Photograph by JOHNNY MILLER
Read Caption
Near Johannesburg, South Africa, a highway separates two starkly contrasting communities: the middle-class, mostly white Primrose neighborhood and the Makause settlement, where unemployed gold miners took over land in the 1990s. Almost all its residents are black.
Photograph by JOHNNY MILLER

Why Do We See So Many Things as ‘Us vs. Them’?

We are wired at birth to favor our own group over others. But science might be able to help us move past what divides us.

This story is part of The Race Issue, a special issue of National Geographic that explores how race defines, separates, and unites us. Tell us your story with #IDefineMe.

Despite their many differences, Solomon Igbawua and Dahiru Bala were close friends. It began when they were schoolboys, running back and forth between Igbawua’s village and Bala’s settlement, only a mile or two apart in eastern Nigeria’s Benue state. They expected their friendship would endure the rest of their lives.

Compact and barrel-chested, Igbawua, now 40, is a Christian and a member of the Tiv, who have farmed Benue’s gently rolling green plains for centuries. Tall and thin, Bala, 42, is a Hausa Muslim. His people—the tightly intertwined Hausa and Fulani—live by herding sinewy long-horned cattle that range over much of western Africa. In many places such differences—of ethnicity, religion, language, culture, politics—are deadly. A few hundred miles to the north of where I met the men, Boko Haram wages war against all who don’t adhere to its version of Islam. Elsewhere in West Africa and beyond, herdsmen and farmers engage in violent attacks over access to resources. And other groups (races, tribes, nations, religions, sects) are locked in other conflicts throughout the world.

Until recently, though, that did not happen in Zongo, Igbawua’s village, or Daudu, where Bala lives. For most of their lives, they told me, there was enough good land for everyone. If cattle trampled a farmer’s field or a herder found his route to a stream cut off by a new fence, there were ways to settle those kinds of quarrels. “There was peace here, and harmony,” says Elizabeth Anyom, a Tiv.

This story helps launch a series about racial, ethnic, and religious groups and their changing roles in 21st-century life. The series runs through 2018 and will include coverage of Muslims, Latinos, Asian Americans, and Native Americans.

But as the two friends grew into men with children of their own, Benue’s population rose. A warming climate dried up lands to the north, sending more herders south. Good land became scarcer. More and more often farmers found crops ruined by herds, and herders found their cattle routes choked off with fences or newly planted fields. Relations were no longer free and easy among farmers and herders, Hausa-Fulani and other groups.

Still, Zongo and Daudu kept to their peaceful shared routines. No one thought conflict could happen here, says Igbawua’s wife, Katrin.

And then it did.

In 2014 “the crisis” fell upon the two communities. The lines between tribes, religions, and culture turned to walls—and being on the wrong side could be fatal. Rumors would spread; then there were raids and counterraids. Crops were ruined; animals were slaughtered. The Tiv village was burned. Men and women were killed.

Igbawua and Bala told me they didn’t attack anyone, but once groups are pitted against one another, that scarcely matters. During the crisis most Tiv farmers acted as if all herders were the same, and vice versa.

The crisis changed people’s norms of behavior. Getting along wasn’t valued; getting revenge was. “I thought they should not take the law into their own hands,” Bala says. “But I didn’t have the courage to approach my own people and tell them that.” He and Igbawua became refugees, visiting their own homes only briefly and during the day, watching out for ambushes.

View Images
During the Unite the Right rally in Charlottesville, Virginia, on August 12, 2017, alt-right supporters block entrance by counterprotesters to the park that’s home to a statue of Confederate general Robert E. Lee. Fights soon broke out between the groups. Later a counterprotester was killed by a rally participant who police say intentionally drove his car into a crowd.

It’s a common misfortune around the world: People get along well enough for decades, even centuries, across lines of race or religion or culture. Then, suddenly, the neighbors aren’t people you respect, invite to dinner, trade favors with, or marry. Those once familiar faces are now Them, the Enemy, the Other. And in that clash of groups, individuality vanishes and empathy dries up, as does trust. It can happen between herders and farmers in Nigeria or between native-born people and immigrants in France or the United States. The situations are very different, and the differences are important. But so is the shared root of their problems: People everywhere are “identity crazed,” as the evolutionary psychologist John Tooby has put it. We can’t help it: We’re wired from birth to tell Us from Them. And we inevitably (and sometimes unconsciously) favor Us—especially when we feel threatened.

Of course, humans share that trait with many other creatures, from ants to salmon to macaques. What other creatures almost never do, though, is change their group perceptions and actions. The birds and bees kept to their tribes when Yugoslavs turned into warring Croats, Serbs, and Bosnians. Only humans—Hutu and Tutsi—could decide they are no longer countrymen, after peacefully sharing a homeland for centuries. Only humans can switch from feeling united as one American nation to feeling divided between conservative red states and liberal blue ones.

Our capacity to change our perceptions also offers some hope, because it permits people to shift in the direction of more inclusion, more justice, more peace. In Nigeria and other places around the world, communities torn apart by group conflict are putting themselves back together with help from a surprising source: scientists who study the mind. Their methods are also helping to improve community relations with police in Toledo, Indianapolis, and other U.S. cities.

View Images

Maria Uwambaje, who lost four of her children during Rwanda’s 1994 genocide, stands with Boniface Twagiramungu, who led killers to those children. Uwambaje has forgiven Twagiramungu. A program led by the Karuna Center for Peacebuilding is helping such perpetrators and victims heal their trauma.

I am a leopard.

Jay Van Bavel, a neuroscientist at New York University who studies group identity, gave me that label last summer when I was lying in an fMRI scanner near his office. While in the machine I was shown photos of faces—12 young white men and 12 young black men. The scanner tracked my brain’s activity as I connected these individuals to group identities. Having been raised in the United States, I have lived with my country’s racial categories all my life, and it wasn’t difficult to do one of my experimental tasks: classify each face according to its skin color as either black or white. However, I also had to work with another set of categories. The men in the photos were on one of two teams, I was told: Tigers or Leopards. The screen told me who was on which team and drilled me on the details until I had it down. But I wasn’t a neutral observer: I’d been told that I was a Leopard.

My scanner tasks (based on an experiment Van Bavel and his colleagues conducted in 2008) allowed Van Bavel to compare my brain’s activity as it worked, first with a familiar and consequential group identity (race in America) and then with a group identity that was effectively meaningless.

Like the brains in the actual experiment, mine lit up differently depending on whether I perceived an in-group face (for me, a Leopards team member) or an out-group (Tiger) face. For example, my orbitofrontal cortex, a brain region associated with liking, sparked up more when I saw a face from my in-group. So did the fusiform gyrus, a region tied to processing the identity of faces.

The experiment—and dozens of others like it during the past 20 years—confirmed several important facts about exactly how the human brain is “identity crazed.” The scans show, for one, that a lot of our perceptions and emotions about groups happen outside our awareness or control. I have no conscious preference for white people over black people. On the contrary, like most Americans, I abhor racism. Yet, had I not been told I was a Leopard, I almost certainly would have shown an unconscious preference for white faces over black ones. That I did not illustrates a different important finding in Van Bavel’s research: New team identities can easily supplant old ones in our minds. All Van Bavel had to do was tell me about two teams and inform me that I was on one. That was enough for my brain to prefer Leopards over Tigers as quickly and strongly as it normally distinguishes blacks and whites.

View Images
Viewed from the Mexican side of the U.S.-Mexico border near San Diego, a corrugated metal barrier between the two nations is dwarfed by four of eight prototypes for the wall that U.S. President Donald Trump’s administration proposes to build. Walls are not just physical barriers—they also reinforce feelings of separation and alienation.

The scans reflected a key fact about human groupishness: We have keen mental radar that seeks to learn what groups matter around us and which ones we are members of. And this radar is always on. Even as we sit comfortably in our racial, religious, national, and other identities, our minds are alert to the possibility of new coalitions.

It’s not hard to see why humans should have evolved to care about their teams and their place on those teams. Relying on each other is a sound survival strategy for a frail, noisy creature without a lot of built-in weapons. Living in groups is a ticket to survival, which is why most primates live in them. In fact, there is no human society without clear lines that distinguish various groups.

“This is how person perception generally works,” Van Bavel told me. “In the first split second, we judge people on the basis of their group memberships.” Caring about your group memberships isn’t something you have to learn, like reading or driving. It’s something you do automatically, like breathing.

View Images
South Korean soldiers stand guard at the border with North Korea in the village of Panmunjom in July 2017. Despite a common language, culture, and ethnicity (as well as a civilization more than 4,000 years old), the people of the Korean Peninsula have been divided since 1945 into two fiercely opposed states that are still technically at war.

In fact, much of our sensitivity to groups begins long before we can speak. Very young babies prefer adults who look like their caretakers over adults who look different; some evidence shows they also prefer the foods their mothers ate while pregnant or breastfeeding over novel ones, and they like the sound of the language they heard in the womb and early in life much better than an alien tongue. These preferences continue. In adulthood most of us are better at recognizing the faces and reading the emotions of people who look and act like us.

Psychologists have long established how remarkably easy it is to awaken our tribal minds. In a classic experiment conducted in 1954, for example, researchers from the University of Oklahoma made and unmade two warring tribes out of 22 local boys. All were sixth graders, came from similar neighborhoods, and were white. Divided into two groups and bused separately to Robbers Cave State Park, the kids were turned loose with just a few guidelines from the experimenters. Each group soon set itself up with a bunkhouse and a swimming hole, gave itself a name, and established norms (one, the Rattlers, cursed a blue streak, while their rivals, the Eagles, prided themselves on clean language). Then, a week in, each tribe discovered the other.

Within days they were at war—raiding each other’s bunkhouses and eating only with members of their own group. Baseball games and other competitions turned into exchanges of insults. Angry talk about “those n***** campers” and “communists” and “sissies” escalated. Then, in the third week of the camp, the experimenters faked some challenges (pulling a disabled truck, unpacking food delivered in crates) that forced the Rattlers and Eagles to work together. The experience of cooperating toward a common goal united them. By the end of the three-week camp, the boys were singing “The Star-Spangled Banner” together and letting bygones be bygones.

As the Robbers Cave experiment illustrated, human beings can shift their group perceptions in both directions. Sometimes we turn Us into Them. But we can also turn Them into Us.

View Images
At Checkpoint 300 near Bethlehem, Palestinians from the West Bank, some climbing the walls to cut the line, wait to be cleared for entry into Israel. Thousands of workers endure the daily ordeal in exchange for better paying work in Israel. Disparities in economic opportunity often reinforce divisions based on religion, ethnicity, or rival territorial claims.

When I met them last October, Solomon Igbawua and Dahiru Bala were attending a general meeting of Tiv and Hausa-Fulani from both of their communities. It was the first time in three years, the first time since the crisis, that Tiv like Igbawua were willing to visit once familiar Daudu. The meeting began with prayers (one Christian, one Muslim) and continued with speeches in praise of restored peace. Then, through my interpreters, I spoke with men and women from both sides. They told us about losing loved ones, losing homes, hiding in the bush for days, becoming refugees. Yet, they said, they get along now. Igbawua and Bala say they can be friends again.

As we spoke, roosters crowed, goats bleated, and children chased each other in the red-brown dust. After a while the mosque rang out with a recorded call to prayer. Now and then a tremendous boom would echo from the valley below, startling goats and chickens, which made people laugh and joke. These were tests being conducted at a Nigerian military installation, a mile or so down the road, testing ordnance. It was hard to imagine these polite, relaxed people fearing and hating each other.

It’s a striking change. It is as if the herders and farmers had taken a medicine to tamp down fear and hate and bring back trust and sympathy for people outside their own groups. In a sense they did. But the treatment was not a pill.

In 2015 a team from Mercy Corps, a nongovernmental organization that works in the region to promote peace, arrived in Zongo and Daudu, along with local peace groups. They brought an offer. The NGO would supply materials and money to construct borehole wells. This would bring clean water to the two settlements. Residents of both would take part in a program that teaches negotiating skills and conflict prevention. Then they’d have to use those skills, building the boreholes together and jointly operating them.

The program, developed by social scientists and rooted in what science has learned about human groupishness, is one example of how scientists, after studying why we’re so “identity crazed,” are trying to do something about it by applying their theories and methods to real cases of Us-versus-Them trouble.

“The idea is to diminish the psychological benefits of conflict and increase the psychological benefits of cooperation,” says Christopher Grady, a graduate student in political science at the University of Illinois at Urbana-Champaign who is helping to measure the project’s results.

Learning to be a good negotiator is “almost a meditative practice, of stopping before acting,” says Arthur Martirosyan, a senior consultant at CMPartners, a consulting firm that designed the training program. “It’s about saying, I recognize this situation, and because I recognize it, I know it might trigger very destructive emotions in me. I want to be in control.”

There is no single theory of why people fall prey to an Us-versus-Them mind-set, nor is there a single approach to helping them out of that trap. But the growing number of researchers who work on the problem do share the scientific method. They begin with verifiable facts about human minds, human behavior, and human society. They use those to create an intervention. Then they test the intervention, as a pharmaceutical company would test a drug: People in a conflict are divided randomly into groups that get the “treatment” and those that do not. Then the researchers compare the groups to judge whether the treatment really led to less violence, more justice, and greater peace. Around Zongo and Daudu are villages that didn’t get to try negotiation training and a joint project. But if analysis by Grady and other scholars shows that this approach works as well as it seems to have in Zongo, Daudu, and some other villages in the study, the program will be offered to many more people.

View Images

Conflict between groups inflicts pain on people who have no say in the fight. After fleeing Myanmar army troops, hundreds of Rohingya children struggle for food at Balukhali refugee camp in southern Bangladesh. In 2017 at least 700,000 Rohingya were forced to flee a campaign of murder, rape, and devastation by Myanmar’s army.

What flight simulators are to pilots, “use of force” simulators are to cops. The version used at Washington State University in Spokane is typical in many ways. You stand in front of the screen, gun at the ready. Each situation playing out on the screen requires a split-second decision, and the stakes couldn’t be higher. One, for example, is a “domestic violence call.” The camera takes you through a short hallway into a kitchen, where a man and a woman are fighting. The man drags the woman out of view. When you catch up, the man has something in his hand, pointed at her. In another situation you confront an uncooperative driver stopped for a traffic violation. He’s reaching for something inside his car. If the thing is a cell phone and you shoot him, you might have taken a life for no reason. If it’s a gun and you don’t shoot, you might be dead.

But there’s a difference between the WSU simulator and most others that police use: It varies the demographic profiles of the people encountered, including their ethnic and racial identities, and places them in scenarios designed to measure and counter bias. So even as officers get practice in the basics of handling tense situations, they’re also learning when and how they treat blacks and Hispanics differently from whites. The Counter Bias Training Simulator (CBTsim) was developed by Lois James, an assistant professor in the university’s College of Nursing, to show officers how they may treat black, Hispanic, and white people differently in the exact same situations.

“The goal of CBTsim is in fact to remove race or any other demographic from this decision-making process,” James says, “by teaching officers to focus solely on the objective level of threat.”

Such objectivity is precisely what society needs from people in occupations in which equal treatment is essential—doctors, lawyers, teachers, military officers, and, of course, anyone in law enforcement. But given the power of our innate tendency to sort people into groups that we don’t think or feel the same about, it is a tough demand. The United States is now in a national conversation about how much and how often police don’t treat all civilians equally. On a range of activities, from traffic stops and low-level arrests to incidents in which police shoot civilians, evidence shows that there are significant disparities in how American police treat nonwhites compared with how they treat whites.

“I was bothered by what I saw as a paradox,” says Lorie Fridell, a criminologist at the University of South Florida, noting that lab studies suggest bias in policing is strong. “On the other hand I know most police want to serve the public and are very offended at the idea that they might be prejudiced. They aren’t racists.”

The key to unlocking the paradox, Fridell says, is understanding that the group biases involved are often not conscious. They’re implicit—happening without our intent and without our knowledge. Like my preference for Leopards over Tigers, which I learned about only after I’d seen a map of it in my brain scan.

The science of group perception reveals how those biases work, Fridell says. It also shows why cops are right to resent the implication that they are especially biased. The unconscious biases that make a cop favor people like himself are just as strong in Fridell, and me, and you.

View Images

Officers in the Spokane Police Department, including Nick Briggs, develop their skills with the Counter Bias Training Simulator at Washington State University in Spokane. Realistic simulations train officers to use real clues—not stereotypes— when deciding whether or not to use deadly force.

Fridell’s Fair and Impartial Policing (FIP) program is a training course for cops. Like James’s simulator, it is designed to get law enforcement to think about managing their often unnoticed preference for Us and against Them so that it doesn’t affect their obligation to treat all people equally. For an officer, Us might be “Us law-abiding people” versus Them, the kind of person who has a rap sheet; or it might be Us cops versus Them, the civilians. But in the United States, with its long history of race-based injustice against African Americans and other minorities, Us is often “Us white people,” and Them is everyone else.

Since 2007 Fridell has received over $1.5 million in grants from the U.S. Justice Department and delivered training to thousands of officers in hundreds of police departments across the U.S. and Canada.

“It is probably the norm that they don’t want to be in the room,” Fridell says of the cops. Police usually start in a range from “defensive to hostile,” she says. This is because police expect to be cast as the “bad guys” in a drama in which everyone else is a good guy. “The debate about police bias has characterized it as being about explicit bias,” Fridell says, as if the main problem is cops who are deliberately racist. Approaching the issue as a scientific take on the way all minds work, she says, lets a more honest conversation take place. “We’re not pointing fingers,” she says. “We’re not talking about police bias. We’re talking about human bias.”

When I visited the Indianapolis Metropolitan Police Department in 2017, some top brass were sitting down to take Fridell’s workshop with community leaders. Like other cities, Indianapolis has a problem of trust between police and minority communities. In many neighborhoods whose residents are mostly African Americans, the police feel more like occupiers than public servants. African Americans, though 28 percent of the city’s population, are only 15 percent of its police force.

Indianapolis’s police chief, Bryan K. Roach, appointed in January 2017, wants to improve relations. He was attracted to FIP, he says, because it is based on scientific research, not opinion. “What caught our eye is there is science behind it, which makes it a little more palatable to people who do not believe that this is a problem.”

Roach decided to have all 1,600 officers in the department undergo implicit-bias training. He began with himself, his top officers, and representatives of communities throughout the city.

“I thought it went very well,” says Patricia Payne, a retired teacher and school administrator who runs anti-racism workshops for the Indianapolis school system. The fact that the training was about the science of groupishness, she says, helped foster a respectful conversation among the activists and the police commanders. “I realized it was the first time I had ever really been in a room listening to the police perspective,” she says.

It’s certainly a good idea to improve communications between police departments and the people they serve. But the bottom line for implicit-bias training is the same one society wants from any other novel “treatment”: Does it work? Getting an answer is a fiendishly hard problem. How do you prove that an attitude expressed in June affected your behavior in October? Scientists are just beginning to get a handle on that.

For example, James and her colleagues recently began a two-year project that will track some police departments in Ohio as they deal with the people they speak with, stop, arrest, or otherwise encounter. In 2018 the study will simply record how cops behave, using randomly selected moments of body-camera footage, citizen complaints, and other sources of data. Next year each force will be divided at random into four groups. One group will just carry on as they always have, without any training (they’re the control group, the behavioral equivalent of patients who get a placebo during a drug trial). Another group will receive only training in the Counter Bias Training Simulator. The third group will go through classroom-based training. And the last one will take both simulator and classroom training.

In the nine months that follow, the researchers will gather the same kind of data as before. The result of this randomized controlled trial, in 2020, will be a mountain of data about how police with different kinds of training compare with cops who didn’t do any work on implicit bias. And that data should establish whether and how such training results in fairer policing.

This kind of careful testing is taking place in other contexts more and more around the world. In a few years it may show that we’ve finally discovered a true science of human groupishness—one that can help us master these instincts, before they master us. This could make it easier to establish a more peaceful and just world.

View Images
In Ramiro, Rwanda, Cyrille Namubonye, a genocide perpetrator, walks with survivor Maria Nyirambarushimana during a Karuna Center for Peacebuilding workshop aimed at building trust and forgiveness. “In their process of healing, Cyrille realized he needs Maria, and Maria realized she needs Cyrille,” says Rosette Sebasoni, Karuna’s project manager there.

But no one imagines it will be easy. Since my visit to Benue last October, violent conflict in the region between farmers and herders has only gotten worse, fueled in part by Benue’s new anti-grazing law, which herders believe aims to drive them out of the state. (It requires that cattle be confined to ranches—ranches that largely don’t exist and that the law makes no provision for creating.)

At the time of writing, however, the negotiated peace between the farmers of Zongo and the herders in Daudu was still intact. In fact, herders threatened in other areas have come to the Zongo-Daudu area for refuge. When that influx alarmed nearby farmers, the residents of the two communities trained in negotiation were able to defuse the tension and foster mutual respect.

Like many people on both sides of the conflict, Solomon Igbawua has linked the new ideas of negotiation training to his deep religious faith. Forgiveness for past misdeeds, he says, is both a good idea for negotiators and an obligation for Christians. He has forgiven the herders, he says, for the most part. But sometimes he sees an older man and he’s reminded that his father was killed in the fighting three years ago.

“I can forgive,” he says. “But I cannot forget.”

David Berreby’s award-winning latest book, Us and Them: The Science of Identity, explores the human instinct to separate into groups. John Stanmeyer, who has worked for National Geographic since 2004, uses his photography, writing, and filmmaking to explore issues around the world.