Reviewed:1 Duncan Watts, Everything Is Obvious (Once You Know The Answer), Crown Business, New York, 2011. Paperback, 269 pp. plus bibliography, notes and index.
Lest we be paralyzed with indecision, we confidently go about our everyday lives making assumptions, reasoning from the few to the many, seeking regularity and choosing, optimistically, to interpret the outcomes of our decisions as confirmation that our haphazard approach was the best one. What's more, we do the same—and eagerly accept arguments developed in the same way—when we aim to digest current events and important societal trends.
'Twas ever thus, from childhood's hour. In Everything Is Obvious (Once You Know The Answer), Duncan Watts—sociologist, and researcher at first Yahoo!, now Microsoft—walks readers fondly through the decay of this confidence in common-sense explanations of phenomena from politics, history, economics, culture and policy. A litany of cognitive biases, logical fallacies, unpredictable emergent behaviours and irreducible complexities are invoked to show how attractively simple explanations are, quite often, dead wrong. Making use of results from social science research, psychology, network theory and management literature—both famous and not, new and old—Everything is Obvious lays out the limits of prediction, planning and causal reasoning in a complex world.
Watts' attention ranges widely, from the philosophy of Arthur Danto, to Steven Levitt and Stephen Dubner's Freakonomics and back to John Rawls. Tolstoy and Harry Potter; the Mars Climate Orbiter and Twitter; Enron and AIG; Steve Jobs, Jane Jacobs, Newton and Paul Revere all make appearances. But though his cautionary messages may seem discouraging, Watts' style is welcoming and accessible. His explanations of the many obstacles to social science are developed in simple language, moving easily from familiar examples to general concepts and their implications. In the numerous endnotes are acknowledgements of the larger rhetorical leaps and, for the curious reader or academic, extra implications, caveats and references to an extensive bibliography. Later chapters reach back to earlier concepts, authors and references to tie the work together, and the author's several descriptions of his own research fit neatly into the flow of the book.
Lest he be mistaken for other authors bridging the gap between social science and the under-appreciated complexity of everyday affairs, Watts reserves his most direct criticism for Malcolm Gladwell's The Tipping Point, and “gadfly” statistician Nassim Nicholas Taleb's The Black Swan. Gladwell's “law of the few” (suggesting that a few ‘super-influencers’ can cause the viral takeoff of a trend) is debunked, in a chapter on “Special People,” as a hopeful fable that ignores the structure of real-world social networks and ultimately proves a post-hoc fallacy. Black swans—rare, highly unlikely but equally consequential events—are only visible from the far side of the paradigm shifts they trigger, and even then vanish into their context on close inspection or suffer from imprecise labelling. Everything is hard, it seems, and even that which is deceptively obvious may deceive for one of many reasons.
In the latter half, Watts moves to outline what (little) can be done, in light of the challenges laid out—and it's here that his enthusiasm for the opportunities of information technology gets the better of his methodical approach and the care that marks the opening chapters. A chapter on “The Measure of All Things” focuses heavily on the efforts of Internet advertisers to act on the voluminous user data from website visitors. Even if this content is useful for improving the same mechanism that generated it, its usefulness beyond advertising is not argued. Watts acknowledges, but does not thoroughly deal with, the objection that the “friendships” and “likes” of social media users may bear little or irregular relationship to social interactions in the offline world—and he omits that the companies creating these services may profit from creating and eliding such differences.
The chapter on “Fairness and Justice” treats the example of a New York City police officer convicted of drunk driving, but fails to examine the fairness and justice of the same force's stop-and-frisk policy—surely a more consequential topic. Similarly, after dispelling “The Dream of Prediction,” Watts advises that readers “don't predict; react,” or at least not aim to do more than to “predict the present.” Yet on uncertain, global issues such as overpopulation2 or the “super wicked” problem3 of climate change, the luxury to retreat into reaction may not exist; or even if it does, could entail abandoning whole categories of effective and just solutions.
Still, in dealing with these problems—as in daily life—we will need to understand the surprising behaviours of crowds, networks, and ourselves, and Watts' well-written book serves well as a spotter's guide to the inevitable missteps.
for ESD.83, the Engineering Systems doctoral seminar, Fall 2014. ↩
Levin, Cashore, Bernstein, and Auld. “Overcoming the tragedy of super wicked problems: constraining our future selves to ameliorate global climate change,” in Policy Sciences 45:123–152, 2012. 10.1007/s11077-012-9151-0. ↩