Posner Reviews Blink
Blink: The Power of Thinking Without Thinking
By Malcolm Gladwell
(Little, Brown, 277 pp., $25.95)
There are two types of thinking, to oversimplify grossly. We may call them intuitive and articulate. The first is the domain of hunches, snap judgments, emotional reactions, and first impressions--in short, instant responses to sensations. Obviously there is a cognitive process involved in such mental processes; one is responding to information. But there is no conscious thought, because there is no time for it. The second type of thinking is the domain of logic, deliberation, reasoned discussion, and scientific method. Here thinking is conscious: it occurs in words or sentences or symbols or concepts or formulas, and so it takes time. Articulate thinking is the model of rationality, while intuitive thinking is often seen as primitive, "emotional" in a derogatory sense, the only type of thinking of which animals are capable; and so it is articulate thinking that distinguishes human beings from the "lower" animals.
When, many years ago, a judge confessed that his decisions were based largely on hunch, this caused a bit of a scandal; but there is increasing recognition that while judicial opinions, in which the judge explains his decision, are models of articulate thinking, the decision itself--the outcome, the winner--will often come to the judge in a flash. But finally the contrast between intuitive and articulate thinking is overdrawn: it ignores the fact that deliberative procedures can become unconscious simply by becoming habitual, without thereby being intuitive in the sense of pre-verbal or emotional; and that might be the case with judicial decisions, too.
Malcolm Gladwell, a journalist, wishes to bring to a popular audience the results of recent research in psychology and related disciplines, such as neuroscience, which not only confirm the importance of intuitive cognition in human beings but also offer a qualified vindication of it. He argues that intuition is often superior to articulate thinking. It often misleads, to be sure; but with an awareness of the pitfalls we may be able to avoid them.
As Exhibit A for the superiority of intuitive to articulate thinking, Gladwell offers the case of a purported ancient Greek statue that was offered to the Getty Museum for $10 million. Months of careful study by a geologist (to determine the age of the statue) and by the museum's lawyers (to trace the statue's provenance) convinced the museum that it was genuine. But when historians of ancient art looked at it, they experienced an "intuitive revulsion," and indeed it was eventually proved to be a fake.
The example is actually a bad one for Gladwell's point, though it is a good illustration of the weakness of this book, which is a series of loosely connected anecdotes, rich in "human interest" particulars but poor in analysis. There is irony in the book's blizzard of anecdotal details. One of Gladwell's themes is that clear thinking can be overwhelmed by irrelevant information, but he revels in the irrelevant. An anecdote about food tasters begins: "One bright summer day, I had lunch with two women who run a company in New Jersey called Sensory Spectrum." The weather, the season, and the state are all irrelevant. And likewise that hospital chairman Brendan Reilly "is a tall man with a runner's slender build." Or that "inside, JFCOM [Joint Forces Command] looks like a very ordinary office building.... The business of JFCOM, however, is anything but ordinary." These are typical examples of Gladwell's style, which is bland and padded with cliches.
But back to the case of the Greek statue. It illustrates not the difference between intuitive thinking and articulate thinking, but different articulate methods of determining the authenticity of a work of art. One method is to trace the chain of title, ideally back to the artist himself (impossible in this case); another is to perform chemical tests on the material of the work; and a third is to compare the appearance of the work to that of works of art known to be authentic. The fact that the first two methods happened to take longer in the particular case of the Getty statue is happenstance. Had the seller produced a bill of sale from Phidias to Cleopatra, or the chemist noticed that the statue was made out of plastic rather than marble, the fake would have been detected in the blink of an eye. Conversely, had the statue looked more like authentic statues of its type, the art historians might have had to conduct a painstakingly detailed comparison of each feature of the work with the corresponding features of authentic works. Thus the speed with which the historians spotted this particular fake is irrelevant to Gladwell's thesis. Practice may not make perfect, but it enables an experienced person to arrive at conclusions more quickly than a neophyte. The expert's snap judgment is the result of a deliberative process made unconscious through habituation.
As one moves from anecdote to anecdote, the reader of Blink quickly realizes, though its author does not, that a variety of interestingly different mental operations are being crammed unhelpfully into the "rapid cognition" pigeonhole. In one anecdote, Dr. Lee Goldman discovers that the most reliable quick way of determining whether a patient admitted to a hospital with chest pains is about to have a heart attack is by using an algorithm based on just four data: the results of the patient's electrocardiogram, the pain being unstable angina, the presence of fluid in the lungs, and systolic blood pressure below one hundred. There is no diagnostic gain, Goldman found, from also knowing whether the patient has the traditional risk factors for heart disease, such as being a smoker or suffering from diabetes. In fact, there is a diagnostic loss, because an admitting doctor who gave weight to these factors (which are indeed good long-term predictors of heart disease) would be unlikely to admit a patient who had none of the traditional risk factors but was predicted by the algorithm to be about to have a heart attack.
To illustrate where rapid cognition can go wrong, Gladwell introduces us to Bob Golomb, an auto salesman who attributes his success to the fact that "he tries never to judge anyone on the basis of his or her appearance." More unwitting irony here, for Gladwell himself is preoccupied with people's appearances. Think of Reilly, with his runner's build; or John Gottman, who claims to be able by listening to a married couple talk for fifteen minutes to determine with almost 90 percent accuracy whether they will still be married in fifteen years, and whom Gladwell superfluously describes as "a middle-aged man with owl-like eyes, silvery hair, and a neatly trimmed beard. He is short and very charming...." And then there is "Klin, who bears a striking resemblance to the actor Martin Short, is half Israeli and half Brazilian, and he speaks with an understandably peculiar accent." Sheer clutter.
Golomb, the successful auto salesman, is contrasted with the salesmen in a study in which black and white men and women, carefully selected to be similar in every aspect except race and sex, pretended to shop for cars. The blacks were quoted higher prices than the whites, and the women higher prices than the men. Gladwell interprets this to mean that the salesmen lost out on good deals by judging people on the basis of their appearance. But the study shows no such thing. The authors of the study did not say, and Gladwell does not show, and Golomb did not suggest, that auto salesmen are incorrect in believing that blacks and women are less experienced or assiduous or pertinacious car shoppers than white males and therefore can be induced to pay higher prices. The Golomb story contained no mention of race or sex. (Flemington, where Golomb works, is a small town in central New Jersey that is only 3 percent black.) And when he said he tries not to judge a person on the basis of the person's appearance, it seems that all he meant was that shabbily dressed and otherwise unprepossessing shoppers are often serious about buying a car. "Now, if you saw this man [a farmer], with his coveralls and his cow dung, you'd figure he was not a worthy customer. But in fact, as we say in the trade, he's all cashed up."
It would not occur to Gladwell, a good liberal, that an auto salesman's discriminating on the basis of race or sex might be a rational form of the "rapid cognition" that he admires. If two groups happen to differ on average, even though there is considerable overlap between the groups, it may be sensible to ascribe the group's average characteristics to each member of the group, even though one knows that many members deviate from the average. An individual's characteristics may be difficult to determine in a brief encounter, and a salesman cannot afford to waste his time in a protracted one, and so he may quote a high price to every black shopper even though he knows that some blacks are just as shrewd and experienced car shoppers as the average white, or more so. Economists use the term "statistical discrimination" to describe this behavior. It is a better label than stereotyping for what is going on in the auto-dealer case, because it is more precise and lacks the distracting negative connotation of stereotype, defined by Gladwell as "a rigid and unyielding system." But is it? Think of how stereotypes of professional women, Asians, and homosexuals have changed in recent years. Statistical discrimination erodes as the average characteristics of different groups converge.
Gladwell reports an experiment in which some students are told before a test to think about professors and other students are told to think about soccer hooligans, and the first group does better on the test. He thinks this result shows the fallacy of stereotypical thinking. The experimenter claimed it showed that people are so suggestible that they can be put in a frame of mind in which they feel smarter and therefore perform smarter. The claim is undermined by a literature of which Gladwell seems unaware, which finds that self-esteem is correlated negatively rather than positively with academic performance. Yet, true or false, the claim is unrelated to statistical discrimination, which is a matter of basing judgments on partial information.
The average male CEO of a Fortune 500 company is significantly taller than the average American male, and Gladwell offers this as another example of stereotypical thinking. That is not very plausible; a CEO is selected only after a careful search to determine the candidate's individual characteristics. Gladwell ignores the possibility that tall men are disproportionately selected for leadership positions because of personality characteristics that are correlated with height, notably self-confidence and a sense of superiority perhaps derived from experiences in childhood, when tall boys lord it over short ones. Height might be a tiebreaker, but it would be unlikely to land the job for a candidate whom an elaborate search process revealed to be less qualified than a shorter candidate.
Gladwell applauds the rule that a police officer who stops a car driven by someone thought to be armed should approach the seated driver from the rear on the driver's side but pause before he reaches the driver, so that he will be standing slightly behind where the driver is sitting. The driver, if he wants to shoot the officer, will have to twist around in his seat, and this will give the officer more time to react. Gladwell says that this rule is designed to prevent what he calls "temporary autism." This is one of many cutesy phrases and business-guru slogans in which this book abounds. Others include "mind-blindness," "listening with your eyes," "thin slicing"--which means basing a decision on a small amount of the available information--and the "Warren Harding error," which is thinking that someone who looks presidential must have the qualities of a good president.
Autistic people treat people as inanimate objects rather than as thinking beings like themselves, and as a result they have trouble predicting behavior. Gladwell argues that a police officer who fears that his life is in danger will be unable to read the suspect's face and gestures for reliable clues to intentions (Gladwell calls this "mind reading") and is therefore likely to make a mistake; he is "mind-blinded," as if he were autistic. The rule gives him more time to decide what the suspect's intentions are. It seems a sensible rule, but the assessment of it gains nothing from a reference to autism. Obviously you are less likely to shoot a person in mistaken self-defense the more time you have in which to assess his intentions.
Gladwell endorses a claim by the psychologist Paul Ekman that careful study of a person's face while he is speaking will reveal unerringly whether he is lying. Were this true, the implications would be revolutionary. The CIA could discard its lie detectors. Psychologists trained by Ekman could be hired to study videotapes of courtroom testimony and advise judges and jurors whom to believe and whom to convict of perjury. Ekman's "Facial Action Coding System" would dominate the trial process. Gladwell is completely credulous about Ekman's claims. Ekman told him that he studied Bill Clinton's facial expressions during the 1992 campaign and told his wife, "This is a guy who wants to be caught with his hand in the cookie jar, and have us love him for it anyway." This self-serving testimony is no evidence of anything. The natural follow-up question for Gladwell to have asked would have been whether, when Ekman saw the videotape of Clinton's deposition during the run-up to his impeachment, he realized that Clinton was lying. He didn't ask that question. Nor does he mention the flaws that critics have found in Ekman's work.
As with Gladwell's other tales, the Ekman story is not actually about the strengths and the weaknesses of rapid cognition. It took Ekman years to construct his Facial Action Coding System, which Gladwell tells us fills a five-hundred-page binder. Now, it is perfectly true that we can often infer a person's feelings, intentions, and other mental dispositions from a glance at his face. But people are as skillful at concealing their feelings and intentions as they are at reading them in others--hence the need for the FACS, which is itself a product of articulate thinking.
So Gladwell should not have been surprised by the results of an experiment to test alternative methods of discovering certain personal characteristics of college kids, such as emotional stability. One method was to ask the person's best friends; another was to ask strangers to peek inside the person's room. The latter method proved superior. People conceal as well as reveal themselves in their interactions with their friends. In arranging their rooms, they are less likely to be trying to make an impression, so the stranger will not be fooled by prior interactions with the person whose room it is. The better method happened to be the quicker one. But it wasn't better because it was quicker.
Remember JFCOM? In 2002, it conducted a war game called "Millennium Challenge" in anticipation of the U.S. invasion of Iraq. As commander of the "Red Team" (the adversary in a war game), JFCOM chose a retired Marine general named Paul Van Riper. Oddly, Gladwell never mentions that Van Riper was a general. This omission, I think, is owed to Gladwell's practice of presenting everyone who gets the psychology right as an enemy of the establishment, and it is hard to think of a general in that light, though in fact Van Riper is something of a maverick.
The Blue Team was equipped with an elaborate computerized decision-making tool called "Operational Net Assessment." Van Riper beat the Blue Team in the war game using low-tech, commonsense tactics: when the Blue Team knocked out the Red Team's electronic communications, for example, he used couriers on motorcycles to deliver messages. Was Van Riper's strategy a triumph of rapid cognition, as Gladwell portrays it? Operational Net Assessment was and is an experimental program for integrating military intelligence from all sources in order to dispel the "fog of war." The military is continuing to work on it. That Van Riper beat it two years ago is no more surprising than that chess champions easily beat the earliest chess-playing computers: today, in a triumph of articulate "thinking" over intuition, it is the computers that are the champs.
Gladwell also discusses alternative approaches in dating. (The procession of his anecdotes here becomes dizzying.) One is to make a list of the characteristics one desires in a date and then go looking for possible dates that fit the characteristics. The other, which experiments reveal, plausibly, to be superior, is to date a variety of people until you find someone with whom you click. The distinction is not between articulate thinking and intuitive thinking, but between deduction and induction. If you have never dated, you will not have a good idea of what you are looking for. As you date, you will acquire a better idea, and eventually you will be able to construct a useful checklist of characteristics. So this is yet another little tale that doesn't fit the ostensible subject of his book. Gladwell does not discuss "love at first sight," which would be a good illustration of the unreliability of rapid cognition.
In discussing racial discrimination, Gladwell distinguishes between "unconscious attitudes" and "conscious attitudes. That is what we choose to believe." But beliefs are not chosen. You might think it very nice to believe in the immortality of the soul, but you could not will yourself (at least if you are intellectually honest) to believe it. Elsewhere he remarks of someone that when he is excited "his eyes light up and open even wider." But eyes don't light up; it is only by opening them wider that one conveys a sense of excitement. The metaphor of eyes lighting up is harmless, but one is surprised to find it being used by a writer who is at pains to explain exactly how we read intentions in facial expressions--and it is not by observing ocular flashes.
This book also succumbs to the fallacy that people with good ideas must be good people. Everyone in the book who gets psychology right is not only or mainly a bright person, he is also a noble human being; so there is much emphasis, Kerry-like, on Van Riper's combat performance in the Vietnam War, without explicitly mentioning that he went on to become a lieutenant-general. Such pratfalls, together with the inaptness of the stories that constitute the entirety of the book, make me wonder how far Gladwell has actually delved into the literatures that bear on his subject, which is not a new one. These include a philosophical literature illustrated by the work of Michael Polanyi on tacit knowledge and on "know how" versus "know that"; a psychological literature on cognitive capabilities and distortions; a literature in both philosophy and psychology that explores the cognitive role of the emotions; a literature in evolutionary biology that relates some of these distortions to conditions in the "ancestral environment" (the environment in which the human brain reached approximately its current level of development); a psychiatric literature on autism and other cognitive disturbances; an economic literature on the costs of acquiring and absorbing information; a literature at the intersection of philosophy, statistics, and economics that explores the rationality of basing decisions on subjective estimates of probability (Bayes's Theorem); and a literature in neuroscience that relates cognitive and emotional states to specific parts of and neuronal activities in the brain.
Taken together, these literatures demonstrate the importance of unconscious cognition, but their findings are obscured rather than elucidated by Gladwell's parade of poorly understood yarns. He wants to tell stories rather than to analyze a phenomenon. He tells them well enough, if you can stand the style. (Blink is written like a book intended for people who do not read books.) And there are interesting and even illuminating facts scattered here and there, such as the blindfold "sip" test that led Coca-Cola into the disastrous error of changing the formula for Coke so that it would taste more like Pepsi. As Gladwell explains, people do not decide what food or beverage to buy solely on the basis of taste, let alone taste in the artificial setting of a blindfold test; the taste of a food or a drink is influenced by its visual properties. So that was a case in which less information really was less, and not more. And of course he is right that we may drown in information, so that to know less about a situation may sometimes be to know more about it. It is a lesson he should have taken to heart.
Richard A. Posner is a judge of the United States Court of Appeals for the Seventh Circuit and a senior lecturer at the University of Chicago Law School.