Against Empathy: The Case for Rational Compassion

by Paul Bloom

The idea I’ll explore is that the act of feeling what you think others are feeling—whatever one chooses to call this—is different from being compassionate, from being kind, and most of all, from being good. From a moral standpoint, we’re better off without it. Many people see this as an unlikely claim. Empathy in

I think this is all mistaken. The problems we face as a society and as individuals are rarely due to lack of empathy. Actually, they are often due to too much of it. This isn’t just an attack on empathy. There is a broader agenda here. I want to make a case for the value of conscious, deliberative reasoning in everyday life, arguing that we should strive to use our heads rather than our hearts. We do this a lot already, but we should work on doing more.

There is a lot of evidence that the foundations of morality have evolved through the process of natural selection. We didn’t think them up.

I am going to argue three things, then. First, our moral decisions and actions are powerfully shaped by the force of empathy. Second, this often makes the world worse. And, third, we have the capacity to do better. But how could empathy steer us wrong? Well, read on. But in brief: Empathy is a spotlight focusing on certain people in the here and now. This makes us care more about them, but it leaves us insensitive to the long-term consequences of our acts and blind as well to the suffering of those we do not or cannot empathize with. Empathy is biased, pushing us in the direction of parochialism and racism. It is shortsighted, motivating actions that might make things better in the short term but lead to tragic results in the future. It is innumerate, favoring the one over the many. It can spark violence; our empathy for those close to us is a powerful force for war and atrocity toward others. It is corrosive in personal relationships; it exhausts the spirit and can diminish the force of kindness and love. When you’re done with this book, you might ask

But there is a related sense that has to do with the capacity to appreciate what’s going on in the minds of other people without any contagion of feeling. If your suffering makes me suffer, if I feel what you feel, that’s empathy in the sense that I’m interested in here. But if I understand that you are in pain without feeling it myself, this is what psychologists describe as social cognition, social intelligence, mind reading, theory of mind, or mentalizing. It’s also sometimes described as a form of empathy—“cognitive empathy” as opposed to “emotional

we are capable of all sorts of moral judgments that aren’t grounded in empathy. Many wrongs, after all, have no distinct victims to empathize with. We disapprove of people who shoplift or cheat on their taxes, throw garbage out of their car windows, or jump ahead in line—even if there is no specific person who appreciably suffers because of their actions, nobody to empathize with.

First, the gap between consequentialism and principle-based moral views might not be as large as it first seems. Many seemingly nonconsequentialist abstract principles can actually be defended in consequentialist terms; they can be seen as useful rules that we are better off applying absolutely, even if they sometimes make things worse. Think about a rule like “Always stop at a red light.” In a sense this isn’t very consequentialist; when the road is clear and you need to get home on time, it’s best overall if you just keep driving. But still, it makes good sense for a society to enforce an absolute rule rather than trusting people to figure it for themselves. The benefits of people not making foolish mistakes outweigh the costs of some lost time at intersections. Maybe we should think about “do not torture” in the same way: Even if there are cases in which torture would be justified, we are all better off with an absolute prohibition. Second, regardless of what abstract moral principles there are, nobody denies that consequences also matter. If Immanuel Kant had to decide whether to hurt someone mildly or kill her, he might well complain that both acts are wrong, but I assume that he would agree that the second is worse. (If not, so much the worst for Kant.)

Empathy is like a spotlight directing attention and aid to where it’s needed. But spotlights have a narrow focus, and this is one problem with empathy. It does poorly in a world where there are many people in need and where the effects of one’s actions are diffuse, often delayed, and difficult to compute, a world in which an act that helps one person in the here and now can lead to greater suffering in the future.

Further, spotlights only illuminate what they are pointed at, so empathy reflects our biases.

empathy distorts our moral judgments in pretty much the same way that prejudice does.

To see these weaknesses, consider the example that I raised in the prologue, the murders of twenty children and six adults at Sandy Hook Elementary School in Newtown, Connecticut, in 2012. Why did this give rise to such a powerful reaction? It was a mass shooting, and over the last thirty years in the United States, these have caused hundreds of deaths. This is horrible, but the toll from these mass shootings equals about one-tenth of 1 percent of American homicides, a statistical nonevent. (That is, if you could wave a magic wand and end all mass shootings forever, nobody looking at the overall homicide rates would even notice.) Actually, in the year of the Sandy Hook killings, more schoolchildren were murdered in one American city—Chicago—than were murdered in Newtown, and yet I’ve never thought about those murdered Chicago children before looking that up, and I’m not likely to think about them again . . . while my mind often drifts back to Newtown. Why? Part of the answer is that Sandy Hook was a single event. The murders in Chicago are more of a steady background noise. We’re constituted so that novel and unusual events catch our attention and trigger our emotional responses. But it’s also in large part because it’s easy for people like me to empathize with the children and teachers and parents of Newtown: They’re so much like those I know and love. Teenage black kids in Chicago, not so much.

Empathy is particularly insensitive to consequences that apply statistically rather than to specific individuals. Imagine

Many believe this. After all, to call someone “empathic” (or sometimes “empathetic,” but let’s not get into that argument about words) is a compliment, with empathy probably ranking close to intelligence and a good sense of humor. It’s a good thing to put in an online profile for a dating site. But this claim about the relationship between empathy and certain good traits is an empirical one, something that can be tested using standard psychological methods. For instance, you can measure someone’s empathy and then look at whether high empathy predicts good behaviors such as helping others. Now this is easier said than done. It’s hard to accurately measure how empathic a person is. But there have been various efforts, and it turns out that the relationship between empathy and goodness is weak. In fact, we’ll see that there is some evidence that high empathy for the suffering of others can paralyze people, lead them to skewed decisions, and often spark irrational cruelty.

One could write a book recounting the good things that arise from empathy. But this is a limited argument in its defense. There are positive effects of just about any strong feeling. Not just empathy but also anger, fear, desire for revenge, and religious fervor—all of these can be used for good causes. Consider racism. It’s easy to think of cases where the worst racist biases are exploited for a good end. Such biases can motivate concern for someone who really does deserve concern, can push one to vote for a politician who really is better than the alternative, can motivate enthusiasm for a war when going to war is the just decision, and so on. But that’s not a sufficient argument for racism. One has to show that the good that racism does outweighs the bad and that we are better off using racism to motivate good action rather than alternatives such as compassion and a sense of fairness and justice. The same holds for empathy. We are often quick to point out the good that empathy does but blind to its costs. I think this is in part because there is a natural tendency to see one’s preferred causes and beliefs as bolstered by empathy. That is, people often think about actions that are kind and just (assistance that works, just wars, appropriate punishments) as rooted in empathic feelings, while they view those that are useless or cruel (assistance that fails, unjust wars, brutal punishments) as having other, less empathic sources. But this is an illusion.

We do best when we rely on reason. Michael Lynch defines reason as the act of justification and explanation—to provide a reason for something is to justify and explain it, presumably in a way that’s convincing to a neutral third party. More specifically, reasoning draws on observation and on principles of logic, with scientific practice being the paradigmatic case of reason at work.

Scientific inquiry is the finest example of how individuals who accept certain practices can work to surpass their individual limitations. Take my attack on empathy, for instance. I really do want to be fair, honest, and objective. But I’m only human, so it’s probably true that this book contains weak arguments, cherry-picked data, sneaky rhetorical moves, and unfair representations of those I disagree with. Fortunately, there are many who are in favor of empathy, and they’ll be highly motivated to poke holes in my arguments, point out counterevidence, and so on. Then I’ll respond, and they’ll respond back, and, out of all this, progress will be made.

You can see this overlap between self and other as a clever evolutionary trick. To thrive as a social being, one has to make sense of the internal lives of other individuals, to accurately guess what other people are thinking, wanting, and feeling. Since we’re not telepathic, we have to infer this from information we get from our senses. One possible solution is that we come to understand people in the same way that we come to understand any other phenomenon, like the growth of plants or the movement of stars in the night sky. But there’s an alternative. We can take advantage of the fact that we have minds ourselves, and we can use our own minds as a laboratory to bring ourselves up to speed on how others will behave and think.

Our empathic experience is influenced by what we think about the person we are empathizing with and how we judge the situation that person is in. It turns out, for instance, that you feel more empathy for someone who treats you fairly than for someone who has cheated you. And you feel more empathy for someone who is cooperating with you than for someone you are in competition with.

We see how reactions to others, including our empathic reactions, reflect prior bias, preference, and judgment. This shows that it can’t be that empathy simply makes us moral. It has to be more complicated than that because whether or not you feel empathy depends on prior decisions about who to worry about, who counts, who matters—and these are moral choices. Your empathy doesn’t drive your moral evaluation of the drug user with AIDS. Rather it’s your moral evaluation of the person that determines whether or not you feel empathy.

This separateness has some interesting consequences. Consider how to make sense of criminal psychopaths. One recent scientific article struggles with the question of whether these troubling individuals are high in empathy or low in empathy. For the authors, the evidence suggests both: “Psychopathic criminals can be charming and attuned while seducing a victim, thereby suggesting empathy, and later callous while raping a victim, thereby suggesting impaired empathy.” So which is it? The authors try to resolve this apparent paradox in terms of a distinction between ability (one’s capacity to deploy empathy) and propensity (one’s willingness to do so). They suggest that these criminal psychopaths have normal empathic ability but adjust it like the dial of a radio—turn it up when you want to listen to the lyrics, turn it down if you want to focus on passing a slow truck on the I-95. Turn up empathy when you want to figure out how to charm people and win their trust; turn it down when you’re assaulting them.

This theory has the advantage of simplicity, as it explains the moral power of empathy in terms of the obvious fact that nobody (well, almost nobody) likes to suffer. It suggests that empathic motivations are, in the end, selfish ones. It’s not clear, though, that selfishness can explain the good acts that empathy leads to. When empathy makes us feel pain, the reaction is often a desire to escape. Jonathan Glover tells of a woman who lived near the death camps in Nazi Germany and who could easily see atrocities from her house, such as prisoners being shot and left to die. She wrote an angry letter: “One is often an unwilling witness to such outrages. I am anyway sickly and such a sight makes such a demand on my nerves that in the long run I cannot bear this. I request that it be arranged that such inhuman deeds be discontinued, or else be done where one does not see it.” She was definitely suffering from seeing the treatment of the prisoners, but it didn’t motivate her to want to save them: She would be satisfied if she could have this suffering continue out of her sight. This feeling shouldn’t be that alien to many of us. People often cross the street to avoid encountering suffering people who are begging for money. It’s not that they don’t care (if they didn’t care, they would just walk by), it’s that they are bothered by the suffering and would rather not encounter it. Usually, escape is even easier. Steven Pinker writes: “For many years a charity called Save the Childr...

To get a sense of the innumerate nature of our feelings, imagine reading that two hundred people just died in an earthquake in a remote country. How do you feel? Now imagine that you just discovered that the actual number of deaths was two thousand. Do you now feel ten times worse? Do you feel any worse? I doubt it. Indeed, one individual can matter more than a hundred because a single individual can evoke feelings in a way that a multitude cannot. Stalin has been quoted as saying, “One death is a tragedy; one million is a statistic.” And Mother Teresa once said, “If I look at the mass, I will never act. If I look at the one, I will.” To the extent that we can recognize that the numbers are significant when it comes to moral decisions, it’s because of reason, not sentiments.

am not arguing that all the biases I have been discussing reflect the workings of empathy. Some do. It’s a lot easier to empathize with someone who is similar to you, or someone who has been kind to you in the past, or someone you love, and because of this, these are the individuals you are more likely to help. The same empathic biases that show up in neuroscience laboratories influence us in our day-to-day interactions. But other biases have causes that go deeper than empathy. We are constituted to favor our friends and family over strangers, to care more about members of our own group than people from different, perhaps opposing, groups. This fact about human nature is inevitable given our evolutionary history. Any creature that didn’t have special sentiments toward those that shared its genes and helped it in the past would get its ass kicked from a Darwinian perspective; it would falter relative to competitors with more parochial natures.

If one is going to do something good, and empathy motivates one to do something less good, then empathy is to blame. But if one isn’t going to do something good, and empathy motivates one to do it, then empathy is a plus.

Sometimes an obsessive concern with unintended consequences is just an excuse for selfishness and apathy. But doing actual good, instead of doing what feels good, requires dealing with complex issues and being mindful of exploitation from competing, sometimes malicious and greedy, interests. To do so, you need to step back and not fall into empathy traps. The conclusion is not that one shouldn’t give, but rather that one should give intelligently, with an eye toward consequences.

More to the point, it is too crude to associate liberal policies with empathy. Consider that many policies associated with liberalism are also endorsed by libertarians, who are, by standard empathy measures, the least empathic individuals of all. Liberals and libertarians share common cause over issues such as gay marriage, the legalization of some drugs, and the militarization of the police. If such policies are grounded in empathy, it is mysterious why the least empathic people on earth would also endorse them.

We’ve seen how conservatives can rely on empathy just as much as liberals. More than that, certain perspectives associated with liberal philosophies aren’t that empathic at all. The best example of this is climate change, something that progressives care more about than conservatives. Here, empathy favors doing nothing. If you do act, many identifiable victims—real people who we can feel empathy for—will be harmed by increased gas prices, business closures, increased taxes, and so on. The millions or billions of people who at some unspecified future date will suffer the consequences of our current inaction are, by contrast, pale statistical abstractions. When liberals argue that we should act, something other than empathy is involved.

It is because of empathy that citizens of a country can be transfixed by a girl stuck in a well and largely indifferent to climate change. It is because of empathy that we often enact savage laws or enter into terrible wars; our feeling for the suffering of the few leads to disastrous consequences for the many. A reasoned, even counterempathic analysis of moral obligation and likely consequences is a better guide to planning for the future than the gut wrench of empathy. This is not a partisan point; it’s a sensible one.

There is a neural difference: Empathy training led to increased activation in the insula and anterior cingulate cortex (both of which we discussed in relation to the neuroscience-of-empathy studies in an earlier chapter). Compassion training led to activation in other parts of the brain, such as the medial orbitofrontal cortex and ventral striatum.

As I’ve mentioned a few times by now, it’s hard to know what to make of these claims, given all of the everyday instances in which we care for people and help them without engaging in emotional empathy. I can worry about a child who is afraid of a thunderstorm and pick her up and comfort her without experiencing her fear in the slightest.

As Cicero said about the merits of friendship—but he could just as well have been talking about close relationships in general—it “improves happiness and abates misery, by the doubling of our joy and the dividing of our grief.” I

But the situations in everyday life force us to confront the problem, to balance self versus family versus stranger. If you’re mathematically inclined, you can think of it in terms of the following formula: Self + Close People + Strangers = 100% Now fill in the numbers. Someone who had Self = 100% would be a pure egoist, and would surely be a monster; someone who had Self = 0% would be some sort of crazy saint. Throughout history, many people have had Strangers = 0%; in my last book, Just Babies, I argued that this is the default mode of human nature. But I can’t imagine that many people have it now; few people would let a stranger die—at least someone in front of them—if a rescue cost them very little. So I know what the numbers shouldn’t be. But I don’t know what they are, or how one could find out, or even whether this is the best way to frame the problem.

The myth of pure evil has many sources. One is what Steven Pinker calls “the moralization gap”—the tendency to diminish the severity of our own acts relative to the acts of others. You can see this in reports of violent criminals who are puzzled why people are making such a big deal of their crimes. The most extreme example is Frederick Treesh, one of a group of three “spree killers,” who allegedly told a police officer, “Other than the two we killed, the two we wounded, the woman we pistol-whipped, and the light bulbs we stuck in people’s mouths, we didn’t really hurt anybody.” In one study, Baumeister and his colleagues asked people to recall either an instance where they angered someone or one where they were angered by someone else. When people remembered incidents in which they were the perpetrator, they often described the harmful act as minor and done for good reasons. When they remembered incidents in which they were the victims, they were more likely to describe the action as significant, with long-lasting effects, and motivated by some combination of irrationality and sadism. Our own acts that upset others are innocent or forced; the acts that others do to upset us are crazy or cruel. The finding isn’t surprising if you consider that violent or harmful acts matter a lot more to the victim than to the perpetrator. If John punches Bill, the event will usually mean more to Bill than to John; both the physics and psychology of punching mean that it has more of an imp...

Indeed, some argue that the myth of pure evil gets things backward. That is, it’s not that certain cruel actions are committed because the perpetrators are self-consciously and deliberatively evil. Rather it is because they think they are doing good. They are fueled by a strong moral sense. As Pinker puts it: “The world has far too much morality. If you added up all the homicides committed in pursuit of self-help justice, the casualties of religious and revolutionary wars, the people executed for victimless crimes and misdemeanors, and the targets of ideological genocides, they would surely outnumber the fatalities from amoral predation and conquest.” Henry Adams put this in stronger terms, with regard to Robert E. Lee: “It’s always the good men who do the most harm in the world.” This might seem perverse. How can good lead to evil? One thing to keep in mind here is that we are interested in beliefs and motivations, not what’s good in some objective sense. So the idea isn’t that evil is good; rather, it’s that evil is done by those who think they are doing good.

I concede that empathy can serve as the brakes in certain cases. But I will argue here that it’s just as often the gas—empathy can be what motivates conflict in the first place. When some people think about empathy, they think about kindness. I think about war.

This conclusion about psychopaths fits well with what we know about aggressive behavior in nonpsychopaths. As we discussed in an earlier chapter, a meta-analysis summarized the data from all studies that looked at the relationship between empathy and aggression, including verbal aggression, physical aggression, and sexual aggression. It turns out that the relationship is surprisingly low. So here’s what we can say about psychopaths and empathy: They do tend to be low in empathy. But there is no evidence that this lack of empathy is responsible for their bad behavior.

I’ve been arguing throughout this book that fair and moral and ultimately beneficial policies are best devised without empathy. We should decide just punishments based on a reasoned and fair analysis of what’s appropriate, not through empathic engagement with the pain of victims. We should refrain from giving to a child beggar in India if we believe that our giving would lead to more suffering. None of this denies that pain and suffering exist, and none of this is dehumanization in the sense that we should worry about. It’s just that we are better off focusing on some things and not others in order to achieve certain good ends. Since the ends matter, this is not cruel; it’s kind. We’ve seen that empathy’s relationship to violence and cruelty is complicated. It’s not true that those who do evil are necessarily low in empathy or that those who refrain from evil are high in empathy. We’ve seen how empathy can make us worse people, not merely in the sense that it leads to bad policy and can mess up certain relationships but in the stronger sense that it can actually motivate savage acts. As we think about empathy, it’s useful to compare

These are hard times for anyone who wishes to defend Cartesian dualism—the idea that our minds are somehow separate from the workings of the material world, that our thinking is not done in our brain. There is evidence from neuroscience—both regular neuroscience and its sexier children, cognitive neuroscience, affective neuroscience, and social neuroscience—making it abundantly clear that the brain really is the source of mental life. It’s long been known that damage to certain brain areas can impair capacities such as moral judgment and conscious experience, and over the last few decades we’ve developed the technology to create pretty, multicolored fMRI maps that show the material manifestations of thought.

Just as one example among many, yes, we often favor those who are adorable more than those who are ugly. This is a fact about our minds worth knowing. But we can also recognize that this is the wrong way to make moral decisions. It’s this ability to critically assess our limitations—with regard to our social behavior, our reasoning, and our morality—that makes all sorts of things possible.

Indeed, high intelligence is not only related to success; it’s also related to good behavior. Highly intelligent people commit fewer violent crimes (holding other things, such as income, constant), and the difference in IQ between people in prison and those in the outside world is not a subtle one. There is also evidence that highly intelligent people are more cooperative, perhaps because intelligence allows one to appreciate the benefits of long-term coordination and to consider the perspectives of others.

Adam Smith’s Theory of Moral Sentiments,

The concern about empathy is not that its consequences are always bad, then. It’s that its negatives outweigh its positives—and that there are better alternatives.