In describing novels from the first half of the 19th Century, David Shields writes in Reality Hunger: A Manifesto that “All the technical elements of narrative—the systematic use of the past tense and the third person, the unconditional adoption of chronological development, linear plots, the regular trajectory of the passions, the impulse of each episode toward a conclusion, etc.—tended to impose the image of a stable, coherent, continuous, unequivocal, entirely decipherable universe.”

I’m not so sure; the more interesting novels didn’t necessarily have “the unconditional adoption of chronological development” or the other features Shields ascribes to them. Caleb Williams is the most obvious example I can immediately cite: the murderers aren’t really punished in it and madness is perpetual. Gothic fiction of the 19th Century had a highly subversive quality that didn’t feature “the regular trajectory of the passions.” To my mind, the novel has always had unsettling features and an unsettling effect on society, producing change even when that change isn’t immediately measurable or apparent, or when we can’t get away from the fundamental constraints of first- or third-person narration. Maybe I should develop this thought more: but Shields doesn’t in Reality Hunger, so maybe innuendo ought to be enough for me too.

Shields is very good at making provocative arguments and less good at making those arguments hold up under scrutiny. He says, “The creators of characters, in the traditional sense, no longer manage to offer us anything more than puppets in which they themselves have ceased to believe.” Really? I believe if the author is good enough. And I construct coherence where it sometimes appears to be lacking. Although I’m aware that I can’t shake hands with David Kepesh of The Professor of Desire, he and the characters around him feel like “more than puppets” in which Roth has ceased to believe.

Shields wants something made new. Don’t we all? Don’t we all want to throw off dead convention? Alas: few of us know how to successfully, and that word “successfully” is especially important. You could write a novel that systematically eschews whatever system you think the novel imposes (this is the basic idea behind the anti-novel), but most people probably won’t like it—a point that I’ll come back to. We won’t like it because it won’t seem real. Most of us have ideas about reality that are informed by some combination of lived experience and cultural conditioning. That culture shifts over time. Shields starts Reality Hunger with a premise that is probably less contentious than much of the rest of the manifesto: “Every artistic movement from the beginning of time is an attempt to figure out a way to smuggle more of what the artist thinks is reality into the work of art.” I can believe this, though I suspect that artists begin getting antsy when you try to pin them down on what reality is: I would call it this thing we all appear to live in but that no one can quite represent adequately.

That includes Shields. Reality Hunger doesn’t feel as new as it should; it feels more like a list of N things. It’s frustrating even when it makes one think. Shields says, “Culture and commercial languages invade us 24/7.” But “commercial languages” only invade us because we let them: TV seems like the main purveyor, and if we turn it off, we’ll probably cut most of the advertising from our lives. If “commercial languages” are invading my life to the extent I’d choose the word “invade,” I’m not aware of it, partially because I conspicuously avoid those languages. Shields says, “I try not to watch reality TV, but it happens anyway.” This is remarkable: I’ve never met anyone who’s tried not to watch reality TV and then been forced to, or had reality TV happen to them, like a car accident or freak weather.

Still, we need to think about how we experience the world and depict it, since that helps us make sense of the world. For me, the novel is the genre that does this best, especially when it bursts its perceived bounds in particularly productive ways. I can’t define those ways with any rigor, but the novel has far more going on than its worst and best critics imagine.

Both the worst and best critics tend to float around the concept of reality. To use Luc Sante’s description in “The Fiction of Memory,” a review of Reality Hunger:

The novel, for all the exertions of modernism, is by now as formalized and ritualized as a crop ceremony. It no longer reflects actual reality. The essay, on the other hand, is fluid. It is a container made of prose into which you can pour anything. The essay assumes the first person; the novel shies from it, insisting that personal experience be modestly draped.

I’m not sure what a “crop ceremony” is or how the novel is supposed to reflect “actual reality.” Did it ever? What is this thing called reality that the novel is attempting to mirror? Its authenticity or lack thereof has, as far as I know, always been in question. The search for realism is always a search and never a destination, even when we feel that some works are more realistic than others.

Yet Sante and Sheilds are right about the dangers of rigidity; as Andrew Potter writes in The Authenticity Hoax: How We Get Lost Finding Ourselves, “One effect of disenchantment is that pre-existing social relations come to be recognized not as being ordained by the structure of the cosmos, but as human constructs – the product of historical contingencies, evolved power relations, and raw injustices and discriminations.”

Despite this, however, we feel realism—if none of us did, we’d probably stop using the term. Our definitions might blur when we approach a precise definition, but that doesn’t mean something isn’t there.

Sante writes, quoting Shields, that “‘Anything processed by memory is fiction,’ as is any memory shaped into literature.” Maybe: but consider these three statements, if I were to make them to you (keep in mind the context of Reality Hunger, with comments like “Try to make it real—compared to what?”):

Aliens destroyed Seattle in 2004.

I attended Clark University.

Alice said she was sad.

One of them is, to most of us, undoubtedly fiction. One of them is true. The other I made up: no doubt there is an Alice somewhere who has said she is sad, but I don’t know her and made her up for the purposes of example. The second example might be “process by memory,” but I don’t think that makes it fiction, even if I can’t give you a firm, rigorous, absolute definition of where the gap between fact and interpretation begins. Jean Bricmont and Alan Sokal give it a shot in Fashionable Nonsense: “For us, as for most people, a ‘fact’ is a situation in the external world that exists irrespective of the knowledge that we have (or don’t have) of it—in particular, irrespective of any consensus or interpretation.”

They go to observe that scientists actually face some problems of definition that I see as similar to those of literature and realism:

Our answer [as to what makes science] is nuanced. First of all, there are some general (but basically negative) epistemological principles, which go back at least to the seventeenth century: to be skeptical of a priori arguments, revelation, sacred texts, and arguments from authority. Moreover, the experience accumulated during three centuries of scientific practice has given us a series of more-or-less general methodological principles—for example, to replicate experiments, to use controls, to test medicines in double-blind protocols—that can be justified by rational arguments. However, we do not claim that these principles can be codified in a definite way, nor that the list is exhaustive. In other words, there does not exist (at least present) a complete codification rationality, is always an adaptation to a new situation.

They lay out some criteria (beware of “revelation, sacred texts, and arguments from authority”) and “methodological principles” (“replicate experiments”) and then say “we do not claim that these principles can be codified in a definite way.” Neither can the principles of realism. James Wood does as good a job of exploring them as anyone. But I would posit that, despite our inability to pin down realism, either as convention or not, most of us recognize it: when I tell people that I attended Clark University, none have told me that my experience is an artifact of memory, or made up, or that there is no such thing as reality and therefore I didn’t. Such realism might merely be convention or training—or it might be real.

In the first paragraph of his review of Chang-Rae Lee’s The Surrendered, James Wood lays out the parameters of the essential question of literary development or evolution:

Does literature progress, like medicine or engineering? Nabokov seems to have thought so, and pointed out that Tolstoy, unlike Homer, was able to describe childbirth in convincing detail. Yet you could argue the opposite view; after all, no novelist strikes the modern reader as more Homeric than Tolstoy. And Homer does mention Hector’s wife getting a hot bath ready for her husband after a long day of war, and even Achilles, as a baby, spitting up on Phoenix’s shirt. Perhaps it is as absurd to talk about progress in literature as it is to talk about progress in electricity—both are natural resources awaiting different forms of activation. The novel is peculiar in this respect, because while anyone painting today exactly like Courbet, or composing music exactly like Brahms, would be accounted a fraud or a forger, much contemporary fiction borrows the codes and conventions—the basic narrative grammar—of Flaubert or Balzac without essential alteration.

I don’t think literature progresses “like medicine or engineering.” Using medical or engineering knowledge as it stood in 1900 would be extremely unwise if you’re trying to understand the genetic basis of disease or build a computer chip. Papers tend to decay within five to ten years of publication in the sciences.

But I do think literature progresses in some other, less obvious way, as we develop wider ranges of techniques and social constraints allow for wider ranges of subject matter or direct depiction: hence why Nabakov can point out that “Tolstoy, unlike Homer, was able to describe childbirth in convincing detail,” and I can point out that mainstream literature effectively couldn’t depict explicit sexuality until the 20th Century.

While that last statement can be qualified some, it is hard to miss the difference between a group of 19th Century writers like Thackeray, Dickens, Trollope, George Eliot, George Meredith, and Thomas Hardy (who J. Hillis Miller discusses in The Form of Victorian Fiction) and a group of 20th Century writers like D.H. Lawrence, James Joyce, Norman Rush, and A.S. Byatt, who are free to explicitly describe sexual relationships to the extent they see fit and famously use words like “cunt” that simply couldn’t be effectively used in the 19th Century.

In some ways I see literature as closer to math: the quadratic equation doesn’t change with time, but I wouldn’t want to be stuck in a world with only the quadratic equation. Wood gets close to this when he says that “Perhaps it is as absurd to talk about progress in literature as it is to talk about progress in electricity—both are natural resources awaiting different forms of activation.” The word “perhaps” is essential in this sentence: it gives a sense of possibility and realization that we can’t effectively answer the question, however much we might like to. But both question and answer give a sense of some useful parameters for the discussion. Most likely, literature isn’t exactly like anything else, and its development (or not) is a matter as much of the person doing the perceiving and ordering as anything intrinsic to the medium.

I have one more possible quibble with Wood’s description when he says that “the basic narrative grammar—of Flaubert or Balzac without essential alteration.” I wonder if it really hasn’t undergone “essential alteration,” and what would qualify as essential. Novelists like Elmore Leonard, George Higgins, or that Wood favorite Henry Green all feel quite different from Flaubert or Balzac because of how they use dialog to convey ideas. The characters in Tom Perrotta’s Election speak in a much more slangy, informal style than do any in Flaubert or Balzac, so far as I know. Bellow feels more erratic than the 19th Century writers and closer to the psyche, although that might be an artifact of how I’ve been trained by Bellow and writers after Bellow to perceive the novel and the idea of psychological realism. Taken together, however, the writers mentioned make me think that maybe “the basic narrative grammar” has changed for writers who want to adopt new styles. Yes, we’re still stuck with first- and third-person perspectives, but we get books that are heavier on dialog and lighter on formality than their predecessors.

Wood is a great chronicler of what it means to be real: his interrogation of this seemingly simple term runs through the essays collected in The Irresponsible Self: On Laughter and the Novel, The Broken Estate: Essays on Literature and Belief, and, most comprehensively, in the book How Fiction Works. Taken together, they ask how the “basic narrative grammar” of fiction works or has worked up to this point. In setting out some of the guidelines that allow literary fiction to work, Wood is asking novelists to find ways to break those guides in useful and interesting ways. In discussing Reality Hunger, Wood says, “[Shields’] complaints about the tediousness and terminality of current fictional convention are well-taken: it is always a good time to shred formulas.” I agree and doubt many would disagree, but the question is not merely one of “shred[ing] formulas,” but how and why those formulas should be shred. One doesn’t shred the quadratic formula: it works. But one might build on it.

By the same token, we may have this “basic narrative grammar” not because novelists are conformist slackers who don’t care about finding a new way forward: we may have it because it’s the most satisfying or useful way of conveying a story. Although I don’t think this is true, I think it might be true. Maybe most people won’t find major changes to the way we tell stories palatable. Despite modernism and postmodernism, fewer people appear to enjoy the narrative confusion and choppiness of Joyce than do enjoy the streamlined feel of the latest thriller. That doesn’t mean the latter is better than the former—by my values, it’s not—but it does mean that the overall thrust of fiction might remain where it is.

Robert McKee, in his not-very-good-but-useful book Story: Substance, Structure, Style and The Principles of Screenwriting, gives three major kinds of plots, which blend into one another: “arch plots” that are causal in nature and finish their story lines; “mini plots,” which he says are open and “strive for simplicity and economy while retaining enough of the classical […] to satisfy the audience,” and antiplot, which are where absurdism and the like fall.

He says that as one moves “toward the far reaches of Miniplot, Antiplot, and Non-plot, the audience shrinks” (emphasis in original). From there:

The atrophy has nothing to do with quality or lack of it. All three corners of the story triangle gleam with masterworks that the world treasures, pieces of perfection for our imperfect world. Rather, the audience shrinks for this reason: Most human beings believe that life brings closed experiences of absolute, irreversible change; that their greatest sources of conflict are external to themselves; that they are the single and active protagonists of their own existence; that their existence operates through continuous time within a consistent, causally interconnected reality; and that inside this reality events happen for explainable and meaningful reasons.

The connection between this and Wood’s “basic narrative grammar” might appear tenuous, but McKee and Wood are both pointing towards the ways stories are constructed. Wood is more concerned with language; although plot and its expression (whether in language or in video) can’t be separated from one another, they can still be analyzed independently enough of one another to make a distinction.

The conventions that underlie the “arch plots,” however, can become tedious over time. This is what Wood is highlighting when he discusses Roland Barthes’ “reality effect,” which fiction can achieve: “All this silly machinery of plotting and pacing, this corsetry of chapters and paragraphs, this doxology of dialogue and characterization! Who does not want to explode it, do something truly new, and rouse the implication slumbering in the word ‘novel’?” Yet we need some kind of form to contain story; what is that form? Is there an ideal method of conveying story? If so, what if we’ve found it and are now mostly tinkering, rather than creating radical new forms? If we take out “this silly machinery of plotting and pacing” and dialog, we’re left with something closer to philosophy than to a novel.

Alternately, maybe we need the filler and coordination that so many novels consist of if those novels are to be felt true to life, which appears to be one definition of what people mean by “realistic.” This is where Wood parts with Barthes, or at least makes a distinct case:

Convention may be boring, but it is not untrue simply because it is conventional. People do lie on their beds and think with shame about all that has happened during the day (at least, I do), or order a beer and a sandwich and open their computers; they walk in and out of rooms, they talk to other people (and sometimes, indeed, feel themselves to be talking inside quotation marks); and their lives do possess more or less traditional elements of plotting and pacing, of suspense and revelation and epiphany. Probably there are more coincidences in real life than in fiction. To say “I love you” is to say something at millionth hand, but it is not, then, necessarily to lie.

“Convention may be boring, but it is not untrue simply because it is conventional,” and the parts we think of as conventional might be necessary to realism. In Umberto Eco’s Reflections on The Name of the Rose, he says that “The postmodern reply to the modern consists of recognizing that the past, since it cannot really be destroyed, because its destruction leads to silence, must be revisited: but with irony, not innocently.” That is often the job of novelists dealing with the historical weight of the past and with conventions that are “not untrue simply because [they are] conventional.” Eco and Wood both use the example of love to demonstrate similar points. Wood’s is above; Eco says:

I think of the postmodern attitude as that of a man who loves a very cultivated woman and knows he cannot say to her, ‘I love you madly,’ because he knows that she knows (and that she knows that he knows) that these words have already been written by Barbara Cartland. Still, there is a solution. He can say, ‘As Barbara Cartland would put it, I love you madly.’ At this point, having avoided false innocence, having said clearly that it is no longer possible to speak innocently, he will nevertheless have said what he wanted to say to the woman: that he loves her, but he loves her in an age of lost innocence. If the woman goes along with this, she will have received a declaration of love all the same. Neither of the two speakers will feel innocent, both will have accepted the challenge of the past, of the already said, which cannot be eliminated […]

I wonder if every age thinks of itself as “an age of lost innocence,” only to be later looked on as pure, naive, or unsophisticated. Regardless, for Eco postmodernism requires that we look to the past long enough to wink and then move on with the story we’re going to tell in the manner we’re going to tell it. Perhaps Chang-Rae Lee doesn’t do so in The Surrendered, which is the topic of Wood’s essay—but like so many essays and reviews, Wood’s starts with a long and very useful consideration before coming to the putative topic of its discussion. Wood speaks of reading […] “Chang-Rae Lee’s new novel, “The Surrendered” (Riverhead; $26.95)—a book that is commendably ambitious, extremely well written, powerfully moving in places, and, alas, utterly conventional. Here the machinery of traditional, mainstream storytelling threshes efficiently.” I haven’t read The Surrendered and so can’t evaluate Wood’s assessment.

Has Wood merely overdosed on the kind of convention that Lee uses, as opposed to convention itself? If so, it’s not clear how that “machinery” could be fixed or improved on, and the image itself is telling because Wood begins his essay by asking whether literature is like technology. My taste in literature changes: as a teenager I loved Frank Herbert’s Dune and now find it almost unbearably tedious. Other revisited novels hold up poorly because I’ve overdosed on their conventions and start to crave something new—a lot of fantasy flattens over time like opened soda.

Still, I usually don’t know what “something new” entails until I read it. That’s the problem with saying that the old way is conventional or boring: that much is easier to observe than the fix. Wood knows it, and he’s unusually good at pointing to the problems of where we’ve been and pointing to places that we might go to fix it (see, for example, his recent essay on David Mitchell, who I now feel obliged to read). This, I suspect, is why he is so beloved by so many novelists, and why I spend so much time reading him, even when I don’t necessarily love what he loves. The Quickening Maze struck me as self-indulgent and lacking in urgency, despite the psychological insight Adam Foulds offers into a range of characters’ minds: a teenage girl, a madman, an unsuccessful inventor.

I wanted more plot. In How Fiction Works, Wood quotes from Adam Smith writing in the eighteenth century regarding how writers use suspense to maintain reader interest and then says that “[…] the novel [as an art form; one could also say the capital-N Novel] soon showed itself willing to surrender the essential juvenility of plot […]” Yet I want and crave this element that Wood dismisses—perhaps because of my (relatively) young age: Wood says that Chang-Rae Lee’s Native Speaker was “published when the author was just twenty-nine,” older than I am. I like suspense and the sense of something major at stake, and that could imply that I have a weakness for weak fiction. If so, I can do little more than someone who wants chocolate over vanilla, or someone who wants chocolate despite having heard the virtues of cherries extolled.

When I hear about the versions of the real, reality, and realism that get extolled, I often begin to think about chocolate, vanilla, and cherries, and why some novelists write in such a way that I can almost taste the cocoa while others are merely cardboard colored brown. Wood is very good at explaining this, and his work taken together represents some of the best answers to the questions that we have.

Even the best answers lead us toward more questions that are likely to be answered best by artists in a work of art that makes us say, “I’ve never seen it that way before,” or, better still, “I’ve never seen it.” Suddenly we do see, and we run off to describe to our friends what we’ve seen, and they look at us and say, “I don’t get it,” and we say, “maybe you just had to see it for yourself.” Then we pass them the book or the photo or the movie and wait for them to say, “I’ve already seen this somewhere before,” while we argue that they haven’t, and neither have we. But we press on, reading, watching, thinking, hoping to come across the thing we haven’t seen before so we can share it again with our friends, who will say, like the critics do, “I’ve seen it before.”

So we have. And we’ll see it again. But I still like the sights—and the search.