John Barth’s The Sot-Weed Factor (1960, eighteenth-century poem of same name)SO WHO’S THE FIRST postmodern genius? You’re thinking Beckett, maybe, or Alain Robbe-Grillet? Think what you want, but I’m going with Chuck Jones. Yup. That one. Animator extraordinaire for the Brothers Warner. In one famous episode, “Duck Amuck,” Daffy Duck undergoes a series of baffling and ridiculous transformations, acquiring a tutu, a head like a flower, the body of a lion, and so on. At the end, we’re amused but hardly surprised to learn that the impish wielder of the cartoonist’s pencil is, indeed, Bugs Bunny. Warner Bros, took this self-reflexivity even further in the television series Animaniacs, which plays like an extended dialogue with the history of cartooning. In fact, cartoons seem uniquely well suited to this whole self-referential thing, whether made for television (The Simpsons, Family Guy, South Park) or feature films (Who Framed Roger Rabbit?). But even live-action shows get in on the game. The television series Moonlighting (admittedly not representative of anything in Hollywood) ended one season in midchase to collect the costumes and props because the season had ended before the episode had. Somehow, though, when serious writers do it, some readers may suspect they’ve been cheated. Yet cheated is the last thing we are; rather than closing down possibilities, the strategies we’ve come to know and love as metafiction opens them up.
It’s just that readers, and student readers especially, get a little impatient with funny business in their reading.
How impatient? Read on.
Sometimes we speak of a novel soaring, of narrative flights. Sounds nice, doesn’t it? Impressive. One envisions the passion of Morrison, the wit of Wodehouse, the scope and scale of Tolkien, the invention of García Márquez. No question about it, there are some wonderful novels in the world. Some of them seem to have wings. But that’s not what I mean. I’ll give you a soaring book: The French Lieutenant’s Woman.
The most thrown book in American colleges.
I know when, and when will get you why. What, you want the exact paragraph? We have two options, so take your pick. Sometimes it flies in Chapter 13, when the narrator, having just asked at the end of Chapter 12 who Sarah is and “Out of what shadows does she come,” answers with this little gem: “I do not know.” Not the answer we typically expect. Here’s the rule: when a narrator asks a question, he is obliged to answer. Fowles’s narrator does answer, only not to our satisfaction. But wait, it gets worse. Not satisfied confessing ignorance, he actively breaks the illusion he has maintained for a dozen chapters:
This story I am telling is all imagination. These characters I create never existed outside my own mind. If I have pretended until now to know my characters’ minds and innermost thoughts, it is because I am writing in (just as I have assumed some of the vocabulary and “voice” of) a convention universally accepted at the time of my story: that the novelist stands next to God. He may not know all, yet he tries to pretend that he does. But I live in the age of Alain Robbe-Grillet and Roland Barthes; if this is a novel, it cannot be a novel in the modern sense of the word.
At which point, wham! Against the wall, shaking the Sports Illustrated swimsuit calendar and rattling the South Padre Island souvenir beer mug nearly off the shelf.
The second place is perhaps less definite, but here’s where it begins: “No. It is as I say. You have not only planted the dagger in my breast, you have delighted in twisting it.” Nothing in there, in itself, to create flying paperbacks. Except that we’ve read it before. Exactly. And quite recently. This particular passage comes early in a final chapter, which had, a moment before, seemed superfluous. Chapter 60 ends with the finality of a novel’s ending, complete with a small child, a forever-after embrace, and (I’m not making this up) a thousand violins. So who needs another chapter? Well, Fowles does, for one. Chapter 61 offers us a different ending—no evermore, no string section. He uses the device of an author-surrogate setting his watch back fifteen minutes as the pretext for creating an alternative ending, one considerably less happy, if perhaps more in tune with modern sensibilities. And when some students find out that THE conclusion is merely A conclusion, and there is also a B, then Air Fowles flies again. To be sure, some readers see the extra chapter—and the extra alternative—as a gift.
There’s something maddening, at least for those reading under a deadline, about the revelation that novels are illusions. At some level we know, don’t we? Novels are made-up things? But they’re more than that. Novels are in a sense learned activities—learned by cultures and learned by individual readers and writers. A novel is, as Fowles suggests, a compilation of conventions, a series of if-then propositions: if the main character is a child at the beginning, then the goal of the novel is to deposit him at some state of adulthood by the end; if the narrator is limited to only knowing one character’s mind in Chapters 1 through 5, she can’t suddenly pretend to know everyone’s mind when the crunch comes in Chapter 23. Or this: if a narrator pretends to omniscience throughout the first twelve chapters and then poses a major question about a character, he damned well better have an answer when we turn the page. None of this suddenly becoming helpless as a babe; there is no off position on the omniscient switch.
Unless it’s a game. Unless you’re merely employing conventions to strip them of their illusion, in which case you had better be prepared for the backlash. This is what drives student-readers crazy: What if he’s just playing around? What if he’s not serious? It seems unfair somehow to the apprentice reader who is working very hard (usually to find things the professor has promised are in there—magical things, mysterious things) that the novelist might have regarded this not work but play. Not only that, but it seems to cause doubt about the whole fictional enterprise: if we can’t trust Fowles, and he’s using the devices and practices of the conventional novel, how can we trust any of them?
Well, we can’t. But is that a problem, really? You know a novel is made up when you open it. Does the illusion have to be “real”? Does that question even make sense? It does and it doesn’t. We’ve just entered the literary equivalent of quantum physics, where propositions can be simultaneously true and false, where waves and particles overlap and matter and energy may be the same thing.
Welcome to metafiction.
Metafiction: fiction (that part is a no-brainer, right?) that is about (meta, meaning “going beyond” or “involving”), well, fiction. It’s a relatively new term for a very old practice. If you own the first edition of The American Heritage Dictionary from the early 1970s, you’ll find it missing. The novelist and philosopher William H. Gass came up with the term in his book of criticism and theory, Fiction and the Figures of Life (1970). He means by the term those works of fiction that self-referentially take as at least part of their concern the making of fiction; words like “self-referential,” “reflexive,” and “self-conscious” routinely pop up in discussions of metafiction. In the first instance, Gass was trying to describe a phenomenon of his own time: the many stories and novels that were then appearing by writers like John Barth, Robert Coover, B. S. Johnson, Christine Brooke-Rose, Italo Calvino, Fowles, and of course Gass himself. But the sort of story he means goes way back. How far back is “way back”? Maybe all the way. It goes something like this: on night one, Caveman Alley told a story, and on night two Caveman Oop told one that, at some point included the phrase, “Now here’s what stories typically do, so I will (a) do it or (b) not do it with my story.” Or he would have done that had parentheses been invented. In fact, Caveman Alley may even have started off with, “So you know how stories go . . .” After all, why wait for night two? In any case, metafiction was born.
You doubt? Tom Stoppard’s Rosencrantz and Guildenstern Are Dead (1966) is a play that toys with a couple of hapless minor figures from Hamlet. So that’s metatheater, a near cousin to metafiction. But so is its source. Hamlet has that whole business of a play within a play and instructions to the players on how to present his little drama, and that’s 1600. Shakespeare does that sort of thing all the time, play after play, where either there’s an actual play, as in A Midsummer Night’s Dream, or there’s an act being put on for someone else’s benefit, as in Much Ado About Nothing. And we know he’s not the first. Geoffrey Chaucer’s Canterbury Tales (1384) involves a set of pilgrims going to the great cathedral town for Easter, during which trip they tell stories. Sometimes the stories conflict with other stories, or animosities from real life spill over into the stories (or vice versa), or, as in the case of the Pardoner and his tale, the teller gets so carried away that he forgets where he is and does something inappropriate, for which the other pilgrims give him a tongue-lashing. There’s a lot of material in the Tales for the savvy critic of metafiction.
But wait, as the infomercials say, there’s more. It turns out the idea isn’t original to Chaucer. The recently deceased Giovanni Boccaccio got there first, with his Decameron. His group of ten young people fleeing Florence during an outbreak of the Black Death each tell ten stories over the course of two weeks. This tidy phalanx of narratives covers a pretty full range of possibility, and the framing device necessarily calls attention to the artificiality of the setup: these are stories made to order for a special situation. The work constantly reminds us that it is a work of fiction, that in fact there’s nothing natural about this enterprise. Is that as far back as we go? Hardly. Homer and Virgil both invoke muses, but with a difference. Homer, to the extent that such a personage existed, was an oral poet-singer; he needed all the breath and wit he could muster, so routinely at moments of great stress he calls on the muses to help him out. Virgil, by contrast, is a writer and not a singer, so no such need of breath or immediate recall; when he calls on the muses, he does so in virtual quotation marks, because that’s what composers of epics do. We could go further back if need be, but you get the point.
So this meta-whatever has been going on for a good little while. What’s that you say? Chaucer and Boccaccio, to say nothing of Virgil, aren’t novelists. Ah, I see. Consider the constituent elements of the novel: book-length, fictive, multicharacter, main plot plus subplots, narrative, prose. What’s missing? Only the last one. And I’ll admit, it’s a big one. But in the time of Chaucer and before, virtually no respectable literature was written in prose. Ergo, verse. Given the interests of all three writers, the complexity and shapeliness of their narratives, they clearly provide models for early novelists. Besides, the word is not “metanovel” but “metafiction.” And never forget this other fact: novelists will steal from anywhere. Anyone. Any time. The magpies of the literary world, novelists.
Small wonder, then, that when we come to the “first” novels, they’re already highly self-aware, reflexive. One might almost say postmodern. If one wanted to get hit. One doesn’t. The point is, we tend to think of self-referential writing as something that comes along late in the day, but in the case of the novel, that would be an error. The first metafictional novel? Don Quixote (1605). Also, as it happens, pretty much the first novel. In English, metafiction makes an appearance at least as early as Tristram Shandy (1759–1769). Both works are shot through with jokey, irreverent textual play—puns, reversals of expectation, print-based gambits like full blank pages, anything that Cervantes or Sterne could think of to do with this new form that they obviously weren’t taking too seriously. And the ink was barely dry on Samuel Richardson’s Pamela before Henry Fielding—he of Tom Jones—offered up his parody, Shamela. In some ways, the early days of the novel were like the Wild West: very few rules and a lot of outrageous activity. Part of that was not having any model of how the novelist must behave toward his novel. There’s no reverence toward tradition before there is a tradition. Those early novelists could do anything, and they did.
Only later did novelistic conduct get codified. I’ve talked elsewhere about the specifics of Victorian publishing practices. What we’re interested in here is the way those practices—serial publication, mass audiences, linear narratives, sprawling canvases—tell us about the narratives themselves. Chiefly, they suggest a regularization of the novel: more and more, novels resembled each other in structure, and for reasons philosophical as well as commercial, tended to be less experimental than in other eras. The later Victorians pursued realism, by which we mean not merely accuracy in portraying the world but a systematic effort to create the illusion of reality, with a zeal that was near-religious. Or maybe not merely “near.” The idea was for readers to snuggle right into a novel as if into the story of their friends’ lives (and readers often did think of these characters, with whom they lived for many months, as bosom friends or mortal enemies). In any case, if your worldview, economic principles, and publishing practices conspire to make the realistic novel the highest form imaginable, you’re probably not going to spend a lot of time calling attention to the artifice of the form. This is not a failing. As a reader, would you want to be reminded, with Tess D’Urberville lying on a slab at Stonehenge waiting to be taken by the police, that we’re just playing with form here? I think not.
Does that mean that Hardy doesn’t notice that he’s sitting for months, building up five hundred or so manuscript pages, that he’s making an artistic creation? Of course not. But it does mean that he chooses not to emphasize the made-ness of the book. That emphasis is called “foregrounding” in the criticism trade, meaning that an aspect of the work is moved to the front in order to call special attention to it. Literary realists, for the most part, tend to foreground story and relegate the messy business of artistic self-awareness to the background. Very deep background.
There’s a tendency to see all this metafiction business as “postmodernism” (a term so amorphous as to be almost entirely bankrupt), as merely a function of a certain historical moment when, presumably, writers had run out of ideas. And in truth, there did seem to be some danger of that. Right around the time of Gass’s book there was a lot of death-of-the-novel talk going round. In 1967, John Barth published his celebrated Atlantic Monthly essay, “The Literature of Exhaustion,” in which he argued that the novel had exhausted its possibilities, that perhaps everything that could be done with the form had been done. Yet a mere fourteen years later, he published a companion essay, “The Literature of Replenishment,” announcing a miraculous recovery by the patient, whose narrative pneumonia had turned out to be a mere case of the sniffles. What a relief!
Yet that tendency of fiction to examine its own bases and practices, to doubt its veracity even as it insists on it, has been there from the first and sometimes shows up in very realistic novels. You remember the soaring Fowles? Chapter 13, right? Well, he wasn’t the first. There’s another novel, also Victorian in form, where right in the middle of the narrative the author pauses, in a chapter called “In Which the Story Pauses a Little,” to lay out her theory of novel writing and to talk about the limitations on what she, as novelist, can know. Hers is Chapter 17, not Fowles’s “unlucky” 13. So who is this innovator? Angela Carter? Jane Smiley? Nope. She’s that great postmodernist, George Eliot, in Adam Bede (1859). Now if Eliot, the realist of realists, displays this self-referential element, even a little, it has to be a strong impulse. Well, of course it is. And why?
Because novels don’t grow in gardens. They are made things, and their makers get their know-how by reading lots and lots of other novels. And you can’t read all of those novels, pick up the old quill, and suddenly forget that you’ve ever read a novel. So here it is (you knew it was coming), the Law of Crowded Desks: When a novelist sits down to begin a novel, there are a thousand other writers in the room. Minimum. There are even writers present whom she’s never read. But that way madness lies, or at least a serious headache.
So what does that mean? Yeah, yeah, one room, lots of buddies. But in practical terms, what does it mean for the writer? It means that nobody’s innocent. You simply can’t sit down to write a novel as if the world isn’t already full of novels. It is already full of novels, and so is your head. That’s just how it is. One response I hear in class to any self-conscious novel is that it’s a stunt, an artificial posturing. My students are right. It is a stunt. Well, guess what? So is the novel that ignores all those other writers in the room. To pretend innocence or lack of self-awareness is exactly that: pretense.
And there’s nothing wrong with pretense. The writer assumes a pose. Always. The narrator who is not strictly the author? A pose. Omniscient, godlike narrator? Definitely a pose. We read novels to follow writers pretending to be things and know things that aren’t exactly true. Self-awareness—or its lack—is just one of those things that gets pretended.
But another pair of practical considerations: you can’t do anything that hasn’t already been done; everything is possible. If you know that everything that can be done has already been done, you have a couple of options. One is to shut down completely, or at least never try anything in that vein. We could see this as the path pursued by Alain Robbe-Grillet and the French anti-novelists of the so-called nouveau roman. They attempted to break down the illusions of the novel, to dismantle the customary operations of fiction—you know, get rid of extraneous things like plot, character, narrative continuity, theme—and reduce narrative to a fragmented series of subjective apprehensions of things. There’s a reason Robbe-Grillet never really caught on in the Anglophone reading world. The other option? Get on with it. Write your novel. Of course you know there have been other novels like yours before now. Of course you can’t be first to the quarter pole. It’s okay. Fiction isn’t a race; you don’t have to come in first to win. In fact, no one’s entirely sure what winning is.
Here, too, you have a couple of options. You can pretend that all those other novels don’t already exist and attempt to write a novel that’s like a window on its world. Jerome Klinkowitz calls this pretense “transparency,” the writer’s attempt to make the apparatus of his novel invisible so that readers see straight through to the contents. Or you can call attention to the artifice of the novel in a strategy he calls “opacity.” A work of fiction is opaque to the degree it requires readers to notice it even as they may be trying to focus on the story. Most fiction ever written has strived for transparency. But not all, and not even all of it we might think would. You can think of in terms of the theater. Does the playwright pretend that the audience isn’t there, that the actors and props are “real,” that the proscenium arch is merely a window? Or does she “break the fourth wall,” address or engage the audience directly, remind us that this is only a play? Or do both? Shakespeare chiefly tells his stories and engages us directly, but periodically he throws in the aforementioned plays within plays and discussions of stagecraft. So, too, with fiction writers. Every once in a while, they’ll surprise you.
Is one way better than the other? Yes. The one that works for a particular author in a particular work is the better one. Do I prefer one to the other? That’s like asking if I prefer pie or cake. Why would they be mutually exclusive? I want Dickens or Hardy or Lawrence to insist on verisimilitude. It would be deeply upsetting if my favorite mystery writers suddenly went all metafictional. But I want Fowles to be a metafictionist, and with him Calvino, Barth, B. S. Johnson, Julian Barnes, Angela Carter.
Then there are the O’Briens. Three of the best novelists of the past, or indeed any, century are named O’Brien. Edna, the girl from a repressive Irish village who grew up to smash all the conventions of propriety surrounding Irish women’s fiction, absolutely insists on transparency. Her mode of storytelling requires that readers be drawn in and not let go until finis. From the trilogy The Country Girls in the 1960s through more experimental works like Night (1972) to the sociopolitical later novels House of Splendid Isolation (1994) and In the Forest (2002), her novels either repel readers due to subject matter or hook them and reel them in from the first pages. In part, Edna O’Brien is working unclaimed land; when she began, there were few, if any, models to follow for the sort of novels she wanted to write. But such self-conscious narrating isn’t really her style in any case. Even when, as in Night, she is wrestling with the ghost of James Joyce—the novel is a reworking of the concept of the Penelope episode of Ulysses, a long, rambling, nocturnal, female meditation by a character looking back over her life so far—there’s only the slightest nod to the prior work. The American novelist Tim O’Brien balances narrative immediacy with a certain degree of reflexivity. That may be the more true in his Vietnam novels, Going After Cacciato (1978) and The Things They Carried (1990). It’s hard to write about war without Hemingway looking over your shoulder, stupid to pretend he’s not. But Tim O’Brien mixes sources from fantastic fiction like Alice in Wonderland in with his Hemingway, even as he immerses us in the experience of war. His novels are a heady brew. And then there’s Flann O’Brien, who as I’ve said before is not even an O’Brien. Born Brian O’Nolan, he took his pen name for his first novel, At Swim-Two-Birds, a hilarious if sometimes confusing pastiche of Irish student novel, American western, Irish epic (much of it is drawn from the legends of Finn MacCool and Mad King Sweeney). A novel within a novel within a novel, At Swim has three openings and three endings, a novelist-character whose characters rebel and take him captive (drugging him so he can’t write them into more difficulty), and a pookah, or demon, who takes control of the offspring of the inner novelist and one of his own characters. Now there’s a form of incest we rarely encounter. Mercifully. His other novels, particularly The Third Policeman (1967) and An Béal Bocht (1941, translated as The Poor Mouth, 1973), do similar damage to the novel form; the genre has never fully recovered.
Does that cover the full range of possibilities for metafiction-or-not? Probably not. Someone’s always figuring out something a bit different, and every writer has different strategies. But these three O’Briens give a sense of what can be done. Besides, in their very different ways, they’re all terrific. One could do worse than to adopt a policy of only reading novels by writers named O’Brien.
So maybe we don’t want a name-controlled reading list. What do metafictional novels look like? Well, like a lot of things.
Some are rewrites of classic stories:
John Barth’s The Sot-Weed Factor (1960, eighteenth-century poem of same name)
Valerie Martin’s Mary Reilly (1990, Dr. Jekyll and Mr. Hyde)
John Gardner’s Grendel (1971, Beowulf)
Julian Barnes’s Flaubert’s Parrot (1984, “A Simple Heart” and Madame Bovary)
Jean Rhys’s Wide Sargasso Sea (1966, Jane Eyre)
Jane Smiley’s A Thousand Acres (1991, King Lear) and Ten Days in the Hills (2007, Boccaccio’s Decameron)
Others are recycling earlier or other forms:
John Fowles’s The French Lieutenant’s Woman (1969, Victorian novel), A Maggot (1985, eighteenth-century primary document assembly, in this case, from a private legal prosecution)
Barth’s Letters (1979, eighteenth-century epistolary novel), The Last Voyage of Somebody the Sailor (1991, A Thousand and One Nights)
Iris Murdoch’s The Unicorn (1963, gothic novel)
Tim O’Brien’s Going After Cacciato (1978, Hemingway and Alice in Wonderland—go ponder that for a while)
T. Coraghessan Boyle’s Water Music (1982, mixture of eighteenth-century adventurer’s tale and picaresque)
Vladimir Nabokov’s Pale Fire (1962, narrative poem and poetic exegesis)
Some are intricate forms of play, as with Tristram Shandy:
Fowles’s The Magus or The Collector, in which dueling narratives call into question the very notion of narrative reliability or veracity
Italo Calvino’s If on a Winter’s Night a Traveler, which explores the world of popular novels by using opening chapters from a host of genres, or Invisible Cities, in which Marco Polo seems to have gone to the court of the Great Khan expressly to describe fabulous and even impossible cities he has passed through en route, or, for that matter, any Calvino
Murdoch’s The Black Prince, which, like Fowles’s The Collector, exploits subjective perception by offering contrary versions and interpretations of events through a primary narrative by the protagonist and commentaries by other characters
They’re interesting, baffling, fascinating, sometimes maddening, but always fresh and new. And that’s one purpose of metafiction, to make it new, because . . .
You can do anything. If a writer can imagine it, she can do it. There are no rules that say, you can only go this high, can only stretch the truth this far, can only be this outrageous.
We readers are sometimes too serious. I think we have to grant our writers a bit of latitude. Let them horse around, joke, trick us and trip us up. What is the novel, after all, but play? Unlike all those nonfiction genres—exposé, pamphlet, sermon, harangue, lab report, news article, biography, history, and, yes, literary essay—fiction cuts itself loose from the moorings of reality so that it can give pleasure, so that the imagination can run free of any constraints but its own. Best of all, it invites us in to play as well. We forget that at our peril.
We are less put off, perhaps, when these sorts of authorial hijinks occur in our popular culture, but we should have the same delight with Angela Carter that we have with Bugs and Daffy. When writers play these metafictional games, they’re merely inviting us to play in their world. Writers since at least Sir Philip Sidney have been telling us that the purpose of literature is to instruct and delight. Fiction about fiction can teach us a lot about our own psychology, about what we expect from stories, and about the nature of, well, fiction. Besides, it’s a whole lot of fun.