I didn't realize that Nate Silver's The Signal and the Noise was one of the best books I read last year until about three weeks after I finished it. Why am I still thinking about this book? I would think, riding the bus, going over one of his examples in my head for the twelfth time.
It's not that I'm so into baseball, or politics, or stock prices, or that I want to get better at predicting them. It's that Silver's book is an argument, and a challenge, for how I read stories in the future -- and how I write them.
Silver's core point is this:
Our predictions may be more prone to failure in the era of Big Data. As there is an exponential amount of available information, there is likewise an exponential increase in the number of hypotheses to investigate. For instance, the U.S. government now publishes data on about 45,000 economic statistics. If you want to test for a relationship between all combinations of two pairs of these statistics -- is there a causal relationship between the bank prime loan rate and the unemployment rate in Alabama? -- that gives you literally one billion hypotheses to test.
But the number of meaningful relationships in the data -- those that speak to causality rather than correlation and testify to how the world really works -- is orders of magnitude smaller. Nor is it likely to be increasing at nearly so fast a rate as the information itself; there isn't any more truth in the world that there was before the Internet or the printing press. Most of the data is just noise, as most of the universe is filled with empty space.
This is not just a problem for Big Data. We're not just surrounded by more quantitative information, more numbers, than ever before. We're also surrounded by qualitative data too. From Longreads to UpWorthy, we're have access to more stories, more characters, more anecdotes, more illustrations and examples than ever before. But just as more numbers don't inherently produce more truth, more stories don't inherently provide more lessons necessitating them.
Silver's book reminded me of one of the worst books I read last year, Malcolm Gladwell's David and Goliath: Underdogs, Misfits, and the Art of Battling Giants.
In his most talked-about chapter, Gladwell profiles superlawyer David Boies, the dude who represented Gore against Bush at the Supreme Court, gay marriage against the Mormons in California.
Boies is dyslexic. It takes him hours to read a legal brief, and he foundered in odd jobs until his mid-20s, when he went to NYU Law School, then worked his way up through government and law firms to become a legal goliath.
Boies's dyslexia, as Gladwell tells it, is a 'desirable difficulty'. Being bad at reading volumes of case law made Boies focus on being good at listening and arguing, skills other lawyers neglected. Hollywood producer Brian Grazer, also at the top of his field, also dyslexic, became a great negotiator to compensate for his difficulty reading. Goldman Sachs CEO Gary Cohn grew up dyslexic and found that it made him an outsider, gave him the skill of presenting a persona. He used that skill to blag his way into his first job in finance.
"You wouldn't wish dyslexia on your child," Gladwell ominously concludes, "Or would you?" You can almost hear the music cue: Bum bum BUMMMM.
If all this sounds a bit too easy, that's because it is. When Gladwell's book came out, critics (most prominently Christopher F. Chablis in the Wall Street Journal) pointed out that dyslexia is, rather obviously, not a magic formula for success, decidedly not something you would give to your child if you had the choice.
Gladwell says 'an extraordinarily high percentage of entrepreneurs are dyslexic.' A dyslexia researcher points out that this claim is based on a survey of 102 entrepreneurs and 37 corporate managers (out of 2,000 people contacted) and that it wasn't even designed to detect dyslexia, only dyslexia-like traits such as difficulty with spelling.
Gladwell admits that dyslexics are over-represented in the prison population; Chablis, like a high school English teacher, says 'develop this further': It's kinda sorta a major counterpoint to your argument. Gladwell points to a study where people did better on an intelligence test when it was in hard-to-read font (Lesson: difficulty makes you concentrate harder). Chablis says this test was on just 40 people, all Princeton students, and hasn't been replicated on a larger scale.
The perennial critique of Gladwell is that his conclusions do not offer any new insights, only reformulations of what we already know. This seems unfair. As Silver says, there is only so much truth in the world, only so many insights to be pointed out and illustrated. Most people go their whole lives without coming up with a profound insight into anything. Gladwell can hardly be faulted for pointing out and reformulating the insights we already know.
The more generous critique of David and Goliath is, why didn't Gladwell tell us all this himself? Why is his chapter, his book, written with this false certitude, these capitalized lessons? Boies's story would be no less interesting, no less well-told, if it was juxtaposed with the story of one of those dyslexic prisoners. I might have actually enjoyed that chapter more if it was fortified with the contradictions and arguments in the academic literature, with bright orange caveats highlighting the places where Boies's story is not typical, not indicative of something larger. Lay it on me, Gladwell, I can handle it.
Gladwell is a talented writer, a diligent researcher and interviewer, a monster intellect. If anyone could present the contradictions and paradoxes of the idea of 'desirable difficulty,' it's him. Gladwell is too smart, too curious, too skeptical, to genuinely believe that parents should be giving their kids dyslexia because it is a surefire way to end up a Hollywood produces or a finance CEO.
Hedging against the challenge of 'more information, same amount of truth,' says Silver, requires giving predictions and conclusions with confidence intervals. We're not certain that this hurricane will make landfall in Tampa, but there's a 60 percent chance. Obama is not a surefire bet to beat Romney, but he has more plausible paths to victory. These statements don't remove certainty, but they reduce it.
Gladwell's cardinal sin, to me, is not crediting his readers with enough intelligence to disclose his confidence intervals. Does he really think that the 'desirable difficulty' of dyslexia explains 100 percent of Boies's success? Probably not. If Gladwell is making the argument that dyslexia explains 5 percent of his success, or 20 percent, why not just tell us? It's as if Gladwell is trying to avoid the unfair criticism of his work -- these insights are profound, I swear! -- and in doing so steps right into the more compelling criticism. There's nothing highbrow hates more than middlebrow, and nothing says middlebrow like massaging complicated phenomena into chicken soups for the soul.
In 2012, when Jonah Lehrer was caught fabricating quotes and misrepresenting scientific findings in his book Imagine: How Creativity Works, Ta-Nehisi Coates, (one of the best working practitioners of journalistic uncertainty -- I mean that in a good way) wrote:
Great long-form journalism comes from the author's irrepressible need to answer a question. Fictional long-form journalism comes from the writer's irrepressible need to be hailed as an oracle. In the former fabulism isn't just wrong because it cheats the reader, it's wrong because it cheats the writer. Manufactured evidence tends not to satiate an aching curiosity. But it does wonders for those most interested in oraculism.
I'm not implying that Gladwell fabricated anything. His sense of curiosity is palpable in everything he writes, and a major component of what makes his best work so interesting.
But Silver's book is a justified, albeit indirect, criticism of Gladwell's approach. Silver is arguing for more curiosity and less certitude, not just for people who predict events, but for those who explain them afterwards.
Ultimately, the fault may be ours. Gladwell is under pressure from publishers, from readers, to write books of big ideas, to deliver conclusions, to expand stories into insights that make us feel like we are reaching them ourselves. Book buyers and magazine readers may not tolerate an investigation into adversity, or creativity, or decision-making that finds them too complicated for capitalized lessons, one that concludes there is nothing to conclude.
But maybe that is changing. In the same way Silver has changed what we expect from political forecasts, maybe next generation's Malcolm Gladwell will be someone who dips into subjects, guides us through contradictory evidence and leaves us with no certitude, with more questions than when we arrived. I'm ready to read that kind of journalism. I hope someone out there is ready to write it.