Six-hundred-page books about economics translated from French don’t usually become best-sellers. Part of the reason Thomas Piketty’s Capital in the Twenty-First Century has been so widely read is that it refuses to be just a book about economics. It traces the history of economic inequality with graphs of wealth and income, arguing that the past several decades have seen soaring disparity between the 1 percent and the rest of us. But it also shows how inequality shaped individual lives with stories drawn from novels. When Piketty spoke to President Barack Obama’s Council of Economic Advisers in April, even they responded to the literary aspect of his work, quibbling over his interpretation of Balzac.
As literary historians, we’re thrilled to see economists arguing over details in a novel. But Piketty’s claims about fiction and inequality are important enough to probe in more depth, which is why we decided to test some of them on a scale only recently made possible by computers. Piketty’s account of literary history turns out to be wrong—but wrong in a way that casts a surprising new light on the way novels do respond to the changing economic fortunes of people in the real world.
Novels by Balzac and Jane Austen matter for Piketty because they dramatize the immobility of a 19th-century world where inequality guaranteed more inequality—a world our own century is beginning to resemble once again. Since returns on capital were reliable, especially for large fortunes, the best way to get ahead was to start out ahead; income from labor could never catch up. The stability of 19th-century wealth is felt not only in plots that center on inheritance, but also, Piketty adds, in the references that flesh out a fictional world. “Specific references to wealth and income were omnipresent in the literature of all countries before 1914,” he writes, because money was a stable social reference point. After 1914, the inflationary shocks associated with two world wars briefly reduced inequality by devaluing capital. But inflation also rendered “the meaning of money ambiguous,” because prices now dated too rapidly to be meaningful for long. In response to this new fluidity, novelists simply stopped counting money; “money—at least in the form of specific amounts—virtually disappeared from literature.”
Surprisingly, Piketty supports this sweeping claim about the history of literature “in all countries” with evidence from only four authors: Balzac, Austen, Naguib Mahfouz, and Orhan Pamuk. And while economists like Piketty have been borrowing vivid examples from literature, literary critics have been borrowing statistical methods from the social sciences to study large collections of digital texts. Unfortunately for Piketty, one thing we’ve learned is that he’s wrong to believe money “gradually dropped out of sight between 1914 and 1945.” But in the process of testing that claim, we’ve also discovered some eye-opening changes in novels that do loosely parallel his argument.
We can use computers—in this case, analytical software we wrote—to figure out how often novelists mention “specific amounts” of money because, to do that, writers usually write a price symbol or mention a unit of currency. It’s not quite a simple problem of counting words: Pounds, for instance, can refer to a lot of things that aren’t money. But by combining statistical models and old-fashioned elbow grease, we’ve counted references to amounts of money in 7,700 English-language works of fiction between 1750 and 1950, drawn in part from HathiTrust Digital Library. This is what we see:
It’s exactly the reverse of Piketty’s story about the disappearance of money. References to specific sums increased in frequency from 1825 to 1950, undeterred by the accelerating pace of inflation. (World literature may not follow exactly the same trajectory as the English-language fiction we surveyed, but since Piketty claimed that money disappeared from “the literature of all countries,” this is at least a significant counter-example.) Some literary critics might say this is why economists like Piketty should stick to their day job. But in truth, his argument echoes things critics themselves have said—in particular, a commonly repeated notion that 20th-century modernism turns away from the gritty details of economic life. In his 1984 book Money and Fiction, John Vernon put this especially bluntly: “Fewer novels dealing with money have been written in this century.”
Do both Piketty and literary critics have it wrong? If so, then why did writers count money more often in 1950 than in 1750? When we sorted our 8,000 novels to find ones that mentioned money especially rarely or often, we saw that references to amounts of money were related to a broader trend—the growing tendency to situate stories precisely in a familiar world, which critics call “realism.” Gothic romances like Vathek (1786) were set too far from their readers’ experience for amounts to make much difference. When the eunuch Bababalouk “showered money from both hands among the crowd,” readers didn’t need to know how much. But 19th- and 20th-century writers tried to create a believable world in part by making characters count every penny. In 1906, O. Henry opens his story “The Gift of the Magi” by having a character count out “one dollar and eighty-seven cents” three times—and notes that “sixty cents of it was in pennies.” This trick of convincing readers with fiscal specificity was shared by writers around the world, and it wouldn’t be surprising if we saw similar trends in other languages, not just in English-language fiction.
Literary critics have long acknowledged the rise of realism. But where money is concerned, they, like Piketty, have argued that its significance declined in the early 20th century. One reason is that their mental picture of the period’s literature has been shaped by a select canon of modernist novels, such as F. Scott Fitzgerald’s The Great Gatsby and Ernest Hemingway’s The Sun Also Rises. In the former novel, money becomes a metaphor to describe an increasingly consumerist and commercial society. Jay Gatsby describes Daisy’s voice as being “full of money,” a sentiment narrator Nick Carraway repeats.* Citing such novels as evidence, a conventional view of early-20th-century literature is that realism gave way to modernism, and with this shift, writers talked about money less as a literal object and more as a symbol of emotion or desire. Yet this is hardly the whole story about money and the novel in early-20th-century fiction. When we surveyed the novels from this period containing the most references to money, the top 100 contained examples of modernist fiction, but also realist, working class, immigrant, detective, and African-American fiction. It seems that readers of all kinds were interested in stories about money.
In short, when Piketty extrapolates from a few 20th-century examples to infer that money disappears from the period’s literature as a whole, he’s making much the same mistake that critics themselves have made by focusing narrowly on a small cohort of famous modernist novels. Small samples can be as misleading in literature as they are in economics.
Readers may nevertheless be right to sense that there’s something diminished about the role of money in early-20th-century fiction. But the difference is not that references to money disappear. Instead they proliferate, but at the same time shrink. The large inheritances that interest Piketty were indeed more common in early-19th-century writers like Austen and Balzac. But as we move forward through the century, novelists start to talk more often about commodity prices, or the money that characters have in their pockets. At the same time, the value of money mentioned at any given moment in a story tends to dwindle. References to money in fiction have always run the gamut from rags to riches. But if we collect a representative sample of these amounts, and use historical exchange rates to convert dollars and francs, we find that the median reference to money in a late-18th-century novel comes to about 10 pounds. That figure drops steadily through the 19th century; by the end of the century it’s less than 5 pounds. And that’s before we consider inflation. In the late 18th century, 10 pounds would have been more than half of a worker’s yearly income. By the early 20th century, the amounts of money mentioned in fiction are usually less than 5 percent of an average worker’s salary. This more-than-tenfold decline may explain why money has seemed less important to readers and scholars of early-20th-century fiction. Money appears very frequently in the period’s stories—but it’s less likely to appear as a grand estate than as a nickel dropped in a jukebox.
In some ways, this general story of a shift in emphasis from large to small is in line with Piketty’s account of economic history, which similarly traces a shift from vast sums of inherited wealth to smaller fortunes. But the arcs and rates of change are very different. On the economic side, the inequalities of capital ownership intensify across the 19th century, only to collapse suddenly after 1914, because of the inflation associated with global war. In literature, the scale of reference to money is already shrinking in the 19th century and is already shifting its emphasis from inheritance to wages and prices.
It may take some time for literary critics to fully understand the trajectory we are seeing here, and it may require more than one explanation. But one part of the story, in the 19th century at least, involves a question that Piketty ignores: Who was reading these books?
In Britain, working-class literacy doubled in the 19th century, and the average price of a book fell steeply as cheaper formats were introduced. A reading audience once drawn disproportionately from the upper and middle classes had, by the end of the century, become nearly universal, as William St Clair writes in The Reading Nation in the Romantic Period. It’s hardly surprising that this massive change in the reading audience was associated with shifts of subject matter that tended to replace gold sovereigns with ha’pennies. In this way, novels are a bit different from other kinds of social evidence: They give us less a window on the past than a picture crafted to appeal to a particular set of viewers. When the picture changes, it may tell us less about the macroeconomic landscape than about shifts in a specifically literary marketplace.
And yet literature is social evidence, as Piketty has so successfully shown. Capital in the Twenty-First Century has made novels central to public debate, both by reading individual works carefully and by integrating them in a panoramic history of inequality, built out of data gathered across two centuries. If there’s something to be learned from Piketty’s occasional slips, it’s not that we need to reinforce the wall separating literary from economic history. On the contrary, these projects have even more in common than Piketty assumed. Novels can certainly be used as close-up case studies that enliven a larger quantitative story. But they’re also part of the larger story themselves. If we approach literature with the scrupulous breadth of attention that Piketty brings to economics, we find that it too is a panoramic story about the aspirations and fears that shaped millions of human lives.
*Update, Dec. 11, 2014: This sentence has been updated to clarify that Jay Gatsby described Daisy’s voice as being “full of money” and Nick Carraway repeated the description. (Return.)