by Sam Sommers
In reality, there’s so much more to discrimination than hate. To assume otherwise—to pretend that as long as our hearts are in the right place, we’re off the hook—is just another misconception driven by WYSIWYG and the desire to see the self in a positive light. This chapter offers a more nuanced exploration of intergroup bias, exploring this darker side of humanity without relying on the bad apple excuse. Because it turns out that we all engage in a number of seemingly innocuous tendencies that, in the end, contribute to group stereotypes and disparities.
For starters, there’s the impact of “us,” the generosity of spirit and resources that we offer to members of in-groups but not out-groups. As with me and the elevator barger.
Or the voter able to rationalize away the lack of experience of her nominee but convinced that the opposition candidate’s thin résumé is a fatal flaw.
Or the baseball fan willing to give the benefit of the doubt to the hometown player accused of steroid use but ready to pounce on comparable rumors about other teams.
Or the police officer who sees a bit of his younger self in a juvenile offender and decides to give a second chance to a “good kid from a good family” but throws the book at a less fortunate, less familiar teenager from the other side of town.
No, you don’t have to believe in some widespread racist conspiracy to explain obvious disparities such as those in death penalty cases. Most prosecutors and jurors are well-intentioned individuals working hard to uphold their oaths. But crime seems just a bit more surprising and reprehensible when it happens to someone on the “right” side of the tracks. Jurors feel just a little more outrage and anger when the victim reminds them of someone in their own family or neighborhood.
Sometimes our affinity for us is just as problematic as our dislike of them. Even when you’re convinced that you don’t have a hateful bone in your body, the way we see the world can fuel the fires of conflict and inequity.
SEEING SPOTS
Animosity between some groups is understandable, whether because of historical tension, competition for scarce resources, or incompatible ideologies. Like Democrats and Republicans. Red Sox fans and Yankees fans. Mac users and PC users. Those of us who back into parallel parking spots and the mouth-breathing Neanderthals who barrel in front-first. These are irreconcilable differences.
But something about the simple context of groupness also skews thought and action. That is, even without the particular baggage associated with any of the oppositional pairs above, the mere experience of being split into separate groups in and of itself promotes self-interest and conflict.
How little does it take to nudge someone into an us-versus-them mentality? I’ll show you. On the following page is a cluster of dots. Take a quick look and, within a second or two, give your best estimate of how many there are. No complex geometrical analyses required—just a hair-trigger, first-glance impression. In the spirit of heading off any potential gender differences, I will assure you ahead of time that men and women do not show performance differences on this dot task. Dalmatian owners, on the other hand, should have a built-in advantage.
Ready? Then turn the page and settle on your estimate before reading further.
There are 407 dots in this cluster. I know because I copied and pasted each one myself, having learned the hard way that finding a non-copyrighted image of dots is not as easy as you’d think. After 407, the carpal tunnel kicked in.
How did your guess stack up? Are you a dot overestimator or underestimator? Or do you claim to have gotten it exactly right, in which case you’re more of a dot prevaricator?
Actually, whether your guess was too high or too low isn’t important. What is important is what I tell you about your performance. Were this a research study of intergroup relations, I would have shown you a series of clusters like this one and then, at random, informed you that you’re either a chronic dot overestimator or a chronic dot underestimator. That’s all it would take to push you into the realm of us versus them.
In one such study, a team of British and French researchers led by Henri Tajfel used this dot task to place school-age boys into one of two groups.4 They showed forty different clusters of dots, each for less than half of one second. Again, it didn’t matter if the boys actually guessed too high, too low, or haphazardly—given the outside possibility that overstimators and underestimators really do differ on some dimension, the researchers took matters into their own hands. At random, each boy was told that he was either an overestimator or an underestimator and was then grouped with fellow students of the same supposed tendency.
Having completed the dot portion of the program, the students were told that they’d be moving to an unrelated judgment task. Led to individual cubicles, they were given a series of numerical charts to use for allocating financial rewards between two classmates. For example, one chart presented options for divvying up fifteen cents between Student A and Student B. Another forced them to choose an amount of money for one student to receive and a separate amount for the other to lose. Each decision they made translated into real financial gains or losses for specific classmates at the end of the study.
Ostensibly a measure of their ability to make sense of the numerical charts, the task was anonymous: the boys never knew the true identity of their Student A or Student B. But they did know group membership. They did know whether Student A was, for example, a fellow member of their recently created group of dot overestimators or belonged to the other underestimator group.
Even though group assignments had been determined minutes earlier—and at random, no less—affiliation had a dramatic impact on the students’ allocation decisions. Boys who had been told they were dot overestimators made financial choices that favored fellow overestimators 69 percent of the time. Underestimators showed an even more egregious bias for some reason, clocking in with a 94 percent rate of favoritism.
Such increased generosity toward the in-group transcends the financial. Years later, American researchers used Tajfel’s dot task to examine how group affiliation affects college students’ social expectation and memory.5 After estimating dots, participants were given two decks of index cards. Each card listed a behavior that had supposedly been disclosed by a previous respondent. Some were positive, such as “I took two disadvantaged kids on a vacation.” Others were negative: “I had two affairs while married.”
Both decks included the same number of positive and negative behaviors, and the specific cards in each deck were rotated at random. So the only meaningful difference between them was that participants believed that one deck listed behaviors previously written by dot overestimators, while the other deck supposedly came from underestimators.
After browsing through the cards, participants were given a surprise memory test. When it came to memory for positive behaviors, they did an equally good job with in-group and out-group members, correctly identifying just under 70 percent of the positive cards from each deck. For negative behaviors, however, a telling divergence emerged: Respondents correctly remembered only 57 percent of the negative cards from fellow overestimators or underestimators. But their memory was much more accurate for the peccadilloes of the group they weren’t in—82 percent accurate, in fact.
In other words, the bad behaviors of people like us are not only more forgivable, they’re also more forgettable. These are benefits we don’t extend to outsiders, whom we hold to higher, less attainable standards. This tendency is deep-seated enough that it arises even when insider/outsider status is determined by something as trivial as how we see spots. Just imagine how strong these biases become when the two groups in question have a history of bad blood.
CATEGORICALLY SPEAKING
Why this propensity to see the in-group in a positive light? Some of it owes to our more general generosity in how we view the self. The same rose-colored glasses you use to evaluate your own characteristics and behaviors also filter how you see the groups with which you affiliate. Just as the effort to feel good about the self some
times points you toward downward comparisons with individuals worse off than you are, so does derogating other groups help restore more collective pride and self-esteem.
The other lesson of the dot studies is just how dependent we are on categories. There are too few hours in the day and too many people in the world to permit careful, thoughtful scrutiny of everyone we meet. We have only so much mental energy at our disposal, and categorization is one way to cut cognitive corners.
For a moment, forget about how we perceive people—just think about objects. Early in life we start constructing categories based on the salient features of things in our environment: Furniture. Vegetables. Birds. Trucks. By using categories like these, we save time and effort. Even if you can’t articulate a precise definition of “furniture,” much like the Supreme Court and obscenity, you know it when you see it. This ability to relate the new to the familiar makes it that much easier to deal with novel items in your environment, freeing up cognitive resources for more important pursuits.
But our reliance on categories leads to problematic side effects as well. One is an exaggerated sense of how different groups of objects are. Like vegetables. The U.S. Department of Agriculture groups fruits and vegetables on the same level of its food pyramid, suggesting that we eat five to nine daily servings. (This, in addition to six to eleven servings of grain, three of meat, and three of dairy, which begs the question of whether they also recommend that we sit down to eight meals a day.) Still, despite the USDA’s willingness to group them together, if I so much as include the words fruit and vegetable in the same sentence with my younger daughter, she scoffs at my galling naïveté. To her and most of her preschool cronies, fruit is delicious but veggies are the scourge of the earth.
Indeed, I once made the egregious error of trying to reason with her, explaining that tomatoes—which she won’t eat—are actually fruits, whereas strawberries—which she will—are more vegetablish. She yelled at me. And then started to cry. She only stopped when I assured her that I was kidding, just like the time I had announced in jest that we’d be skipping our long-awaited trip to the indoor water park to go to Costco instead. Well, we adults also prefer to keep a healthy mental distance between competing object categories, even if suggestions to the contrary don’t usually move us to tears.
A second side effect of categorization is that it leads us to gloss over differences that exist within groups. Like birds. If asked, most of us have little trouble generating examples: robin, sparrow, crow, blue jay, pigeon. But the category “bird” is actually far more diverse than we give it credit for, including as it does the ostrich, owl, turkey, pelican, and penguin. Just as different categories aren’t as different as we assume them to be, objects within a category aren’t as similar as we believe.
OK, enough of first-grade science class. The moral here is that when it comes to thinking about people, our reliance on categories is no different. First, we turn to categories almost automatically. Second, categorization leads us to exaggerate small differences between groups and overlook big differences within them.
Researchers at the University of Colorado conducted a neuroscien-tific analysis of just how automatic social categorization is.6 Participants were fitted with an elastic cap with sensors that measure through the scalp the electricity associated with brain activity. Respondents were then shown photos of faces that varied by race and gender. So one set of photos included four white men and one black man, another had four white women and one white man, and so on.
The researchers were interested in how the brain reacts to changes in race or gender within a set of faces. For race, it only took 100 milliseconds to see signs of brain response. That’s one-tenth of one second. Gender didn’t take much longer—just 150 milliseconds. In other words, it takes longer for you to sneeze than it does to place the people you meet into social categories.
And what of the side effects of categorization? They also show up when it comes to thinking about people. We maintain rigid boundaries between social categories the same way my daughter does between fruits and vegetables. Consider the deck of cards research described earlier. Not only did overestimators and underestimators have better memory of the other group’s negative actions, but in another study they also anticipated more bad behavior from the out-group to begin with. This expecting and exaggerating of differences between categories is what opens the door for social stereotyping.
One of the best demonstrations of our fondness for absolute categories is the way we talk about biracial individuals, shoehorning them into more clear-cut category labels. What do Barack Obama, Halle Berry, and Tiger Woods have in common? The consensus is that all three achieved African American firsts in their respective professions. But all three are also multiracial, even if you wouldn’t know it by ubiquitous references to, for instance, “the first African American President.” No one talks about Halle Berry as the first biracial woman to win the Best Actress Oscar, or Tiger Woods as the first Asian golfer to win the Masters, even though both statements are factually sound. As with fruits and vegetables, we eschew ambiguity, preferring to see people in either/ or terms.
The other side effect I mentioned—the tendency to overlook differences within categories—also emerges when thinking about people. Look no farther than the popular intuition that members of certain racial groups all look alike. An interesting aspect of the “they all look the same” hypothesis is that it’s equal opportunity: while white Americans think this about Asians, in Asia they think the same about white Americans. Another interesting note is that the claim has no objective support: while dozens of studies have confirmed that we’re better at recognizing faces of our own race, there’s no reliable evidence that certain racial groups actually have reduced variability in facial appearance.7
So if these groups don’t truly “all look the same,” why are we convinced that they do? In large part because of categorization. From a perceptual standpoint, we simply treat faces differently once we’ve decided that they belong to our own group. We process in-group faces holistically, taking in the face as a unit and noting the relationship between its features in addition to the specifics of those features in isolation. We don’t go to those same lengths after determining that a face belongs to the out-group.
Researchers at the University of Texas at El Paso creatively demonstrated this tendency in a study that assessed Hispanic participants’ memory of faces.8 Using a computer program, the researchers created a series of racially ambiguous faces, such that each one could be plausibly categorized as either Hispanic or black. When a “Hispanic-looking” hairdo—medium-length dark hair, often slicked back—was added to a face, respondents’ memory for it was reasonably good. But another set of Hispanic participants saw the very same photos with close-cropped, “black-looking” hairdos, and their ability to remember and later identify the faces was substantially impaired. In fact, participants made more than twice as many errors on the subsequent memory test when a face had a “black hairdo” as when the very same face had a hairdo that made it look Hispanic.
In short, categorization drives how we see one another. After putting someone into the category “us,” you’re more generous with her, you rationalize away her mistakes, you value her as an individual, and you even visually process her facial features more carefully. Upon categorizing someone as “them,” however, you lower expectations, harp on his missteps, and see him as epitomizing general truths about “people like that.” This asymmetry is most pronounced between rival factions but even emerges when category lines are drawn on the most minimal of criteria.
The repercussions wrought by social categorization are immediate and pervasive. As yet one more example, think of your experiences as a pedestrian. About how easy it is to work up animosity toward the drivers who won’t yield to you, who stop in the middle of the crosswalk, who splash you as they speed by. It takes me all of thirty seconds of walking in a congested area to work up a healthy distaste for all things motorist-related—to decide tha
t we walkers are far and away the morally superior breed.
Then I get back behind the wheel of my own car. Before long, I’m muttering under my breath at jaywalkers and those who cross against the light, fully ensconced in the feeling of superiority now bestowed upon me by my newfound category affiliation.
THERE’S NO BETTER WAY to illustrate our facility with categorization than a quick reader-participation exercise. So let’s go for it. If you were in my research lab, I’d sit you in front of a computer monitor and ask you to use a keypad to categorize a series of words as quickly as possible. That’s too high-tech for us right now, so we’ll turn to the next most rigorous scientific method for assessing categorization ability: thigh slapping.
Allow me to clarify: In a few paragraphs I’m going to ask you to place both of your hands palm down on your lap. Your right palm will rest on your right thigh and your left palm on your left thigh. You don’t have to assume the position yet, as doing so will inevitably impair your ability to hold this book or turn pages. But that’ll be the pose I’ll ask you to adopt in a moment.
Your task will be to read through a list of words, classifying each one as either pleasant or unpleasant as you go. For pleasant words, you’ll slap your right hand on your right leg. For unpleasant words, you’ll slap your left hand on your left leg. You should move from the top to the bottom of the word list in order and as quickly as possible—if you make a mistake (and you’ll likely realize it if you do), just correct it before continuing. There are twenty-five words in the list, so anyone walking by as you do this should hear twenty-five separate, arrhythmic slaps. They may also start to ponder the appropriateness of a psychiatric consult, so you may prefer to close the door before continuing.