Does Misery in Literature Really Follow Miserable Economics?

Heads up, we're about four years out from some pretty depressing books, at least according to new research that links economic misery to misery in literature. According to a study published in PLOS One, a team of British researchers analyzed millions of books according to a "literary misery index" that subtracted the frequency of words associated with joy from those associated with sadness, and they found that spikes in the literary misery index were correlated with spikes in the economic misery index—a measurement of unemployment and inflation.

But literature lagged behind economics by an average of about 11 years, leading the researchers to conclude that the emotional mood of literature was reflecting the mood of the economy over the previous decade. So by 2018, we'll be reaping the literary fruits of The Great Recession.

Finding all of this very difficult to understand, I read The New York Times article on the study, which gave The Grapes of Wrath as an example. It was published in 1939, and therefore reflects the economic mood from 1929-1939, the Great Depression. I think I'd use the words "great" and "depressing" to describe Grapes of Wrath, so I guess that one passes.

The actual researchers didn't analyze any books specifically, they just filtered everything that Google scanned through their literary misery index, but something seems amiss when I test other examples.

I pulled up the economic misery index for the second half of the 20th century and looked for spikes in misery. The George H.W. Bush years are pretty rough, but the highest years under his bespectacled watch are still lower than the least miserable years in the decade from 1974 through 1984, which includes peak misery, in 1980.

So 1984 should have some real downers right? And it sort of does: There was a sequel to The Godfather by Mario Puzo, the Dr. Seuss that came out that year was a metaphor for war, Gore Vidal published a book about Lincoln, and we all know how that ended.

But to honest, it doesn't seem that much darker any other bestsellers list. The top-selling Stephen King book wasn't even a horror novel, it was the fantastical The Talisman, which might start with someone's mother dying from cancer, but by Stephen King standards is still pretty upbeat. The nonfiction list had a book by John Madden and a book called Moses the Kitten, in addition to your standard CEO biographies (Leo Iacocca, in this case) and books about motherhood. Shel Silverstein's A Light in the Attic was in there too.

Hunting for sadness on the bestseller list didn't seem to be working so I tried to think of the most depressing book I could think of. Sophie's Choice came to mind, and it was published in the infamous year of 1980, which meant it was missing out on some primo economic misery to follow in the next few years. It's pretty hard to get bleaker than Cormac McCarthy's The Road: Post-apocalyptic, cannibalistic, a stone cold bummer to the end. Published in 2006—so all of that misery would economically describe 1996-2006. So the prosperous and gay 1990s, and only the earliest hints of the coming economic spiral that America still hasn't really emerged from. Those were some of America's least economically miserable years since the Eisenhower administration. It just doesn't seem to follow.

If there is big data applications in literature, this study might not be the best example. The Times sort of hits on how the literary misery index might not be the best metric of mood:

For one thing, the lists of emotion words, created by other researchers and used for years, include some surprising choices: words like "smug" and "wallow" were among 224 words on the "joy" list; potentially neutral adjectives like "dark" and "low" were among 115 words on the "sadness" list.

There may also be the problem of looking at bestsellers, especially as they're often dominated by legacy acts—your Danielle Steeles, James Pattersons, and so on. It actually seems like these books don't really change that much from year to year—it's the same mysteries and courtroom drama until Harry Potters finally start showing up. And obviously just cherry-picking one depressing story doesn't tell you much—obviously this isn't a rigorous methodological critique of this study. Still, oversimplifying one thing—the economy—and oversimplifying another—the emotional content of literature—and then comparing them seems like a pretty shaking way to start your study.

It is, however, a question we'll have to face more and more in era of "big data": Every now and then something's going to come back that's in the numbers that just doesn't feel correct. And we'll be left to determine whether this is a flaw in perception or in the data collection.

And that unsureness, that human misgiving and distrust, the individual's ego versus the mechanism, why, isn't that what literature explores best?