By Michael Washburn
Sunday, August 1, 2010; B01
Adventures in the Margin of Error
By Kathryn Schulz
Ecco. 405 pp. $26.99
Why Experts Keep Failing Us -- and How to Know When Not to Trust Them
By David H. Freedman
Little, Brown. 295 pp. $25.99
Error arrives cloaked in certainty. In our politics, in our relationships and in the advice we solicit, we're at the mercy of an ever-present unreliability. Such are the lessons taught by Kathryn Schulz's "Being Wrong" and David H. Freedman's "Wrong," complementary explorations of our relentless genius for getting it . . . wrong.
The good news, from Schulz's perspective, is that mistakes shouldn't be condemned, at least not in any traditional sense. Schulz draws on philosophers, neuroscientists, psychoanalysts and a bit of common sense in an erudite, playful rumination on error. "We are wrong about what it means to be wrong," she writes. "Far from being a sign of intellectual inferiority, the capacity to err is crucial to human cognition . . . [and] it is inextricable from some of our most humane and honorable qualities: empathy, optimism, imagination, conviction, and courage." By understanding the dynamics of error, we open a space for tolerance of both our own and others' failings. This is error as pedagogy, and it does not come naturally.
Being wrong, Schulz notes, feels exactly like being right. We often fall victim to the "cuz it's true" dynamic: the self-serving circularity of taking one's own belief in an idea as a sign of that idea's veracity. We engage in this thinking all the time, about everything: We decide what flavor ice cream is best or what the long-term prospects of the "tea party" are, and we promote these beliefs, often fiercely. Humans are engines of strong, often ill-formed or arbitrary opinion, and any assertion of one's knowledge carries with it an accusation of others' errors.
We judge others' mistakes more harshly than our own, often assuming that others' errors stem from one of three flaws: idiocy, ignorance or evil. As Schulz writes, "Moral and intellectual wrongness are connected not by mere linguistic coincidence but by a long history of associating error with evil -- and, conversely, rightness with righteousness."
"Being Wrong" traverses disciplines and eras, deftly interweaving etymology (Schulz reminds us that "error" derives from the Latin word for "to stray or wander") with such sources as Saint Augustine's "Confessions," contemporary neuroscience and vivid examples of radical mistakes. For instance, to illuminate the resilience of error, she tells the disquieting tale of a woman suffering from anosognosia, the unawareness or denial of a disability: The woman remained convinced that she could see despite having recently gone totally blind, going so far as to describe in vivid yet absolutely incorrect detail her hospital room.
Through such cases, Schulz lays bare the inductive failures, misperceptions and biased assumptions that exist in less extreme form in everyone and that everyone should find instructive.
In "Wrong," Freedman takes a darker view of our errors, and while his book is less artfully written than Schulz's, it is more forcefully argued, focusing on the point where error shades into deceit.
Why do experts fail? Political pundits and business writers are paid to give opinions, not to be right. And as Freedman points out, despite life's complexity, we prefer the simple advice proffered by informal experts, particularly when it reinforces our preexisting ideas or affirms our hopes (sleep your way to a six-pack!). The startling lesson of "Wrong," however, is how often medical researchers engage in similar crowd-pleasing charlatanism.
"If," Freedman writes, "a scientist wants to or expects to end up with certain results, he will likely achieve them, often through some form of fudging, whether conscious or not." He supports this assertion with a torrent of research -- experts on experts -- hoping to disenthrall readers from the seduction of scientific expertise. He notes this irony and devotes an appendix to explaining why his sources should prove more reliable: "Experts who study other experts' failings are better equipped . . . to avoid those troubles." On balance, Freedman is convincing, even if his explanation sometimes exhibits a bit of "cuz it's true."
"Wrong" makes a powerful case for the prevalence of scientific ineptitude. For example, data are often disregarded if they contradict the results that were predicted in research proposals written to secure funding. Moreover, most studies honored with publication benefit from journals' preference for provocative, positive findings. This may, at first, seem reasonable, until Freedman points out that when multiple studies address the same question, the ones with positive results are much more likely to see print, even if the majority present negative findings. In this sense, prestigious scientific journals are no better than tabloids: They seek out attention-grabbing headlines, and researchers are eager to oblige.
Brazen fraud also plays a role in research. Among other examples, Freedman explains how an oncologist faked the results of a widely praised genetics experiment by coloring white mice with a Magic Marker.
Mice themselves are a problem, too. Researchers' reliance on different species, subjected to extreme conditions, to shed light on human ailments falls under Freedman's withering gaze. For instance, anyone who sees a correlation between human depression and the length of time a lab rat swims before giving up and floating is probably fooling himself.
But Freedman's lazy scientists are Schulz's self-deceived strivers. Experts are only human and, therefore, unaware of many of their mistakes, even if they profit from our culture of credulity. Despite this rash of errors, we remain in denial over our own fallibility -- we're our own worst experts. In her discussion of medical expertise, Schulz writes, "What is both philosophically and practically interesting about [the question of medical error] is the paradox that lurks at its heart: if you want to try to eradicate error, you have to start by assuming that it is inevitable."
Perhaps this quasi-philosophical first premise should supplement Descartes's cogito: Not only I think, therefore I am, but I'm certain, therefore I'm often wrong.
Michael Washburn is the assistant director of the Center for the Humanities at the Graduate Center of the City University of New York.