A decade ago this August, veteran New York City cop Joseph Gray went to a Brooklyn strip club, drank himself silly, and then got into his burgundy van and accidentally smashed into and killed three people, including a pregnant 24-year-old woman and her 4-year-old son, Andy. Twelve hours later, the undelivered child had died, too. Gray pleaded with the judge for leniency, but in her words, barreling down the street drunk, in a thousand-pound van, was like "waving a loaded gun around a crowded room." Gray was given the maximum of five to 15 years for second-degree manslaughter. The victims' community was certain it wasn't enough.
Duncan Watts, principal research scientist at Yahoo! Research and former sociology professor at Columbia University, chronicles this tale of misfortune in his new book, Everything Is Obvious: Once You Know the Answer, in order to make a simple point. "I couldn't help but think," Watts writes, "about what would have happened had Joseph Gray come along an instant later." In Watts' view, had this tragedy been averted by, say, Gray taking one last trip to the bathroom, his actions "would have been exactly as bad." Gray's case, Watts says, shows that "it doesn't follow that we should overlook the role of chance in determining outcomes. And yet we do tend to overlook it." (Never mind that the law does consider intent.)
Everything Is Obvious argues that just as the legal system may have downplayed the role of chance when punishing Gray, common sense reasoning, by focusing on actual outcomes rather than possible outcomes, makes deductions about the world that obscure just how unpredictable complex social interactions can be. Because common sense tries to explain "events only after the fact, our explanations place far too much emphasis on what actually happened relative to what might have happened." Perhaps Watts is on to something. After an event, people do find explanations that make what happened seem inevitable, without considering outcomes that were equally possible but that, by sheer chance, failed to materialize.
Leonardo da Vinci's Mona Lisa, for example, is thought to be the most beautiful painting in the world, but couldn't that be, Watts wonders, more due to historical happenstance than its intrinsic attributes?
Because most after-the-event explanations are not scientific, unlike say, medical trials with placebo control groups, Watts feels confident enough to conclude that "These days, common sense serves the same purpose as mythology." It's a bit much. "By analogy, in ancient times," he writes, "when our ancestors were startled by lightning bolts descending from the heavens, accompanied by claps of thunder, they assuaged their fears with elaborate stories about the gods." Yet despite Watts' repeated claims that common sense reasoning is flawed because people apply it after events, it's clear that his attacks on common sense have nothing at all to do with when we apply it.
In fact, Watts admits, not only do complex events often defy causal explanations, but, in a complex world, we can't predict much either. Even if we make a correct prediction, he argues, we can't be sure of why it was correct. Apple's Steve Jobs seemed like a good CEO, but how can we be sure? A stock broker may have made a string of impressive investments, but still, there's "simply not enough data to estimate" whether he's any good at predicting. History is not a randomized medical trial with a control group, Watts laments, and as one of his chapter subtitles helpfully explains, from an experimental perspective, history is badly designed: "History Is Only Run Once."
Inexplicably understating humankind's keen awareness of luck and historical fluke, Watts is intent on emphasizing the unpredictable role of chance on outcomes, and so he does emphasize it, over and over again, flinging barbs at both common sense and, as luck would have it, at Malcolm Gladwell. Common sense and Gladwell both get it in the teeth for the same reason, and why shouldn't they, if conclusions, especially about anything complicated, are so often impossible to verify. Gladwell's "catnip" speculations on influential individuals are more "perception than reality," Watts sniffs, because "there's no way to decide who is right." As evidence, he offers data from randomized Internet experiments.
Watts is at his best when exploring the Internet as a tool of sociology. Previously, it has been nearly impossible to study things like how people influence each other, because it’s extraordinarily difficult to track such interactions using randomized trial experiments. For the first time, the Internet now allows us to do exactly that, fairly cheaply and at massive scale. Watts hopes, in what in many ways should have been the true thesis of the book, that “by rendering the unmeasurable measurable, the technological revolution in mobile, Web, and Internet communications has the potential to revolutionize our understanding of ourselves and how we interact.”
In late 2009, Watts tracked an astounding 39 million Twitter chains started by 1.6 million users, counting how many times a URL was retweeted in order to identify the characteristics of influential tweeters. The purpose of the study was to see which tweets “went viral” and why. He found that 98 percent of tweets didn’t “cascade” at all, and also that at an individual level, it’s very difficult to predict whether someone who has been an influential tweeter in the past will be influential in the future. In another wonderful experiment, he ran using Amazon’s “ Mechanical Turk,” where Watts paid online workers pennies for small tasks, he found that the quality of work did not change with pay levels.
Perhaps this, Watts' third book, provides a valuable warning about the hubris of prediction, especially in complex contexts like wars. But by attacking common sense itself, he not only drastically overreaches, he distracts readers from the most valuable aspects of his argument. Watts isn't wrong—we’re all victims of happenstance, and many human interactions have been impossible to study using randomized trials. (Many remain so, too.) But had Watts chosen to focus entirely on how technology is changing and will change the way sociologists study complex social realities, the journey could have been far more rewarding. There’s a fascinating book there, to be sure, waiting to be written.
Jamie Holmes is a program associate at the New America Foundation. His writing has appeared in Foreign Policy, the Christian Science Monitor, the Philadelphia Inquirer, and the Huffington Post.