More than a century ago, Frank Norris wrote that "the Great American Novel is not extinct like the dodo, but mythical like the hippogriff," an observation that Philip Roth later used as the epigraph for a spoofy 1973 baseball fantasia called, naturally, "The Great American Novel." It pointedly isn't - no one counts it among Roth's best novels, though what books people do place in that category will turn out to be relevant to our purpose here, which has to do with the eternal hunt for Norris's legendary beast. The hippogriff, a monstrous hybrid of griffin and horse, is often taken as the very symbol of fantastical impossibility, a unicorn's unicorn. But the Great American Novel, while also a hybrid (crossbred of romance and reportage, high philosophy and low gossip, wishful thinking and hard-nosed skepticism), may be more like the yeti or the Loch Ness monster - or sasquatch, if we want to keep things homegrown. It is, in other words, a creature that quite a few people - not all of them certifiably crazy, some of them bearing impressive documentation - claim to have seen. The Times Book Review, ever wary of hoaxes but always eager to test the boundary between empirical science and folk superstition, has commissioned a survey of recent sightings.
Or something like that. Early this year, the Book Review's editor, Sam Tanenhaus, sent out a short letter to a couple of hundred prominent writers, critics, editors and other literary sages, asking them to please identify "the single best work of American fiction published in the last 25 years." The results - in some respects quite surprising, in others not at all - provide a rich, if partial and unscientific, picture of the state of American literature, a kind of composite self-portrait as interesting perhaps for its blind spots and distortions as for its details.
And as interesting, in some cases, for the reasoning behind the choices as for the choices themselves. Tanenhaus's request, simple and innocuous enough at first glance, turned out in many cases to be downright treacherous. It certainly provoked a lot of other questions in response, both overt and implicit. "What is poetry and if you know what poetry is what is prose?" Gertrude Stein once asked, and the question "what is the single best work of American fiction published in the last 25 years?" invites a similar scrutiny of basic categories and assumptions. Nothing is as simple as it looks. What do we mean, in an era of cultural as well as economic globalization, by "American"? Or, in the age of James Frey, reality television and phantom W.M.D.'s, what do we mean by "fiction"? And if we know what American fiction is, then what do we mean by "best"?
A tough question, and one that a number of potential respondents declined to answer, some silently, others with testy eloquence. There were those who sighed that they could not possibly select one book to place at the summit of an edifice with so many potential building blocks - they hadn't read everything, after all - and also those who railed against the very idea of such a monument. One famous novelist, unwilling to vote for his own books and reluctant to consider anyone else's, asked us to "assume you never heard from me."
More common was the worry that our innocent inquiry, by feeding the deplorable modern mania for ranking, list-making and fabricated competition, would not only distract from the serious business of literature but, worse, subject it to damaging trivialization. To consecrate one work as the best - or even to establish a short list of near-bests - would be to risk the implication that no one need bother with the rest, and thus betray the cause of reading. The determination of literary merit, it was suggested, should properly be a matter of reasoned judgment and persuasive argument, not mass opinionizing. Criticism should not cede its prickly, qualitative prerogatives to the quantifying urges of sociology or market research.
Fair enough. But there would be no point in proposing such a contest unless it would be met with quarrels and complaints. (A few respondents, not content to state their own preferences, pre-emptively attacked what they assumed would be the thinking of the majority. So we received some explanations of why people were not voting for "Beloved," the expected winner, and also one Roth fan's assertion that the presumptive preference for "American Pastoral" over "Operation Shylock" was self-evidently mistaken.) Even in cases - the majority - where the premise of the research was accepted, problems of method and definition buzzed around like persistent mosquitoes. There were writers who, finding themselves unable to isolate just one candidate, chose an alternate, or submitted a list. The historical and ethical parameters turned out to be blurry, since the editor's initial letter had not elaborated on them. Could you vote for yourself? Of course you could: amour-propre is as much an entitlement of the literary class as log-rolling, which means you could also vote for a friend, a lover, a client or a colleague. But could you vote for, say, "A Confederacy of Dunces," which, though published in 1980, was written around 20 years earlier? A tricky issue of what scholars call periodization: is John Kennedy Toole's ragged New Orleans farce a lost classic of the 60's, to be shelved alongside countercultural picaresques like Richard FariŅa's "Been Down So Long It Looks Like Up to Me"? Or is it a premonition of the urban-comic 80's zeitgeist in which it finally landed, keeping company with, say, Jay McInerney's "Bright Lights, Big City"? What about story collections - I. B. Singer's, Donald Barthelme's, Raymond Carver's, for instance - that appeared between 1980 and 2005 but gathered up the work of earlier decades? Do they qualify? And - most consequentially, as it happened - what about John Updike's four "Rabbit" novels? Only the last two were published during the period in question, but all four were bound into a single volume and published, by Everyman's Library, in 1995. Considered separately, "Rabbit Is Rich" (1981) and "Rabbit at Rest" (1990) might have split Updike's vote, which "Rabbit Angstrom" was able to consolidate, placing it in the top five. If Nathan Zuckerman had received a similar omnibus reissue, with "The Counterlife," "The Human Stain," "American Pastoral" and the others squeezed into one fat tome, literary history as we know it - or at least this issue of the Book Review - would be entirely different.
THE question "what do you mean by 'the last 25 years'?" in any case turned out to be a live one, and surveying the recent past caused a few minds to wander farther back in time. One best-selling author (whose fat novels seem to have been campaigning for inclusion in this issue long before the editors dreamed it up, even though not even he bothered to vote for any of them) reflected on the poverty of our current literary situation by wondering what the poll might have looked like in 1940, with Hemingway, Faulkner and Fitzgerald - to say nothing of Theodore Dreiser, Willa Cather and Sinclair Lewis - in its lustrous purview. The last time this kind of survey was conducted, in 1965 (under the auspices of Book Week, the literary supplement of the soon-to-be-defunct New York Herald Tribune), the winner was Ralph Ellison's "Invisible Man," which was declared "the most memorable" work of American fiction published since the end of World War II, and the most likely to endure. The field back then included "The Adventures of Augie March," "Herzog," "Lolita," "Catch-22," "Naked Lunch," "The Naked and the Dead" and (I'll insist if no one else will) "The Group." In the gap between that survey and this one is a decade and a half - the unsurveyed territory from 1965 to 1980 - that includes Thomas Pynchon's "Gravity's Rainbow" and William Gaddis's "JR," as well as "Humboldt's Gift," "Portnoy's Complaint," "Ragtime," "Song of Solomon" and countless others.
Contemplation of such glories lent an inevitable undercurrent of nostalgia to some of the responses. Where are the hippogriffs of yesteryear? Could they have been dodos all along? Not to worry: late-20th-century American Lit comprises a bustling menagerie, like Noah's ark or the island of Dr. Moreau, where modernists and postmodernists consort with fabulists and realists, ghost stories commingle with domestic dramas, and historical pageantry mutates into metafiction. It is, gratifyingly if also bewilderingly, a messy and multitudinous affair.
It is perhaps this babble and ruckus - the polite word is diversity - that breeds the impulse of which Sam Tanenhaus's question is an expression: the urge to isolate, in the midst of it all, a single, comprehensive masterpiece. E pluribus unum, as it were. We - Americans, writers, American writers - seem often to be a tribe of mavericks dreaming of consensus. Our mythical book is the one that will somehow include everything, at once reflecting and by some linguistic magic dissolving our intractable divisions and stubborn imperfections. The American literary tradition is relatively young, and it stands in perpetual doubt of its own coherence and adequacy - even, you might say, of its own existence. Such anxiety fosters large, even utopian ambitions. A big country demands big books. To ask for the best work of American fiction, therefore, is not simply - or not really - to ask for the most beautifully written or the most enjoyable to read. We all have our personal favorites, but I suspect that something other than individual taste underwrites most of the choices here. The best works of fiction, according to our tally, appear to be those that successfully assume a burden of cultural importance. They attempt not just the exploration of particular imaginary people and places, but also the illumination of epochs, communities, of the nation itself. America is not only their setting, but also their subject.
They are - the top five, in any case, in ascending order - "American Pastoral," with 7 votes; Cormac McCarthy's "Blood Meridian" and Updike's four-in-one "Rabbit Angstrom," tied with 8 votes each; "Don DeLillo's "Underworld," with 11; and, solidly ahead of the rest, Toni Morrison's "Beloved," with 15. (If these numbers seem small, keep in mind that they are drawn from only 125 votes, and from a pool of potential candidates equal to the number of books of fiction by American writers published in 25 years. Sometimes cultural significance can be counted on the fingers of one hand.)
Any other outcome would have been startling, since Morrison's novel has inserted itself into the American canon more completely than any of its potential rivals. With remarkable speed, "Beloved" has, less than 20 years after its publication, become a staple of the college literary curriculum, which is to say a classic. This triumph is commensurate with its ambition, since it was Morrison's intention in writing it precisely to expand the range of classic American literature, to enter, as a living black woman, the company of dead white males like Faulkner, Melville, Hawthorne and Twain. When the book first began to be assigned in college classrooms, during an earlier and in retrospect much tamer phase of the culture wars, its inclusion on syllabuses was taken, by partisans and opponents alike, as a radical gesture. (The conservative canard one heard in those days was that left-wing professors were casting aside Shakespeare in favor of Morrison.) But the political rhetoric of the time obscured the essential conservatism of the novel, which aimed not to displace or overthrow its beloved precursors, but to complete and to some extent correct them.
It is worth remarking that the winner of the 1965 Book Week poll, Ralph Ellison's "Invisible Man," arose from a similar impulse to bring the historical experience of black Americans, and the expressive traditions this experience had produced, into the mainstream of American literature. Or, rather, to reveal that it had been there all along, and that race, far from being a special or marginal concern, was a central facet of the American story. On the evidence of Ellison's and Morrison's work, it is also a part of the story that defies the tenets of realism, or at least demands that they be combined with elements of allegory, folk tale, Gothic and romance.
The American masterpieces of the mid-19th century - "Moby-Dick," "The Scarlet Letter," the tales of Edgar Allan Poe and, for that matter, "Uncle Tom's Cabin" - were compounded of precisely these elements, and nowadays it seems almost impossible to write about that period without crossing into the realm of the supernatural, or at least the self-consciously mythic. This is surely what ties "Beloved" to "Blood Meridian." Both novels treat primordial situations of American violence - slavery and its aftermath in one case, the conquest of the Southwestern frontier in the other - in compressed, lyrical language that rises at times to archaic, epic strangeness. Some of their power - and much of their originality - arises from the feeling that they are uncovering ancient tales, rendering scraps of a buried oral tradition in literary form.
But the recovery of the past - especially the more recent past - turns out to be the dominant concern of American writing, at least as reflected in this survey, over the past quarter-century. Our age is retrospective. One obvious difference between "Invisible Man" and "Beloved," for instance, is that Ellison's book, even as it flashes back to the Depression-era South and the Harlem of the 1940's, plants itself in the present and leans forward, to the point of risking prophecy. "Beloved," in contrast, concerns itself with the recovery of origins, the isolation of a primal trauma whose belated healing will be undertaken by the narrative itself. And while "Blood Meridian" is far too gnomic and nihilistic to claim such a therapeutic function for itself, it nonetheless shares with "Beloved" a vision of the past as an alien realm of extremity, in which human relations are stripped to the bare essentials of brutality and tenderness, vengeance and honor.
In some ways, the mode of fiction McCarthy and Morrison practice is less historical than pre-historical. It does not involve the reconstruction of earlier times - the collisions between real and invented characters, the finicky attention to manners, customs and habits of speech - that usually defines the genre. But to look again at the top five titles in the survey is to discover just how heavily the past lies on the minds of contemporary writers and literary opinion makers. To the extent that the novel can say something about where we are and where we are going, the American novel at present chooses to do so above all by examining where we started and how we got here.
IF "Beloved" and "Blood Meridian" pull us back to a premodern American scene - a place that exists beyond realism and in some respects before civilization as we know it - the other three novels trace the more recent ups and downs of that civilization. Indeed, it is only a small exaggeration to say that "Underworld," "American Pastoral" and "Rabbit Angstrom" are variations on the same novel, a decades-spanning tale rooted in the old cities of the Eastern Seaboard. Needless to say, the methods, the characters and the voices are quite distinct - no one would mistake Roth for DeLillo or Updike for Roth - but these are differences of perspective, as if three painters were viewing the same town from neighboring hillsides.
The three novels do what we seem to want novels to do, which is to blend private destinies with public events, an exercise that the postwar proliferation of media simultaneously makes more urgent and more difficult. Rabbit Angstrom, high school basketball star, typesetter-turned-car-dealer, as carelessly loyal to his country as he is unfaithful to his wife, is an incarnation of the American ordinary made exemplary by the grace of God and of Updike's prose. Especially in the later novels, his consciousness becomes the prism through which the unsettled experience of the nation is refracted. The war in Vietnam, the racial agitations of the 60's, the moon landing, the Carter-era malaise, the end of the cold war: all of these are filtered through Rabbit's complacent gaze. So are less dramatic but no less consequential shifts in manners and morals, in taste and sensibility. Food, sex, cars, real estate, social class, religion - everything changes from "Rabbit, Run" to "Rabbit at Rest," even as the deep continuities of American life, embodied in the hero's transcendent laziness, appear to triumph in the end.
"Rabbit Angstrom" is not, strictly speaking, a novel of retrospect; it was written in the present tense and in real time, each segment composed before the end of the story could be known. Because of this - because Updike's gift for observing the present has always outstripped his ability to animate the past - "Rabbit," like the great Russian and French realist novels of the 19th century, becomes an unequaled repository of historical detail. Next to it, Updike's attempted multigenerational chronicle of 20th-century American history, "In the Beauty of the Lilies," looks thin and stagy.
Alongside Rabbit there is Zuckerman, his near contemporary, and like him the product of a small, industrial mid-Atlantic city. More pointedly, perhaps, there is Swede Levov, the hero of "American Pastoral" (Zuckerman being the self-effacing narrator), who is, like Rabbit, a star athlete in high school and whose nickname curiously recalls Rabbit's ethnic background. But while Rabbit is, for all the suffering he endures and inflicts, a fundamentally comic character, his destiny arcing toward happiness, Swede's trajectory is tragic. Fate has raised him high in order to see how far he might fall. He contains traces of Job - his fidelity to America tested by brutal and arbitrary misfortune - and also of Lear, snakebit by one of the most floridly and obscenely ungrateful children in all of literature.
The agonized question that ripples through "American Pastoral" is "what happened?" How did the pastoral America of Newark in the 40's and 50's - an Eden only in retrospect - come apart? And its selection over Roth's other books is indicative of how important this question is taken to be. Over the past 15 years, Roth's production has been so steady, so various and (mostly) so excellent that his vote has been, inevitably, split. If we had asked for the single best writer of fiction over the past 25 years, he would have won, with seven different books racking up a total of 21 votes. Within these numbers is an interesting schism. The loose trilogy of which "American Pastoral" is the first installment - "I Married a Communist" and "The Human Stain" are its companions - accounts for 11 votes, while 8 are divided among "Sabbath's Theater," "The Counterlife" and "Operation Shylock," and another 2 go to "The Plot Against America." The Roth whose primary concern is the past - the elegiac, summarizing, conservative Roth - is preferred over his more aesthetically radical, restless, present-minded doppelgnger by a narrow but decisive margin.
A similar split occurs among DeLillo's partisans, who favor the historical inquiry of "Underworld" over the contemporaneity of "White Noise." (There were also two voters who chose "Libra," a more narrowly focused historical fiction and in some ways a rehearsal for "Underworld.") Like "American Pastoral," "Underworld" is a chronologically fractured story drawn by a powerful nostalgic undertow back to the redolent streets of a postwar Eastern city. Baseball and the atom bomb, J. Edgar Hoover and the science of waste disposal are pulled into its vortex, but whereas Updike and Roth work to establish connection and coherence in the face of time's chaos, DeLillo is an artist of diffusion and dispersal, of implication and missing information. But more than his other books, "Underworld" is concerned with roots, in particular with ethnicity. Nick Shay, at first glance another one of his tight-lipped, deracinated postmodern drifters, turns out to be a half-Italian kid from the old East Bronx, and the characteristic rhythms of DeLillo's prose - the curious noun-verb inversions, the quick switches from abstraction to earthiness, from the decorous to the profane - are shown to arise, as surely as Roth's do, from the polyglot idiom of the old neighborhood.
So the top five American novels are concerned with history, with origins, to some extent with nostalgia. They are also the work of a single generation. DeLillo, born in 1936, is the youngest of the five leading authors. The others were born within two years of one another: Morrison in 1931, Updike in 1932, Roth and McCarthy in 1933.
Their seniority, needless to say, is earned - they have had plenty of time to ripen and grow - but it is nonetheless startling to see how thoroughly American writing is dominated by this generation. Startling in part because it reveals that the baby boom, long ascendant in popular culture and increasingly so in politics and business, has not produced a great novel. The best writers born immediately after the war seem almost programmatically to disdain the grand, synthesizing ambitions of their elders (and also some of their juniors), trafficking in irony, diffidence and the cultivation of small quirks rather than large idiosyncrasies. Only two books whose authors were born just after the war received more than two votes: "Housekeeping," by Marilynne Robinson, and "The Things They Carried," by Tim O'Brien. These are brilliant books, but they are also careful, small and precise. They do not generalize; they document. Ann Beattie, born in 1947, is among the most gifted and prolific fiction writers of her generation, but her books are nowhere to be found on this list; not, I would venture, because she fails to live up to the survey's implicit criterion of importance, but because she steadfastly refuses to try.
Expand beyond the immediate parameters of this exercise, and the generational discrepancy grows even more acute: add Thomas Pynchon and E. L. Doctorow, Anne Tyler and Cynthia Ozick, John Irving and Joan Didion and Russell Banks and Joyce Carol Oates and you will have a literary pantheon born almost to a person during the presidency of Franklin Roosevelt. Further expansion - by means of a Wolfe here, a Mailer there - is likely to push the median age still higher. Think back on that 1965 survey; it's hard to find an author on the list of potential candidates much older than 50.
IS this quantitative evidence for the decline of American letters - yet another casualty of the 60's? Or is the American literary establishment the last redoubt of elder-worship in a culture mad for youth? In sifting through the responses, I was surprised at how few of the highly praised, boldly ambitious books by younger writers - by which I mean writers under 50 - were mentioned. One vote each for "The Corrections" and "The Amazing Adventures of Kavalier & Clay," none for "Infinite Jest" or "The Fortress of Solitude," a single vote for Richard Powers, none for William T. Vollmann, and so on.
But the thing about mythical beasts is that they don't go extinct; they evolve. The best American fiction of the past 25 years is concerned, perhaps inordinately, with sorting out the past, which may be its way of clearing ground for the literature of the future. So let me end with a message to all you aspiring hippogriff breeders out there: 2030 is just around the corner. Get to work.
A. O. Scott is a film critic at The Times. He is writing a book on the American novel since World War II.