On Saturday, James Alex Fields Jr. drove his car into a crowd of counter-protesters in Charlottesville, Virginia, injuring 30 people and killing Heather Heyer. Earlier that day, white supremacists nearly beat Dre Harris to death. Throughout the afternoon, violence erupted between white supremacists and counter-protesters.
What drove white supremacists to converge on the town of Charlottesville ready to fight and, in a few cases, kill? Did they see the victims as less than fully human and feel no moral obligations to them? Or could an excess of morality—morality that could only be satisfied by punishing a fellow human being—instead have driven their violence?
A popular explanation for horrific violence is that perpetrators see victims as little more than animals or objects, and so perpetrators feel no remorse in abuse, torture, and murder. This process of dehumanization has been invoked to explain violence ranging from the Holocaust in World War II to the prisoner abuses at Abu Ghraib prison.
But this theory is challenged by research we published this week in Proceedings of the National Academy of Sciences. Dehumanization would predict indifference toward others’ welfare, not an active desire and delight in harming them. To understand why one person would actively desire to inflict suffering upon another, we have to look to a counterintuitive source: human morality.
Could an excess of morality—morality that could only be satisfied by punishing a fellow human being—instead have driven the violence in Charlottesville?
Dehumanization allows perpetrators to commit violence to satisfy instrumental goals and self-interest, but perpetrators do not dehumanize victims when they are morally motivated to see them suffer. Specifically, our experiments reveal two classes of motives for violence. One kind of violence is driven by instrumental gain, wherein people do not desire to harm victims but knowingly harm them in order to achieve some other objective (shooting someone in order to steal their money). Dehumanizing victims enables this kind of instrumental violence. The second kind of violence, however, is driven by moral sentiments, wherein people actively desire to harm victims they feel deserve it. This moral violence only emerges when perpetrators see victims as capable of thinking, experiencing sensations, and having moral emotions. In other words, when perpetrators see their victims as human.
This is not to say that perpetrators like the white supremacists in Charlottesville do not hold disgusting and racist views of their victims. They may also insult their victims by explicitly comparing them to apes and other animals. But when we experimentally manipulate dehumanization and measure it in ways that do not tap into explicit racism, we find no relationship between dehumanization and moral violence.
We first demonstrated this by asking people to report how much they approved of different kinds of violence. Then, we asked them about the victims in these examples. Can the victims think and reason? Can they feel pain and suffering? Are they capable of love and compassion? Of hate and anger? We used these questions to assess whether people thought of victims as completely human independent of other negative attitudes they may hold toward them.
We found that dehumanizing victims predicts support for instrumental violence but not for moral violence. For example, Americans who saw foreign civilians as less human were more likely to support drone strikes in foreign countries. In this case, no one wants to kill innocent civilians, but if they die as collateral damage in the pursuit of some military objective, dehumanizing them eases guilt. In contrast, seeing terrorists as less human predicted nothing about support for drone strikes against them. This is because people want to hurt and kill terrorists. Without their humanity, how could terrorists be guilty, and how could they feel the pain they deserve?
In cases of moral violence, our experiments suggest that engaging a sense of morality, not disengaging it, is more likely to cause aggression.
We also conducted experiments in which we asked people to imagine harming someone either for money or as punishment for an immoral act. In one experiment, we found that people are more likely to harm someone for money when the victim is described using dehumanizing language (for instance, simply calling someone a “man” instead of naming him and describing his appearance and personality). This is just as dehumanization theories predict.
But dehumanizing language does not make people any more likely to harm someone to punish immoral behavior. In another experiment, we found that after imagining harming someone for money, people see the person they harmed as less human. But imagining harming someone as punishment for immoral behavior does not change their perception of the person’s humanity. In a few cases, we even found that people justify punishing immoral behavior by seeing the person they harmed as more human.
Many people believe that violence stems from a breakdown in our moral sensibilities. To reduce violence, according to this argument, we need only restore our sense of morality by generating empathy toward victims. If we could just see them as fellow human beings, then we would do them no harm. Yet our research suggests this is not the case. In cases of moral violence, our experiments suggest that engaging a sense of morality, not disengaging it, is more likely to cause aggression. When a group of white supremacists beat a man nearly to death, the primary driver of their behavior is not dehumanization, but rather moral outrage toward an enemy conceptualized as evil and completely human.
Does this mean that there dehumanization plays no role in violence? Absolutely not. Dehumanization leads to indifference, and indifference enables people to look away from atrocities. Every time Donald Trump fails to condemn violence against minorities or when he and other whites argue that there is violence “on many sides,” the effects of empathy for whites and dehumanization of minorities on violence is clear. Dehumanization may not cause a white supremacist to kill, but it does allow the rest of us to stand aside.