun, Conrad, Italo Svevo, Ralph Ellison’s Invisible Man , and Wittgenstein’s Nephew , in which Thomas Bernhard, following Diderot, raises the possibility that Paul Wittgenstein, the famous philosopher’s nephew, was a greater philosopher than his uncle precisely because he did not write his philosophy down.

Look at what Stendhal does with this inheritance in The Red and the Black , published in 1830: Julien Sorel is strikingly unpredictable. Like Diderot’s Rameau, Julien seethes with satiric callousness, self-interested impropriety, and gratuitous resentments. He is determined to make Madame de Rê nal love him not out of any natural impulse but in the proud belief that this is how to conquer society, and how to repay the slights he feels she has given him: “Julien said to himself, What do I know of this woman’s character? Only this—before I went away, I took hold of her hand and she withdrew it; today I withdraw my hand and she takes hold of it and presses it. There’s a fine opportunity of paying her back all the contempt she has felt for me. Heaven knows how many lovers she’s had. Possibly she only decided to favour me because it’s so easy for us to meet.” 4

Stendhal’s superb addition to this complex creature is subtly to reveal that, whatever he may say to himself, Julien is actually, unconsciously, in love with Madame de Rê nal. (This is the kind of novelistic psychological subtlety not easily available to Diderot’s dialogue form.) Julien is an uncannily wise portrait because he is in fact nobler than his egotism. His mantra is “Each man for himself in this desert of egoism men call life,” a suitably French cynicism. But he can’t actually live like this. He is too passionate, too noble. Like Diderot’s Rameau, he reveres Tartuffe. But he is nothing like as brilliantly intelligent or all-seeing as the Rameau character, and this is Stendhal’s great, novelistic enrichment. Julien poses as a fearsome truth-teller, but he is just an intelligent, undereducated romantic provincial who is not quite clever or supple enough, his mind full of blowsy Napoleonic ardor. We, the readers, can see this. His understanding fluctuates; occasionally he sees clearly, but more often he cannot read upper-class society’s codes as well as he thinks he can. He is proudly hypocritical, but not always hypocritical enough to see the need to hide his obvious hypocrisy—he is always blurting out his heart to people when he should keep it closed.

In Paris, Julien falls in love with the highborn Mathilde, the daughter of his employer. Each lover wants to be the other’s love slave; but each is also too proud for this, and simultaneously wants to be the other’s master. Mathilde is romantically in love with Julien’s proud exceptionalism, but feels that it would be beneath her to marry a servant; Julien loves her but is fearful of being patronized. Dostoevsky, writing between the 1840s and 1881, and a keen reader of the French, would become an even greater novelist of the workings of this kind of pride and abasement. There is a direct link from Rousseau to Diderot to Dostoevsky.

In a famous scene in Notes from Underground , published in 1864, the narrator, an insignificant but proudly rebellious outcast, has an encounter with an impressive-looking cavalry officer in a tavern. The officer, blocked by the narrator, casually picks him up and moves him out of the way. The narrator is humiliated, and can’t sleep for his dreams of revenge. He knows that this same officer walks every day down the Nevsky Prospect. The narrator follows him, “admiring” him from a distance. He determines that he will walk in the opposite direction and that when the two men meet, he, the narrator, will not budge an inch. But whenever the encounter arrives, he panics, and moves out of the way just as the officer strides past. At night he wakes, obsessively turning over the question: “Why is it invariably I who swerves first? Why precisely me and not him?” Eventually he holds his ground, the two men brush shoulders, and the narrator is overjoyed. He returns home singing Italian arias, feeling properly avenged. But the satisfaction lasts only a few days.

Dostoevsky was the great analyst—in a sense, almost the inventor—of the psychological category that Nietzsche called ressentiment . Again and again, Dostoevsky shows how pride is really very close to humility, and how hate is very close to a kind of sick love, in just the way that Ra meau’s nephew is far more dependent on the existence of his famous uncle than he will actually admit to, or in the way that Julien both loves and hates Madame de Rê nal and Mathilde. In the Nevsky Prospect anecdote, the weaker man loathes but “admires” the officer—and in a sense, loathes him because he admires him. His impotence has less to do with his actual circumstances than with his imaginary relation to the officer, which is one of impotent dependence. Dostoevsky would call this psychological torment the “Underground,” meaning a kind of poisonous, impotent alienation, a chronic instability of self, and a vaunting pride that at any moment might unexpectedly crash into its inverse—cringing self-abasement. 5

Nothing in fiction, not even in Diderot and Stendhal, quite prepares one for Dostoevsky’s characters. In The Brothers Karamazov , for in stance, the clownish Fyodor Pavlovich is about to enter a dining room at the local monastery. He has already acted terribly in the cell of the saintly monk, Father Zosima. Fyodor decides that he will act scandalously in the dining room also. Why? Because, he thinks to himself, “it always seems to me, when I go somewhere, that I am lower than everyone else and that they all take me for a buffoon—so let me indeed play the buffoon, because all of you, to a man, are lower than I am.” And as he thinks this, he remembers being asked once why he hated a certain neighbor, to which he had replied: “He never did anything to me it’s true, but I once played a most shameless nasty trick on him, and the moment I did it, I immediately hated him for it.”

Dostoevskian character has at least three layers. On the top layer is the announced motive: Raskolnikov, say, proposes several justifications for his murder of the old woman. The second layer involves unconscious motivation, those strange inversions wherein love turns into hate and guilt expresses itself as poisonous, sickly love. Thus Raskolnikov’s mad need to confess his crime to the police and to Sonia the prostitute presages Freud’s comment on the action of the superego: “In many criminals,” writes Freud, “especially youthful ones, it is possible to detect a very powerful sense of guilt which existed before the crime, and is therefore not its result but its motive.” Or in the case of Fyodor Karamazov and his desire to punish the neighbor to whom he was once nasty, you could say that guilt is causing him, unconsciously, to be horrible to his neighbor; his behavior recalls the quip—both funny and deadly serious—of the Israeli psychoanalyst who remarked that the Germans would never forgive the Jews for the Holocaust. The third and bottom layer of motive is beyond explanation and can only be understood religiously. These characters act like this because they want to be known ; even if they are unaware of it, they want to reveal their baseness; they want to confess. They want to reveal the dark shamefulness of their souls, and so, without knowing quite why, they act “scandalously” and appallingly in front of others, so that people “better” than they can judge them for the wretches they are.

There is something deeply philosophical about Dostoevsky’s analysis of human behavior, and Nietzsche and Freud were attracted to his work. (One chapter of Dostoevsky’s novella The Eternal Husband is entitled “Analysis.”) Proust, who said that all of Dostoevsky’s novels might have the one title, Crime and Punishment , studied him with perhaps more care than he would admit to. It is Proust who elaborates and develops the philosophical analysis of psychological motive. In Proust, you can see every element of characterization—and indeed of fiction-making itself—living happily together, as if you were watching schools of fish underneath a glass-bottomed boat. Thus, his characters are in some sense both externally seen and yet highly inward; they are “flat,” but are extensively analyzed by Proust as if they were “round”; and of course the novel is so massive that their flatness gets elongated over time, and no longer seems like flatness. Proust is not afraid of caricature, and positively loves to “tag” his characters with leitmotifs, or repeated “characteristics,” in a Dickensian way—as, say, Marcel’s grandfather likes repeatedly to say “On guard! On guard!” and Mme Verdurin always complains of getting a headache whenever music is played. He uses this method to “fix” his characters, just as the earliest novelists did, or, nearer to his own time, as Dickens, Tolstoy, and Mann all did.

But his fiction simultaneously mounts a revolt against the tyranny of fixed, Theophrastan “characteristics.” Combray is presented as a closed world in which everyone knows everyone else, and Marcel’s family is seen as having a supremely secure sense—sustained largely by their system of “tagging” friends and acquaintances with leitmotifs—of what people are like. When someone informs Marcel’s aunt that she has just seen a stranger in the village, she wants to send the maid over to Camus, the chemist, 6 to ask him who this can be: the very notion of someone not known to the family is an outrage. But as Proust puts it, “our social personality is a creation of the thoughts of other people.” His characters in fact change in unexpected ways, and we must constantly adjust the optic that we use to view these people. Marcel’s family is sure that they know M. Swann thoroughly; but Proust reveals that they have seen only one side of him, and the least authentic at that. Likewise, Swann falls in love with Odette, because, in part, she reminds him of a woman in a painting; but over the course of many strenuous months, he will find that one of the dangers of love is that it encourages us to fix in our amorous minds a picture of the beloved. Sometimes these alterations are caused by the smallest gestures and revelations, and are in themselves mysterious in origin. Marcel changes his idea of M. Legrandin because he glimpses him talking animatedly to someone and bowing in a peculiar way:

This rapid straightening-up caused a sort of tense muscular wave to ripple over Legrandin’s rump, which I had not supposed to be so fleshy; I cannot say why, but this undulation of pure matter, this wholly carnal fluency devoid of spiritual significance … awoke my mind suddenly to the possibility of a Legrandin altogether different from the one we knew. 7

Progress! In Fielding and Defoe, even in the much richer Cervantes, revelation of this altering kind occurs at the level of plot—an unexpected sister, a lost will, and so on. It does not alter our conception of a character. Don Quixote, though an infinitely deep comic idea, is the same kind of character at the end of the book as at the beginning. (That is why his deathbed change of heart is so disconcerting.)

The Russians and the French essentially set the terms of the modernist novel as it flourished in Britain and America between 1920 and 1945. You can trace the excitement of encounter in Virginia Woolf’s essays, especially those written in the teens and the twenties of the century, as she discovered the new translations of the Russians into English by Constance Garnett. She put it like this in “Mr Bennett and Mrs Brown” (1923):

After reading Crime and Punishment and The Idiot , how could any young novelist believe in “characters” as the Victorians had painted them? For the undeniable vividness of so many of them is the result of their crudity. The character is rubbed into us indelibly because its features are so few and so prominent. We are given the keyword [e.g., “I never will desert Mr. Micawber”] and then, since the keyword is astonishingly apt, our imaginations swiftly supply the rest. But what keyword could be applied to Raskolnikov, Mishkin, Stavrogin, or Alyosha? These are characters without any features at all. We go into them as we descend into some enormous cavern.

Ford Madox Ford agreed (though his master was Flaubert). Other than Richardson, he argued in his deliciously biased book The English Novel , nothing was worthy of adult attention in English fiction until Henry James came along. For Ford, the serious European novel began with Diderot:

It was to Diderot—and still more to Stendhal—that the Novel owes its next great step forward … At that point it became suddenly evident that the Novel as such was capable of being regarded as a means of profoundly serious and many-sided discussion and therefore as a medium of profoundly serious investigation into the human case. It came into its own.

This new approach to character meant a new approach to form. When character is stable, form is stable and linear—the novelist begins at the beginning, telling us about his hero’s childhood and education, moving decisively forward into the hero’s marriage, and then toward the dramatic crux of the book (something is wrong with the marriage). But if character is changeable, then why begin at the beginning? Surely it would be more effective to begin in the middle, and then move backward, and then move forward, and then move backward again? This is just the kind of form Conrad would use in Lord Jim and The Secret Agent , and Ford in The Good Soldier . Ford, again, in his memoir of Conrad:

What was the matter with the Novel, and the British novel in particular, was that it went straight forward, whereas in your gradual making acquaintanceship with your fellows you never do go straight forward. You meet an English gentleman at your golf club. He is beefy, full of health, the model of the boy from an English public school of the finest type. You discover, gradually, that he is hopelessly neurasthenic, dishonest in matters of small change, but unexpectedly self-sacrificing, a dreadful liar, but a most painfully careful student of Lepidoptera and, finally, from the public prints, a bigamist who was once, under another name, hammered on the Stock Exchange … To get such a man in fiction you could not begin at his beginning and work his life chronologically to the end. You must first get him in with a strong impression, and then work backwards and forwards over his past …

Is it contradictory to have defended the flatness of characters while simultaneously arguing that the novel has become a more sophisticated analyst of deep, self-divided characters? No, if one resists both Forster’s idea of flatness (flatness is more interesting than he makes it out to be) and his idea of roundness (roundness is more complicated than he makes it out to be). In both cases, subtlety of analysis is what is important.