A new kind of disenchantment has come over literature. It has to do with what you might call the working myth of the life of literature — the half-conscious way that people decide which texts they consider literature, and how they carry those texts forward. The catalyst, I believe, is the recent revolutionary advance in counting. That may not sound like a startling technological breakthrough, but thanks to computers, we are now able to count with unprecedented speed and thoroughness. Last August, for example, a computer programmer named John Matherly sent a simple “Are you there?” message to every device with a direct, public connection to the Internet. Within five hours, about 400 million machines responded, and after twelve hours of analysis, he was able to draw a map of their locations around the world. Imagine trying to contact, count, and map all the people in the world by yourself; because they aren’t (yet) all connected to the Internet, you wouldn’t be likely to live long enough to finish.
Counting has changed the world before. Consider Europe and America in the two or three centuries before 1750, when society had a structure that was still half-feudal. Government taxes were collected by private individuals, who kept a cut for themselves. No matter how excellent a soldier’s performance, he couldn’t hope for a career as an officer unless he bought his way in. Men in public life were extremely touchy about their honor: if a reputation was slighted, the libeled party had to exchange pistol shots with his libeler at dawn, or else forfeit his social standing. In The Institutional Revolution, Douglas W. Allen argues that these peculiar conventions made economic sense in their day. Because standards of measurement were inconsistent and record-keeping was haphazard, it wasn’t possible to know from a distance precisely what someone else was up to. The best way to keep a person from slacking was to let him skim a little, and the best way to keep him from cheating was to make the occasional exposure of dishonor a matter of life and death. In such a world, loyalty mattered more than talent; Voltaire wasn’t kidding when he wrote that “it pays to shoot an admiral from time to time to encourage the others.”
And then, between 1750 and 1850, everything changed. Lengths and weights became standardized; time-keeping mechanisms were improved and clocks became widely distributed; bureaucracies took charge of record-keeping. People stopped leaning so heavily on trust when the new technologies made it easier to verify. It was no longer necessary to shoot the person who claimed your hand was in the cookie jar once it became possible to show him, instead, an up-to-this-morning inventory of the cookies.
Today’s breakthrough in counting is at least as radical as the one that took place at the onset of the Industrial Revolution, and we now find ourselves in the process of adjusting our social norms to the new transparency of our actions. We are also in the process of fighting over the terms of that adjustment. We fight about whether to replace the personal judgment of teachers with standardized curricula and frequent testing, whether it’s ethical for employers to track the keystrokes and body movements of workers, whether we’re comfortable with retailers having the intimate knowledge of ourselves that they’re able to piece together from our purchasing histories, and whether we trust our governments with the power to monitor our phone calls and emails. We haven’t yet had a good fight about the intrusion of counting into the life of literature, however. Maybe we should.
At the end of 1818, John Keats began a long letter to George and Georgiana Keats, his brother and his sister-in-law, who were trekking down the Ohio River toward Louisville, Kentucky. No telegraph cable yet spanned the Atlantic, and the Keatses depended on private shipping companies to carry their ink-on-paper messages. Transit was slow. A few years earlier, the United States and Great Britain had ended the War of 1812 by signing a peace treaty in Ghent, in modern-day Belgium, but official notification didn’t reach New Orleans for almost two months. In the interval, Andrew Jackson fought a bloody military campaign that a telegram would have rendered superfluous.
Measuring by the delay in their messages, Keats and his brother’s family were farther from each other than it may be possible for people today to be. The poet suggested an unconventional way of bridging the distance. “I shall read a passage of Shakespeare every Sunday at ten o’clock,” he proposed. “You read one at the same time, and we shall be as near each other as blind bodies can be in the same room.”
I’d suggest that Keats was only half joking. In his poetry, after all, he associated literature with the power to transcend time and space, writing, for example, that George Chapman’s translation of Homer made him think of the first glimpse that the Spanish conquistador Cortés had had, centuries earlier, of the vast Pacific Ocean. Keats returned to the idea of a book-mediated connection in a letter written about a year later to his lover, Fanny Brawne. “Do not send any more of my books home,” he wrote on February 24, 1820, after tuberculosis had confined him to bed in Hampstead for three weeks. “I have a great pleasure in the thought of you looking on them.”
There’s something mystical about these passages. If Keats wasn’t in the room with Brawne, he couldn’t know for certain at any given moment whether she was looking at his books. He couldn’t know which pages she turned, or how quickly. In fact, he would have had to take her word for it that she looked at his books at all. But he maintained nonetheless that he felt real pleasure when he thought of her reading his books. The simile in Keats’s letter to his brother and sister-in-law is a strange one, and worth repeating: “As near each other as blind bodies can be in the same room.” “Blind” because presumably Keats and his brother’s family wouldn’t be able to see each other, but “in the same room” because somehow they would be able to feel each other’s presence.
What does it take to believe in such a communion? I think it requires the belief that reading, or at least a certain kind of reading, is sensuous, invisible, and soulful. Each instance of this kind of reading is unique. In its ideal form, it occurs on a plane that is oblique to the physical location of the people doing it, even when they happen to be in the same room.
This isn’t to say that the particular bodies of the readers are irrelevant. If they were, then the communion could be reduced to the content of the text being shared, and Keats wasn’t offering to mail his brother and sister-in-law a copy of Shakespeare’s plays. The hope, in his half-joke, is that Shakespeare’s words would call out a response in their souls, as instantiated in their bodies, that resonated so strongly with the response called out by the same words in his soul, as instantiated in his body, that he and his relatives would feel connected.
A year and a half ago, my husband and I went on vacation. We took a lot of pictures, as one does, and uploaded a batch to Facebook. After we got back to New York, my husband had lunch with his friend Peter Mendelsund, a book designer, as he does almost every week. Mendelsund told my husband that he really liked our vacation pictures.
“If you liked them, how come you didn’t ‘like’ them?” my husband asked.
“I thought I did ‘like’ them,” Mendelsund replied.
“One or two,” my husband said accusingly.
By the end of the day, Mendelsund had “liked” several dozen more.
“But you missed a couple,” my husband called to tell him.
I suspect that my husband, like Keats, was only half joking. Thanks to the mediation of our social lives by computers, we have become so habituated to having evidence of the mental states of other people that we no longer quite believe in those mental states when evidence is lacking. If George Keats didn’t “like” Shakespeare, did he really like him? Pics or it didn’t happen. John Keats couldn’t have known which pages of his books Fanny Brawne turned. But we live in a world in which e-book retailers do know which pages their customers turn. In December 2014, the bookseller Kobo reported that although Solomon Northup’s 1853 memoir Twelve Years a Slave was its ninth-highest seller in Great Britain that year, only 28.2 percent of purchasers got to the end of it.
Is it still possible to believe in a literary communion that takes place in silence and at a distance, and that leaves behind no evidence of any kind?
One temptation, when a thing is countable, is to imagine that instances of it are interchangeable. The average, rather than the ideal, becomes the archetype. There’s little point in counting, after all, if you can’t take the mental shortcut of assuming that the aspects of a thing that can’t be counted don’t matter. That is the basic trade-off at the heart of economics, which treats human desire as more or less fungible, even though most of us experience desire as particular and various. In exchange for this procrustean simplification, economics acquires a powerful predictive capacity. There are signs that the humanities today envy that capacity and are ready to accept a similar simplification.
If you think that works of literature are fungible, though, it begins to seem a little silly to believe that a particular book could speak to you personally. Wouldn’t another do just as well? In Zealot, the religious-studies scholar Reza Aslan pointed out that Jesus was one of a number of Aramaic-speaking magician-messiah figures with a revolutionary message in first-century Jerusalem. Aslan argued that it was reasonable to assume that Jesus resembled his peers, and suggested that an amalgamated portrait of these magician-messiahs would be tantamount to a portrait of Jesus. The trouble is that even though such an assumption might be sensible in economics, it isn’t quite safe in history, and in religion it won’t do at all. Christianity, even in its mildest, least doctrinal forms, involves the belief that Jesus was an outlier — not only unique in his day but unique for all time. People believe something similar about great works of literature, or at least they used to. In an essay that argues tacitly, and somewhat embarrassingly, for his own greatness, Wordsworth wrote that whereas bad popular poetry is immortal as a species — it bolts up and dies out, and is replaced by poetry just as bad and just as popular and just as ephemeral a day later — a good poem is immortal as an individual.
Under the old dispensation, an act of reading, too, could be special. If we no longer believe in this possibility, the humanities classroom looks awfully inefficient. Why grant novice readers a meatspace conversation about literature with an expert when expert knowledge can be broadcast cheaply over a computer network? The luxury doesn’t make sense unless learning about literature is understood to consist of something more than the transmission of data. It’s only if a conversation between a teacher and student is understood to create meaning, rather than merely transfer it, that an opportunity seems to be lost.
A second temptation, in a counted world, is to imagine that no single instance of a thing matters — that the individual case is no more than a rounding error. In the old myth, by contrast, it was possible to believe that a work of literature succeeded if it reached just one person for whom it was a key. A couple of years ago, when I was trying to convince the administrators of the New York Public Library not to knock the bookshelves out of their landmark research building at Fifth Avenue and 42nd Street in Manhattan, one of the library’s publicists pointed out that in the previous year, patrons asked to see only a minute proportion of the books stored there — only 300,000 out of roughly 4 million, or about 7.5 percent. It was impossible to know whether this number was higher or lower than in earlier years — the library had only recently finished computerizing its circulation system — and in public, I argued that a number without a context couldn’t plausibly justify a change in policy. But in private, I had to admit that I found the number strangely demoralizing. Someday it will probably become possible to estimate a book’s chance of finding not its one true reader but any reader at all.
My dismay was naïve; the unreadness of library books has been noticed before. “No place affords a more striking conviction of the vanity of human hopes, than a publick library,” Samuel Johnson wrote in The Rambler in 1751. Lorin Stein, the editor of The Paris Review, has even said that he finds it consoling to visit a bookstore in much the way that it’s consoling to visit a cemetery: there’s a peace to be had in knowing that someday we’ll all be forgotten. Quantifying these melancholy convictions, however, brings them home, at least to me, in a new way. You start to wonder whether there’s a back-of-the-napkin calculation that could determine once and for all whether the creative effort is worth it.
The third, and perhaps most crucial, temptation that besets those who count is to equate the value of a thing with the popularity of it. You may like a singer, but if he were really a genius, wouldn’t more people be downloading his song? It might seem to you that a reporter has exposed a grave threat to the republic, but why aren’t more people clicking on her article?
Writers have long protested against measuring worth by popularity, as H. J. Jackson recounts in Those Who Write for Immortality, a new study of the mechanics of literary fame. The Roman poet Horace, Jackson notes, was frankly elitist, writing that “ ’Tis enough if the knights applaud me.” Elitism remained unabashed as late as 1625, when Francis Bacon wrote of praise, “If it be from the common people, it is commonly false and naught.” The rhetoric shifted in the eighteenth century, as people began to doubt that social hierarchy was a positive good in itself. David Hume suggested in his 1757 essay “Of the Standard of Taste” that “durable admiration” was a better index of aesthetic value than the momentary good opinion of any set of readers. In the preface to his 1765 edition of Shakespeare’s plays, Samuel Johnson argued that the best judgment of literary value was gradual and cumulative. Literary merit was the sort of thing that had to be discovered by experiment and assessed by means of comparisons, Johnson wrote, and “no other test can be applied than length of duration and continuance of esteem.”
Hume and Johnson made it sound as though they were taking the judgment of literary value out of the hands of a sociopolitical elite and submitting it to an impersonal process — time, or maybe history. In 1919, T. S. Eliot called his vision of the process “tradition.” “What happens when a new work of art is created,” Eliot wrote,
is something that happens simultaneously to all the works of art which preceded it. The existing monuments form an ideal order among themselves, which is modified by the introduction of the new (the really new) work of art among them.
One imagines a Greek temple, glowing at night, whose columns are somehow living beings — daunting, ethereal, austere, yet not completely unwelcoming if one happens to be wearing just the right peplum.
To believe in the old myth of literature, it wasn’t absolutely necessary to subscribe to Eliot’s fantasy. But some notion of a literary canon was essential to the ideal of soulful reading, because not all texts repaid soulful attention. If one was building up one’s soul, one wanted to know about as many texts as possible that did. There needed to be a way to pass news of them along.
In my imagination, at least, the transmission happened along the following lines: The canon was entrusted to an elite, but it wasn’t necessarily coterminous with any particular sociopolitical elite. Anyone who could persuade another person to listen to her literary opinions belonged. It was a kind of freemasonry, crossing time as well as space. Though some communications were transmitted instantly, others might not reach another member for years, perhaps centuries.
Within this elite, knowledge of literary tradition was respected because it gave a reader a greater familiarity with what literature was capable of — a wider range for comparison. In most cases, a professor of literature belonged, as did a newspaper reviewer, even though the taste of each probably struck the other as disfigured somewhat by the mental habits of his profession. A publisher, by virtue of choosing which titles to print or reprint, was eligible, as was a bookseller, who exercised a vote when he stocked a title. A reader took part when he made a purchase that supported the publisher in his choice and the bookseller in his seconding of it, and he voted again when he recommended the book to a friend. I worked in the town library when I was in high school, and one of the librarians, a former nun, used to wander through the stacks from time to time and save her favorite books from being discarded by stamping them with false due dates. She, too, was participating.
It was impossible, in the old days, to quantify exactly the power of any one voice in this imaginary elite. A stray remark by someone’s uncle might be decisive, if it led a young listener to a hand-me-down copy of a book that she fell for so hard that she was inspired, years later, to write a critical essay — or even, perhaps, a new work of literature — that cemented the book’s connection to the living tradition. In other words, it was possible to imagine, with E. M. Forster, that “the final test of a novel” — or any book — “will be our affection for it.”
Or rather, posterity’s affection. For better or worse, it wasn’t possible to know posterity’s verdict in advance. To writers who found it hard to live with the uncertainty, Eliot advised writing with a “historical sense,” in the apparent belief that a style that was knowing about the literary tradition would be more likely to be welcomed into it. Wordsworth, more boldly, suggested that a writer with a future should expect controversy in the present, and even welcome it. “Every author,” Wordsworth wrote, “as far as he is great and at the same time original, has had the task of creating the taste by which he is to be enjoyed.”
Jackson shows in Those Who Write for Immortality that a reputation like Wordsworth’s was in fact shaped not by his literary merit alone but also by quirks of publishing history, unforeseeable shifts in readerly taste, and acts of advocacy and partisanship. If she could travel back in time two centuries, she would advise a young Romantic poet to choose a house suitable for conversion into a shrine, to instill devotion in younger relatives who would one day manage the literary estate, to cultivate a personal myth contradictory enough to keep biographers occupied, and to write short lyrics because long narrative verse was about to go out of style.
Jackson considers the canon a bit of a sham, and potentially dangerous. It’s true that the informality of canon formation probably did make it easier to exclude writers on the basis of race, gender, class, and sexual orientation, but Jackson is concerned for the most part with less categorical biases, and I’m not persuaded that her exposure of the gears, and of the grit that has sometimes compromised the gears, puts the myth entirely out of commission. Although she tries to identify poets who deserve a higher place in literary history than they have received, she admits to misgivings about one of her candidates — Barry Cornwall — and compares the work of another — Robert Bloomfield — to the text of greeting cards. She seems most enthusiastic about rehabilitating Leigh Hunt, a mentor of Keats’s, but much of her praise is for Hunt’s personality; her description of his poetic language is brief and general. It turns out to be difficult to alter the canon. Jackson explains that those who are passed over by literary history are at a disadvantage because “their more successful counterparts were used to establish standards” according to which we now make literary judgments — an explanation that uncannily resembles Wordsworth’s claim that a great writer creates the taste by which he is to be enjoyed. Through its resistance to being dislodged, the criterion of literary merit seems to offer some proof of its existence.
Moreover, if Jackson were to prove one of her claimants worthy, she wouldn’t thereby overthrow the canon. She would refine it. The canon has long been understood to be an imperfect and continuous approximation. Jackson quotes a warning that William Godwin issued in 1797: “The most which a successful author can pretend to, is to deliver up his works as a subject for eternal contention.” Instability is part of the myth’s appeal, solacing authors who feel underappreciated — who hope that the judgment of posterity in, say, 2015 might be revised come 2050. The canon is a mystical sum, which can never be tallied; its only true index is written in living and fallible hearts.
This myth of unknowability is being replaced today by an illusion of certainty. As I write this sentence, the Amazon sales rank of John Keats’s Selected Letters is 796,426, and the new Oxford Authors edition of William Wordsworth’s poetry and prose has a rank of 2,337,250. It’s hard to look away from such numbers, which are objective even when the data that they summarize is incomplete and the formula that generates them has been kept secret. Numbers are easier to compare than opinions, which may be why opinions now are so often accompanied by a rating of one to five stars.
What happens to the canon in such a world? Maybe it lapses into no more than a list of books with high network externalities — books that others in your peer group are reading and that it will benefit you to read, too, because they are more likely to come up in conversation and serve as opportunities for you to show off and forge connections. Or maybe the canon becomes what an algorithm guesses you might like, through a Bayesian analysis of preferences that you and others have already registered. Let the computer run its algorithm, and you will, by some lights, be getting rid of literary criticism’s last few sticks of Biedermeier furniture. The preferences that the algorithm taps into are the result of human judgments about literary value, and the algorithm no doubt draws on more of them, more systematically, than any member of my fading freemasonry was ever likely to have been able to. So why does the algorithm leave me uneasy? The Internet companies that manage these recommendation lists are free to sell spots on them to the highest bidder, but leave that aside as an obvious corruption, which could in theory be remedied. What bothers me, I think, is that with an algorithm, humans are cut out of the loop by computer processing of their own earlier judgments — like voters in California preempted by election predictions based on votes cast on the East Coast, except in this case, the voters have somehow been preempted by themselves. Living memory has the power to shape and to create, but artificial memory only repeats what it was told.
“Where wisdom once was, quantification will now be,” Leon Wieseltier, formerly of The New Republic, now a contributing editor at The Atlantic, wrote in a recent lament. And get off my lawn, someone on the Internet inevitably jeers whenever quantification is questioned. There’s something democratic about numbers, after all. Resistance to them, in the field of literary judgment, implies a belief that appreciation by some readers is worth more than appreciation by others. The sophisticated understanding today is that “highbrow” is just another market segment. Whether a show gets renewed depends on its ratings and ad sales, not the caliber of its audience’s taste. Aren’t reviews on Amazon and ratings on Goodreads the voice of the people speaking? Wordsworth raised a similar question, rather defensively, at the end of his essay decrying popularity as a criterion of literature. It would be slander, he insisted, for anyone to claim that he didn’t respect the literary judgment of the people. He just didn’t happen to acquiesce in the judgments being made by the people at the time he was writing.
I sympathize. It can still be hard for a writer to make such an acquiescence. Here’s an excerpt from the first Amazon review I ever received:
Crain is a prissy snob and NYT regular who has written two books one of which is out of print. These volumes will find their place among the unread and unremembered exercises in effete yuppie lifestyle decoration. . . . Trash only of interest to sheltered Yale frat boys going through they’re [sic] mandatory Feminist/Homosexual postmodern semiotics discourse potty training.
Hume and Johnson were right about judgment maturing with time, at any rate. The review stung when I first read it, thirteen years ago, but these days I’m strangely fond of it.
However politically awkward it may be to say so, of course the appreciation of some readers is worth more than that of others. It’s because numbers are democratic that they are usually accurate about existing popular preferences and usually wrong about the ultimate value of literary works. The opinions on the Internet today are a modern form of those that circulated, informally and rather more chaotically, in my imaginary freemasonry. Indeed, the literary judgments that appear on the blog of an avid reader, say, or as a contribution to an online literary review that is developing a coherent sensibility, such as Public Books, the Los Angeles Review of Books, or Open Letters Monthly, strike me as the old myth’s forlorn hope in a new realm. It’s when online companies solicit and host opinions so as to be able easily to quantify and aggregate them that the damage happens.
Hume suggested the nature of the damage in “Of the Standard of Taste.” Numbers and counting may belong to the province of reason, but Hume argued that reason is of limited use in judging literature. It’s impossible, he believed, to deduce the laws of good writing, for one thing. Reason might at first appear capable of producing a few generalizations, but “when critics come to particulars,” he warned, “this seeming unanimity vanishes.” All reason can do is deduce the qualities of a good critic — a delicate palate, practice, a faculty for making comparisons, freedom from prejudice, and good sense — and then conduct an empirical debate about which critics possess those qualities. He believed, in short, that literature knows itself only through critics, and that the critics have to be evaluated. I would argue that only a community can conduct such an evaluation, and that the most common vehicle for such evaluation is, or at least used to be, the book review.
Book reviews are usually thought of as evaluations of books, not of the people who write them, but the judgments involved in a review are in fact taking place on several planes and along several angles. It’s impossible for a critic to judge a book unless she holds it to a standard, which may take the form of a rule, such as “Language should be no fussier than what’s needed to convey its message,” or of a personal touchstone, such as Sybille Bedford’s A Legacy. But where does the authority for the standard come from? Another reviewer might choose a different rule, such as “Language should dazzle,” or a different touchstone, such as James Joyce’s Ulysses. A further challenge is that literature changes, and the standards for judging it must therefore also change. I believe that the critic appeals for her authority to the community that she shares with her readers — a community that in the days of print was largely constituted by the writers for and readers of a particular periodical — and that the community chooses its standards through conversation.
The conversation is asymmetric: the critic proposes, and the reader — often through his representative, the critic’s editor — disposes. In every review, a critic is proposing both a judgment of a work of literature and, sometimes explicitly, a standard for judgment. In every review, a critic asserts, at least tacitly, what she believes literature to be. Her definition may be a little different from the one she proposed last week, and it may be a little different from one proposed by another reviewer at the same publication, but there will usually be a family resemblance. The critic is not free to redefine literature at whim. Readers would exit the community if the critic’s understanding were to drift so far from theirs that her reviews no longer helped to identify what they sought in literature. In the old days, it was the responsibility of editors to forestall the exit of readers by dropping such critics. But a critic who flunked out of one community of readers might be welcomed by another, because every publication had different standards of judgment. The New Republic might be willing to pan a book celebrated in The New York Times Book Review; in the old days, it often did.
The Internet has damaged the coherence of this system in a number of ways. By undermining the profitability of newspapers, the Internet destroyed most local book-reviewing communities in America. But the more general problem now is that literary judgment is usually severed from the context that gave rise to it. Online, the community who reads a review often has no borders. Many readers arrive at online reviews by way of Twitter or Facebook, not because they devotedly visit a homepage, and they aren’t likely to reflect on their loyalty to the publisher of the review. They aren’t, in most cases, subscribers. No one ever canceled his connection to the Internet because he read something on it that he disagreed with. It is one thing to conduct a tacit conversation about literary standards with tens or even hundreds of thousands of subscribers, and it is another to conduct such a conversation with all speakers of the English language with an Internet connection. There is no longer any exit. No reader can get out, and no critic can be kicked out. You may feel that Amazon reviews are so unreliable that you’d rather not read them, and you are free to declare that you’ll never read one again. Good luck with that. A particular customer review may strike you as so mean-spirited and off topic that you think Amazon should take it down. Good luck with that too.
Most reviews today, cut off from the communities that once fostered and disciplined them, have no authority. “All sentiment is right,” Hume wrote, “because sentiment has a reference to nothing beyond itself.” A market can be driven by sentiments alone, and a marketplace can be exploited with great efficiency by those who are able to count sentiments and analyze them. But literature was supposed to be able to do more than fulfill wishes and confirm preferences.
Literature will survive if readers declare war on counting, if they insist that literature is defined by the judgment of the ideal critic and not the average one, and if they are able to build new communities of critics and readers with borders that are porous and expansive but nonetheless meaningful. “For this week past,” Keats wrote to Fanny Brawne, on July 4, 1820, “I have been employed in marking the most beautiful passages in Spenser, intending it for you, and comforting myself in being somehow occupied to give you however small a pleasure.” The communion imagined by Keats here is on a continuum with those he imagined in his other letters to Brawne or to his brother, but in this case no mysticism is required. As soon as Keats was healthy enough, he would be able to visit Brawne and share with her the Spenser verses that he had marked. But I wonder if the sharing, when it took place, would have been able to bring him as much pleasure as his imagination of the sharing had. Or to put it another way, I wonder if what he would have most enjoyed, in the act of sharing, was his imagination of Brawne’s pleasure — which, even if she were sitting beside him, would have been invisible to him — and his imagined perception that it brought their souls together. The deepest literary pleasures, even when they involve others, are a little dreamy and lonely.