Attempts at Description are Stupid
A lesson in the pitfalls of realism from George Eliot.
Note: This morning I saw a post from my friend Alexander Chee about reading Middlemarch for the first time. Someone said, “I love Eliot’s authorial intrusions,” and Alex wrote, “She relishes them.” That took me back to this lecture I gave on the unexpected joys of reading George Eliot as an experimental writer—back in 2009. The audience was MFA students in creative writing, and my tone is, I have to say, kind of dry and pedantic. I don’t write lectures like this anymore. But I agree with every word.
George Eliot’s most quoted statement about the nature of fiction comes from her first novel, Adam Bede, in the form of a chapter-length digression in the middle of the novel (Book Two, Chapter 17). In this chapter, addressing the reader directly, she invokes a hypothetical question any nineteenth-century novelist would be expected to take very seriously: why can’t you make your characters better than they are? That is, more noble, more virtuous, more up-to-date, more expressive of general public opinion, or the author’s presumed opinions? Here is Eliot’s answer, in truncated form:
…It happens, on the contrary, that my strongest effort is to avoid any such arbitrary picture, and to give a faithful account of men and things as they have mirrored themselves in my mind. The mirror is doubtless defective; the outlines will sometimes be disturbed, the reflection faint or confused; but I feel as much bound to tell you as precisely as I can what that reflection is, as if I were in the witness box, narrating my experience on oath.
If this argument seems so self-evident as to not even be worth arguing, of course, it’s because this conception of realism—call it “holding a mirror up to nature”—has become so much a part of our aesthetic DNA that we no longer are aware of its source. Fiction, in short, is not about ideals, but about empirical truth: what can actually be observed. Eliot illustrates this turn in a sentence that travels the distance from Milton to Thomas Hardy in the flick of a phrase:
I turn, without shrinking, from cloud-borne angels, from prophets, sibyls, and heroic warriors, to to an old woman bending over her flowerpot, or eating her solitary dinner, while the noonday light, softened perhaps by a screen of leaves, falls on her mob-cap, and just touches the rim of her spinning wheel, and her stone jug, and all these cheap common things which are the precious necessaries of life to her…
“Cheap common things”: one hears echoes of this phrase in Flaubert’s letters, in Zola, in Chekhov, even to a degree (perhaps without the damning word “cheap”) in Henry James and Virginia Woolf, but more to the point, for our purposes, in every textbook on fiction, every introductory creative writing course in use today. Describe what you see. Or, as Joseph Conrad put it in his famous preface to The N— of the Narcissus: “My job is to make you see.” It’s no coincidence at all that realism in fiction arrived on the scene, in Anglo-American discourse, at the same moment as photography. Reality, what exists in the real world, is the subject of interest here; this is true subjectively (that is, the work is supposed to portray the workings of the mind as it actually happens) but also objectively (the work is supposed to focus on what actually exists in the world, however humble, rather than what we wish was there). Reality is, as Eliot and many other writers of her time (and later) insist, the “proof” on which fiction depends: “I feel as much bound to tell you as precisely as I can what that reflection is, as if I were in the witness box, narrating my experience on oath.”
There’s a logical problem with these analogies that becomes obvious as soon as we detach ourselves a little from Eliot’s argument. This interpretation of realism blurs the line between representation—mimesis, the creation of likenesses—and reproduction, that is, creating an image of something that already exists. It strongly implies that fiction should aspire to reproduction; that reproduction is the only dependable, “provable” art. The difficulty, of course, is that fictional characters and situations don’t already exist. There is no such thing, in fiction, as “reproduction,” in the sense that there is in photography, or in legal testimony. There is no original, no referent, no event, only the carefully crafted perception of such an original. As the English literary critic Terry Eagleton says, this makes the standard Eliot proposes—“narrating my experience on oath”—not only unattainable but a little unfair:
If realism is taken to mean ‘represents the world as it actually is’, then there is plenty of room for wrangling over what counts in this respect. You cannot decide whether a work is realist simply by inspecting it. Suppose we discovered a piece of writing from some long-vanished civilisation which we knew was in some sense fictional, and which paid inordinate attention to the length of men’s noses. We might categorise the work as non-realist, until further archaeological research revealed that the civilisation in question regarded nose-size as an important index of male fertility. In which case the text might shift into the category of realism. Literary critics in the distant future would not be able to tell that Endgame was non-realist unless, for example, they had historical evidence that putting old people in dustbins was not standard geriatric practice in the mid-20th century.
What Eagleton is proposing here is that we treat realism less as a de facto standard for evaluating all works of fiction and more as a question of intent within a certain cultural and historical moment. In other words, realism is only a subjective question: if a writer seeks out “what actually happened,” and produces something recognizably like the material fabric of existence in a certain era, then that is realism, whether it bears the label or not.
The problem, however, from a writer’s point of view, is that when we acknowledge that what we’re doing is representation and not reproduction, we may begin to doubt the accuracy, the persuasiveness, of our own perceptions. On the one hand we have an impossible epistemological demand: make it feel like it actually happened. On the other hand we have our own limited imaginative perception, our own tiny corner of worldly experience. How can we make others believe what we ourselves don’t know if we believe?
I’ve chosen George Eliot as the subject of this essay because I have the feeling that at the end of her writing career she began to struggle with this kind of self-doubt, not so much because she doubted her abilities but, on the contrary, because she was such a virtuoso that she could make her readers believe virtually anything. I believe this not on the basis of research into her letters or biography but because her last novel, Daniel Deronda, confronts the question head-on.
For Eliot the problem of the verifiability of fictional assertions had a profound moral dimension. She came to fiction relatively late in life; as a child and young adult she belonged to an evangelical Christian sect, and later broke with orthodox Christianity when she translated The Life of Jesus, one of the first texts to engage in skeptical historical interpretation of the Bible. (It was at the same time that she took up a twenty-year cohabitation with an older, already married man, which did nothing to endear her to her Christian brethren). Only in her forties, after two decades of translating and writing philosophy and theology, did she turn to writing novels. She brought to fiction not the sense of a religious vocation but rather a post-religious concern with translating Christian values and principles into a secular, historical world. Which is another way of saying that to her realism was a matter of moral and spiritual integrity, as well as aesthetic success. And that, I think, contributes to the uncertainty she demonstrates about the realist “project” in Daniel Deronda.
It’s not really necessary to know the whole plot of the novel to understand these examples, but here’s a rough outline: Gwendolen Harleth is the beautiful and prodigal daughter of a widow; she has been educated in the grand style, expecting that she will enter into high society one day, but at the beginning of the novel, she learns that her mother has lost all her money. Faced with the prospect of becoming a governess, she allows herself to be wooed by an immensely wealthy man, Grandcourt, even though she has grave doubts about his character. Grandcourt turns out to be a brutal, psychologically abusive husband. The novel’s other main character, Daniel Deronda, is the adopted son of Grandcourt’s uncle, a noble and generous young man whom Gwendolen falls in love with, but who ultimately rejects her. (Deronda also, in the course of the novel, learns that he is Jewish and ultimately leaves England for Palestine, but that doesn’t enter into the passages I will quote here).
The first passage I want to look at treats Gwendolen Harleth directly as a subject for debate. It occurs very early in the novel, when Gwendolen has just returned home to her newly impoverished family.
Always she was the princess in exile, who in time of famine was to have her breakfast-roll made of the finest bolted flower from the seven thin ears of wheat, and in a general decampment was to have her silver fork kept out of the baggage. How was this to be accounted for? The answer may seen to lie quite on the surface—in her beauty, in a certain unusualness about her, a decision of will which made itself felt in her gracious movements and clear unhesitating tones…This potent charm, added to the fact that she was the eldest daughter…may seem so full a reason for Gwendolen’s domestic empire, that to look for any other would be to ask the reason of daylight when the sun is shining. But beware of arriving at conclusions without comparison. I remember having seen the same assiduous, apologetic attention awarded to persons who were not at all beautiful or unusual. Some of them were a very common sort of men. And the only point of resemblance among them all was a strong determination to have what was pleasant, with a total fearlessness in making themselves disagreeable or dangerous when they did not get it. I doubt whether even without her potent charm and peculiar filial position Gwendolen might not still have played the queen in exile, if only she had kept her inborn energy of egoistic desire, and her power of inspiring fear as to what she might say or do. However, she had the charm, and those who feared her were also fond of her; the fear and the fondness being perhaps both heightened by what may be called the iridescence of her character—the play of various, nay, contrary tendencies.
Since very few contemporary writers use this kind of explicit authorial commentary at length, to our ears this passage might seem most similar to the voice-over we hear on certain TV shows: Sex and the City, Desperate Housewives, or Gossip Girl. And up to a certain point this is not an inaccurate comparison. George Eliot’s tone here begins, like one of those omniscient female narrators, as snide condescension mixed with envy, luxuriating in the narrator’s power not only to control but to evaluate and ridicule. And this is why the line in the middle of the passage so powerfully brings us up short: “But beware of arriving at conclusions without comparison.”
What kind of thing is that for an omniscient narrator to say? Rather than dangling Gwendolen in front of us like a puppet, Eliot invokes, in a subtle but troubling way, the question of verifiability: is it enough to say that Gwendolen, like the wicked stepsister, gets her way because she’s beautiful and gracious, and the eldest in the family? No, it is not. Gwendolen’s power lies in her “egoistic desire,” and we know this because we can compare her with other similar cases, a kind of control group. (Where these other cases come from is not specified, and this is only one among several ways that Eliot plays with the narrator’s ambiguous status). Moreover, Gwendolen is not, like the wicked stepsister, secretly detested by those around her; “those who feared her were also fond of her; the fear and the fondness being perhaps both heightened by what may be called the iridescence of her character.” In this passage Eliot proposes an empirical problem—how do we know we know how Gwendolen gets her power over the family?—and, by drawing some unspecified, but convincing, bit of observable wisdom, seems to solve it.
The second, and more challenging, passage I want to look at in Daniel Deronda occurs later, when we are introduced to Gwendolen’s future husband, Grandcourt, for the first time. This is the first part.
He was slightly taller than herself, and their eyes seemed to be on a level; there was not the faintest smile on his face as he looked at her, not a trace of self-consciousness or anxiety in his bearing; when he raised his hat he showed an extensive baldness surrounded with a mere fringe of reddish-blond hair, but he also showed a perfect hand; the line of feature from brow to chin undisguised by beard was decidedly handsome, with only moderate departures from the perpendicular. It was not possible for a human aspect to be freer from grimace or solicitous wrigglings; also it was perhaps not possible for a breathing man wide awake to look less animated. The correct Englishman, drawing himself up from his bow into rigidity, assenting severely, and seeming to be in a state of internal drill, suggests a suppressed vivacity, and may be suspected of letting go with some violence when he is released from parade; but Grandcourt’s bearing had no rigidity, it inclined rather to the flaccid.
This is Eliot working at the height of her descriptive powers. Looking at the rhythm of the prose alone for a minute, consider the way she deploys semicolons to mark the movement of the first sentence, each statement rising, cresting, and falling like a wave. Grandcourt is not beautiful the way an Apollonian statue is beautiful; instead he expresses that highest Victorian aspiration, order and proportion: “he showed an extensive baldness surrounded with a mere fringe of reddish-blond hair, but he also showed a perfect hand.” Already extremely wealthy, born into high society, used to getting his way, he projects a sense of wanting nothing and being impressed by nobody—which is just the beginning of his power over Gwendolen.
On one level, we can easily say that Grandcourt feels like a person we might meet (were we to live in England in the late nineteenth century): Eliot describes him in such a richly nuanced way that it’s difficult not to feel that he is a flesh-and-blood person sprung to life. At the same time, the prose is so carefully structured—sculpted—that it is impossible to ignore that Grandcourt is a verbal creation. Again, Eliot runs up against the problem of artifice, of unverifiability. And here she issues a much more strident challenge to herself, and to the reader:
Attempts at description are stupid: who can all at once describe a human being? even when he is presented to us we only begin that knowledge of his appearance which must be completed by innumberable impressions under differing circumstances. We recognize the alphabet; we are not sure of the language. I am only mentioning the points that Gwendolen saw by light of a prepared contrast in the first five minutes of her meeting with Grandcourt: they were summed up in the words, “He is not ridiculous.”
Perhaps the most important phrase in this passage is “by light of a prepared contrast.” Here Eliot is explicitly poking a hole in the analogy of realist fiction to photography, or photo-reproduction. Gwendolen’s introduction to Grandcourt is not “natural” or simply “mirrored in her mind”; it’s carefully staged, backlit, somewhat melodramatically engineered into the narrative. But there’s an even more disconcerting ambivalence in Eliot’s preceding statement: “Even when he is presented to us we only begin that knowledge of his appearance which must be completed by innumberable impressions under differing circumstances.” Innumberable impressions, that is, more impressions than are ever possible in the text of a novel, even one as long and comprehensive as this. In the midst of the most suspenseful moment in the novel thus far, Eliot simultaneously presents and washes her hands of Grandcourt. You could call it virtuosic ambivalence, or, as I like to think of it, a kind of meta-realism.
In his 1979 book Philosophy and the Mirror of Nature, Richard Rorty proposed that philosophers abandon the traditional notion that what we think reflects reality in a determinable, systematic way. Drawing on a century of “anti-foundational” philosophy—that is, philosophy that argues against a stable, universal, unchangeable notion of truth—he argued that instead of constantly trying to establish a common basis of agreement we should pay attention to how reality is understood within certain local, specific conditions, without privileging one set of conditions over the other.
I invoke Rorty here because although it would seem that fiction is by its very nature anti-foundational—after all, as I said earlier, it has no original, no referent—within a realistic narrative our tendency as writers and readers is to cling to certain foundational assumptions with the tenacity of a ten-year-old who insists on explaining the difference between High Elvish and Low Elvish while watching The Lord Of The Rings. No matter how tenuously a character may be described or referred to in a fictional text, our minds have a tendency to grant that person a full, though not necessarily complex, existence. This generalizing capacity is exactly what George Eliot is referring to when she asks, in the passage I quoted above, “Who can all at once describe a human being?” We tend to take it on trust that any fiction writer can; if they’re not fully capable of describing a human being, of vivifying inanimate words through a kind of Frankensteinian juju, what business do they have writing fiction at all?
All the same, what George Eliot is doing in these passages in Daniel Deronda is not metafiction, in the sense of Borges, Robert Coover, or John Barth, or Tristram Shandy, for that matter. She is not interested in disrupting what John Gardner called the “vivid, continuous dream” of the realist novel; she is not destablizing narrative itself at all, but rather taking it for granted that the reader wants to believe in the existence of this fictional world as much as she does. What she is doing is something more Rorty-esque, by asserting, through the narrator, her provisional, not total, authority over the story. When Philosophy and the Mirror of Nature was published, it occasioned howls of protest among philosophers and cultural commentators, from both the right and the left, who accused Rorty of undermining the basis for law, morality, and science, by saying that no one set of principles could ever be declared absolutely or innately true. Rorty’s response, which infuriated his critics, was that everyone should basically relax and keep doing what they were doing before, as long as they acknowledged that “’objective truth’ is no more and no less than the best idea we currently have about how to explain what is going on.”
Part of what makes Eliot’s approach especially interesting is that Daniel Deronda stands at the end of a long period in which the implied author/narrator held sway in the English novel. The generation after her—Thomas Hardy, George Gissing—was much more reticent in its use of authorial commentary. By the turn of the twentieth century, the author had begun to disappear into the modernist convention of authorial silence, which has come down to us in the dictum of “show, don’t tell.” (For more on this turn toward authorial silence in the modernist era, check out Wayne Booth’s classic The Rhetoric of Fiction.) Of course, in its own way, authorial silence is just as absolute in its claim of epistemological authority as the Victorian use of authorial commentary: either way we are expected to take the “existence” of the fictional universe as a given. George Eliot proposes, in these passages, something that is more like a conversation between the narrator and the reader, and this, again, is why Rorty is a particularly interesting parallel, because what he advocates in place of absolutism is a kind of philosophic conversation:
…Coming to understand is more like getting acquainted with a person than following a demonstration. In both cases we play back and forth between guesses about how to characterize particular statements or other events, and guesses about the point of the whole situation, until gradually we feel at ease with that was hitherto strange.
This is not at all far removed from what Eliot said in her little disquisition on Grandcourt: “We recognize the alphabet; we are not sure of the language.” In both cases we’re more focused on an ongoing process than any kind of absolute result. But we’re not focused on the process as a means of undoing the process, as is, say, Alain Robbe-Grillet’s novel Jealousy, which slows down a sequence of events almost to the point of stasis in a deliberate effort to frustrate the reader’s expectations of a temporal sequence. Meta-realist writing, as I like to call what Eliot is doing here, is not experimental as much as interrogative.
I want to sharpen my definition of meta-realism by looking at a contemporary example that couldn’t be more different from Daniel Deronda. This is James Alan McPherson’s short story “Elbow Room,” published in the collection of the same name, which won the Pulitzer Prize in 1978. “Elbow Room” is, as its center, a story about an interracial marriage in San Francisco in the early 1970s, but it is also, simultaneously, a debate about how—and whether—it is possible to tell the story of an interracial marriage in San Francisco in the early 1970s. Echoing the paranoid style of the times, McPherson stages this as a debate between the narrator (here named “Narrator,” with a capital N) and “Editor,” with a capital E, who begins the story by announcing,
Narrator is unmanageable. Demonstrates a disregard for form…questioned closely, he declares himself the open enemy of conventional narrative categories…In order to save this narration, editor felt compelled to clarify slightly, not to censor but to impose at least the illusion of order. This was an effort toward preserving a certain morality of technique.
Editor’s effort, however, is partly unsuccessful; the narration keeps slipping away from his “standards,” and between passages McPherson inserts an ongoing debate in italics between the two:
The above section is totally unclear. It should be cut.
I would leave it in. It was attempting to suggest the nature of the times.
But here the narrative begins to drift. There is a shift in subject, mood, and focus of narration. Cutting is advised.
Back during that time there was little feeling and no focus.
What is it about the story that is so unacceptable to Editor? First, the narrator’s insistence on blurring the distinction between himself and the story he sets out to tell; second, the narrator’s insistence that there is no way of telling the story of this couple without treating them as actors in, and victims of, the circumstances of the times. Consider this passage in which McPherson describes the black woman in the couple, Virginia, who joined the Peace Corps to escape an impoverished background, and returned with a highly cosmopolitan sensibility:
In Boston, New York, Philadelphia, Chicago, people gathered in groups and told similar stories. They thought in terms new to them. In conversation they remarked on common points of reference in the four quarters of the world. The peasants among them had become aristocratic without any of the telling affectations. The aristocrats by birth had developed an easy, common touch. They considered themselves a new tribe.
But then their minds began to shift. In the beginning it was a subtle process. During conversation someone might say a casual “you know?” and there would be a hesitation at first, denying affirmation…People began to feel self-conscious and guilty…Inevitably, many people in conversation began saying, “I don’t understand!” It took several months before they became black and white.
This is a beautifully compressed and heartrending description of the way racial discord evolved among educated, liberal Americans in the Sixties, but it is also, from the Editor’s point of view, a “violation of conventional narrative categories.” What is social history of this kind doing in a short story about two individuals? Shouldn’t we be paying attention to the lived experience of their relationship, and not what’s happening in Boston, New York, Philadelphia, Chicago? Even more galling to the Editor is the narrator’s insistence that this state of affairs had seeped into literature itself:
The life-affirming peasants of Chekhov and Babel sat wasted and listless on their porches, oblivious to the beats in their own blood. Even Pushkin’s firebrands seemed content with the lackluster: mugging old ladies, killing themselves, snatching small change from dollar-and-dime grocers…Men with greatness in them spoke on the telephone, and in private, as if bouncing safe clichés off the ear of a listener into an expectant and proprietary tape recorder.
And the caste curtains were drawn, resegregating all imaginations.
What McPherson is doing here, above all, is working at the far edge of the fiction writer’s ability to generalize, to place characters in a larger social and cultural fabric while still insisting on their fictional autonomy. Rather than push his characters to be exemplars of a particular social tendency—to make them exaggerations, objects of satire—he wants to reserve the space for them to act as individuals without insisting on their absolute individuality. Referring to the Kansas-born Paul Frost, Virginia’s white husband to be, he writes,
I know that when I looked I saw dead Indians living in his eyes. But I also say a wholesome glow in their directness. They seemed in earnest need of answers to honest questions always on the verge of being asked.
Consider how close this is, in a sense, to Eliot’s presentation of Grandcourt:
The correct Englishman, drawing himself up from his bow into rigidity, assenting severely, and seeming to be in a state of internal drill, suggests a suppressed vivacity, and may be suspected of letting go with some violence when he is released from parade; but Grandcourt’s bearing had no rigidity, it inclined rather to the flaccid.
Categorical statement: then individual exception. It’s the tension between these two things—between what we know we know and what we’re willing to concede, for the purposes of getting on with the story—that make meta-realism different from other kinds of experimental fiction that more often want to deny the reader the mimetic experience of the story at all. As Richard Rorty never stopped arguing, the turn away from absolutist forms of epistemology never spelled the end of philosophy, and, in a similar way, the meta-realist practice of upending narrative without destroying makes it impossible to argue, as many cynics do, for the end of the novel as a form of intelligent art.
Image credit: Portrait of Mary Ann Evans (George Eliot), pencil drawing by Samuel Lawrence (Wikimedia commons)