This conversation with psychologist, researcher, and author Paul Bloom began with a cat tweet. Bloom had shared a scientific study titled “Domestic Cats Do Not Show Signs of Secure Attachment to Their Owners.” My own cat was sprawled languidly on my lap, not because he is securely attached to me, but because the heat in my apartment was off. I retweeted, adding “Well, duh.”

Bloom “liked” my retweet, a small gesture that, to me—a psychology-curious interdisciplinary grad student and follower of Bloom’s writing and guest appearances on podcasts like Very Bad Wizards—signified an opening to contact Bloom and ask him to discuss his recent and, to some, controversial work on empathy.

At a time when politicians, journalists, and public figures of all kinds confidently proclaim empathy as a cure for the world’s ills, Bloom argues for exactly the opposite. In his research on the subject, as well as in experiments conducted with his graduate students at the Mind and Development Lab at Yale University, Bloom has made the case that empathy is an unreliable moral guide, one that can be easily exploited.

Clarity of definition remains important for Bloom in his popular writing, which at times engenders public critiques that range from polite to downright furious. And so throughout our interview, Bloom was careful to define his terms. “What I mean by ‘empathy,’” he told me, putting forth a definition that attempts to complicate the more typical understanding, “is putting yourself in other people’s shoes, feeling what they feel.”

I spoke with Bloom over Skype about his research, interest in developmental psychology, and the complexity of empathy.

—Allyson Kirkpatrick for Guernica

Guernica: Your research has covered such diverse areas as the origins of religious belief, art, and pleasure. What are you focusing on now?

Paul Bloom: Over the last few years, I’ve been focusing on questions having to do with the self, and questions having to do with morality. I’m very interested in why we do good things, or bad things, and where moral judgments come from. With my wife, Karen Wynn, I’ve been involved in experiments with babies concerning the origin of morality, and I wrote a book in 2013 called Just Babies: The Origins of Good and Evil.

I’m currently writing a book on empathy, arguing that it has certain important limitations.

Guernica: Can you explain how you study morality and empathy quantitatively? What do those experiments look like?

Paul Bloom: I do all sorts of different experiments. The experiments we do with babies, for instance, involve showing ten- or twelve-month-olds a one-act play. It could be two different characters; one character is trying to get up a hill, and the other character is either helping it up or pushing it down. Or it could be a play in which a puppet is trying to open a box, and the other puppet either helps to open the box or slams it shut. We show babies these scenes, and we test to see who they like to look at or who they prefer to reach for. In some experiments, we offer them a chance to reward or punish the puppets. These baby experiments allow us to explore what babies know about right and wrong.

I also do work with preschool children. We often give them different scenarios and ask them questions: do you think this was right? Do you think this was wrong? Konika Banerjee and I have been looking at whether or not people naturally believe in fate. We give kids and adults stories and we say: did this happen for a reason, was there a greater purpose for it? We find that kids are much more prone to attribute something to fate than adults.

Then with adults I often do experiments where we’re just asking questions about morality, and looking at demographic factors like your political orientation and where you’re from.

By “empathy,” some people mean everything that is good—compassion, kindness, warmth, love, being a mensch, changing the world—and I’m for all of those things. I’m not a monster.

Guernica: What have you discovered in your recent work about empathy? Have you noticed any trends?

Paul Bloom: My book is going to be called Against Empathy, which may give you a feeling for where my argument is going to go. Whenever I talk about this, I have to begin in the most boring of all possible ways: by defining my terms. By “empathy,” some people mean everything that is good—compassion, kindness, warmth, love, being a mensch, changing the world—and I’m for all of those things. I’m not a monster.

What I mean by “empathy” is putting yourself in other people’s shoes, feeling what they feel. That’s the way psychologists tend to use the term. My argument is that empathy zooms you in on an individual and, as a result, it’s narrow, it’s innumerate, it’s racist, it’s very biased. One way to put it is that because of empathy we care more about a little girl stuck inside a well than we do about something like climate change. Because of empathy, stories of the suffering of one person could lead us into a war that could kill millions of people. Because of empathy, we care more for, and devote far more resources to, someone who is familiar, from our country or our group, than a stranger.

So, I argue that we should be kind, we should be compassionate, and we should definitely be reasonable and rational, but that empathy leads us astray.

Guernica: It sounds as though you’re arguing that empathy is also an unreliable guide for moral decision-making. What are other ways that empathy can lead us astray?

Paul Bloom: In some experiments I’ve done with a graduate student, Nick Stagnaro, we tell people about an atrocity—something that has hypothetically occurred in the Middle East with people being kidnapped, people being tortured—and then we ask, how should the American government respond? We give people a continuum of responses, from doing nothing, to some sort of embargo, to airstrikes, all the way to a full-blown ground invasion. We also give them a standard empathy test; you could take one online. It turns out just as we predict: the more empathic you are, the more you want to retaliate and hurt these people.

And so when people want to inspire you to turn against some group of people, they’ll often use empathy. When Obama wanted to bomb Syria, he drew our attention to the victims of chemical warfare. And in both of the Iraq wars, politicians said, “Look at the horrific things that are happening.” I’m not a pacifist. I think the suffering of innocent people can be a catalyst for moral action. But empathy puts too much weight on the scale in favor of war. Empathy can really lead to violence.

The example I use, which is a recent one, motivated me to write an article because it bothered me so much: I was listening to Trump—because that’s what I do, it’s like a fetish—and he was talking about “Kate.” And he just called her “Kate.” So I looked it up, and it turns out this is someone who was murdered by an undocumented immigrant, as a random street killing. And Trump talks about her all the time. Trump uses her to inspire hatred towards this group of people. He calls them rapists, he calls them killers, and he makes this concrete through “Kate.”

I think the response to somebody like Trump should be: show me the numbers. Let’s get some statistics on the table. It’s an unfeeling response, but I think it is ultimately a more moral one. In fact, if you actually look at the numbers, a lot of the claims that people like Trump make are not true, but since they exploit stories to make their argument, and exploit our empathic feelings, they have an effect.

I bought Ann Coulter’s book—I’m embarrassed to say it—Adios America. You would think since the book makes a vehemently anti-immigration argument that she would appeal to our worst side. But she’s smart. She appeals to our best side. She tells us stories—stories about children who are victims. There is a lot of rape in the book. There’s a lot of murder. Real victims. She exploits our empathic feelings towards these victims to motivate terrible things.

Guernica: I want to go back to what you said about how empathy causes people to really zoom in. How do you see empathy affecting global decisions and public policy? Have you seen people legislating with empathy in mind?

Paul Bloom: I think Americans are always going to care more about Americans than about Mexicans. I think we’re going to care more about Americans than Africans. I don’t think that’s ever going to go away, and I don’t think it’s ever going to go away that people care more about their families than strangers, and their communities over other communities. But I think it would transform the world in such a good way if we could just acknowledge, at least intellectually, that an African life and an American life are the same. I have huge admiration for people who think like the effective altruist, who try to rationally think about how they can change the world for the better, and who try not to be swayed by irrational considerations, such as skin color or whether or not someone lives in the same neighborhood.

I mean, in my own life, I do not live like an effective altruist. An effective altruist would really disapprove of my life. I don’t give enough to charity and I still have both my kidneys. Even the charities I give to are related to things that touch my life, like the Special Olympics. I’m not fully rational; I’m swayed by my biases and my emotions. But I think if we could turn the dial a bit, and try to take what the philosopher Henry Sidgwick called “the point of view of the universe”, and look from above, and realize that we are not special, none of us are, I think it would just cause a transformation.

Politically, this idea is terribly taboo. Every president, Democratic or Republican, simply works on the supposition that it’s better to keep jobs in America than let them go to Mexico. And you might argue it’s better to help our own than to help others. You might argue on utilitarian grounds that the best way for the world to work is for everybody to take care of themselves first. And people have made that argument. But I just think we would be so much better off if we could care for distant others even a little bit more.

That something can be used for good isn’t necessarily a knockdown argument for it.

Guernica: I’ve also heard you describe empathy as a “moral spark.” Can it be good? We’ve talked about it being dangerous, but how can it be used well?

Paul Bloom: Something as important and central and encompassing as empathy can’t be all bad. I think empathy plays a role in intimate relationships, where you might want your partner not just to care about you or understand you but to feel what you feel. The philosopher Michael Slote has a good example of this: the father of a girl who collects stamps could say, “I respect your hobby and your interest in your hobby.” But if he could share her enthusiasm, wouldn’t that be terrific? Also, I think empathy is really important for pleasure. It plays a role in sports. It plays a role in sexual pleasure. It plays a role in fiction when you read stories. And, yes, I think empathy can serve as a moral spark, motivating us to do good things. But anything can be a moral spark.

Racism can work that way; it can motivate behavior that turns out to be right, as when a politician might use a racist appeal to generate support for a war that’s actually a just war. And so the fact that something can be used for good isn’t necessarily a knockdown argument for it. You have to weigh the pros and cons. Some people also think that without that spark of empathy we would do nothing, but that’s just flat-out wrong. You could feel compassion for somebody without the spark of empathy. Here I draw a lot from Buddhism, which focuses on compassion and kindness, loving kindness, as they call it, but rejects empathy because it’s a poor moral guide. And I think there’s a lot of evidence suggesting that they’re right.

Guernica: Is there a way to avoid the potentially negative effects of empathy?

Paul Bloom: The question of how to circumvent it is an interesting one, and I don’t pretend to have a quick answer for it. We can try our best, in some ways, to exploit other emotions in humans, but I think the real answer is more on an institutional level. As a parallel case, there is all sorts of evidence in social science and psychology about biases when hiring people. I think the way around these biases isn’t to say, “Well, I will try very hard not to be biased,” because it just doesn’t work that way. The way around it is to set up institutional procedures that don’t allow bias to take place. My favorite example of this occurs with symphony orchestras, which are heavily male. People used to say, “Look, men and women just have different styles. We’re not being sexist here; we’re just listening to the music.” Then some smart people did a simple thing: they had musicians audition behind screens, so you would only hear the music, because that’s what counts. Then, all of a sudden, the sex discrepancy disappeared. It’s not like there were evil men rubbing their hands together in evil sexism; they genuinely thought they were judging the music, but they were influenced by other factors. Here and elsewhere, this suggests that good intentions aren’t the cure. There are other sorts of rational cures that can take place.

Guernica: I’m wondering, how did you come to work in developmental psychology? What were you curious about? Were there any formative experiences that led you here?

Paul Bloom: Nobody ever asks me that. My brother is severely autistic, so when I was a kid I spent a lot of time as a teenager in camps and programs for autistic kids. When I went to McGill as an undergraduate, I figured I’d be a therapist working with these kids. The truth is, and I knew this even back then, I’m just not good at this. I’m too empathic to do this sort of thing. I’m impatient, I’m easily bored. I have none of the traits that would make me good for working with special needs kids. But I went through the psych program, and almost through a chance encounter I ended up working with this brilliant professor, John Macnamara, and he did work connecting philosophical work with psychology—incredibly rich, fascinating work, and I just fell in love with it. It was so much deeper than any of the psychology I had previously studied, and I began to realize that the methods of psychology could be used to explore the deepest questions that exist. It was one of these lucky encounters that changes your life. It was at a party at his house I met my wife. He directed me to go to MIT to work with Susan Carey, who became my graduate adviser, and so I did my dissertation on language development, on how children learn to talk. Then, gradually, my interests drifted to different things: I became interested in how children understand the minds of other people, and in what children enjoy, which brought me to the broader question of pleasure and how pleasure works for both children and adults. I suffer from a little bit of an academic version of Attention Deficit Disorder, where every few years I drift to a different topic. But, to answer your question, it started with my brother, and then things just went on from there.

Guernica: How do you find your work with teaching informs your work with research?

Paul Bloom: I teach this large introduction to psychology course, which I’ve taught for many years to, you know, three hundred, four hundred, five hundred kids. That forces me to take topics that I find interesting and make them real to a broader audience. I actually think, as a result, it helps me think better about those topics. I feel the same way about my popular writing. When you start writing things to try to persuade someone who’s not already part of your guild or your profession that something is interesting, it forces you to ask yourself, “Well, why is this interesting?” I also teach a small seminar on a different topic every year to graduate students and undergraduates by invitation. I’ve taught courses on the cognitive science of good and evil, fiction and imagination, and something on the seven deadly sins. I taught a course called “Habits of Mind” last semester with a friend of mine, Tamar Gendler, who is a philosopher, and we explored questions such as the relationship between emotions and reason, and what it is to live a good life. I also love doing research with graduate students and talking about ideas with them. Overall, I love teaching. I wouldn’t take a job that didn’t include it.

I am never going to write about dogs again. You can write about Islam, you can write about sexuality, but no, not dogs.

Guernica: Speaking of your public writing, people sometimes lash out online in response to your work on empathy. What do you think of this?

Paul Bloom: I published this thing in the New Yorker on empathy and then I immediately went on Twitter wondering what people were saying about it. And the very first tweet—and I’m sort of paraphrasing here—was: “This is probably the stupidest fucking article I’ve ever read in my life.” And then it kind of went downhill after that.

I think there are two reasons why I get backlash. First is that they don’t actually read what I write, and they think I’m some sort of psychopath. But the other reason is—and I want to show proper respect here—some people read it and they just think I’m really wrong, and sometimes they’re not polite in saying so. But, you know, I kind of like social media, and I like hearing from people. I don’t like the ugly stuff, but there are some people—smart people—who have a very different perspective, and I’ll get a backlash from them. And this isn’t necessarily a bad thing.

When I write I’ll sometimes say things which are somewhat controversial—not because I’m seeking out controversy for its own sake, but if I don’t have anything to say which is different, why am I bothering to write stuff down in the first place? It would be nice if everybody who had something interesting to say about my work could say it politely and civilly, but it doesn’t work that way… Sometimes people are just really nasty.

One other thing. When I wrote an op-ed in the New York Times saying there is no such thing as an immaterial soul, I got some polite emails in response saying, “I beg to differ,” and “I disagree.” But when I wrote an article about communication in dogs, I got the worst hate mail I’ve ever gotten in my life. If you write about dogs, people freak out. I am never going to write about dogs again. You can write about Islam, you can write about sexuality, but no, not dogs.

Guernica: Do you think empathy is learned, or are we empathic by nature?

Paul Bloom: I think there’s some evidence that we’re empathic by nature. There is some evidence from studies of babies and young children that they resonate with the pain of others, and there’s some work by Frans de Waal that other primates also resonate with the pain of others. It gets complicated because it’s hard to pull apart empathy from compassion. What is really clear is that we innately care for other people at least to some extent.

Individuals differ in how empathic they are. Some people would really flinch if they watched me hitting my hand with a hammer, and other people would just not care. People differ as well in where they direct their empathy and their compassion. Many people are intensely concerned about the suffering of non-human animals, and some do not care at all. There are cultural differences.

As we discussed, my own view on this is a little bit unusual. I think we should really discourage this sort of empathic engagement when it comes to making moral decisions. I think we should focus on something like compassion, on getting people to care more for others without putting ourselves in their shoes. I think, for instance, there’s some evidence that when it comes to being a doctor or nurse, a police officer or therapist, that empathetic engagement leads to burn-out. Imagine if you’re dealing with severely ill children, and you felt their pain all the time, and the pain of their parents—you wouldn’t be able to do that job for very long. It would kill you. What you need is to care for them, but have some distance. Pull back. That’s what I’m arguing for.

Guernica: Do you think the mainstream conception of empathy conflates it with compassion?

Paul Bloom: I think so. Sometimes people forward me emails or articles where it says, “More empathy is needed in the Middle East,” or “Doctors need to have more empathy,” and they expect me to get upset. But for the most part, people use “empathy” to mean everything good. For instance, many medical schools have courses in empathy. But if you look at what they mean, they just want medical students to be nicer to their patients, to listen to them, to respect them, to understand them. What’s not to like? If they were really teaching empathy, then I’d say there is a world of problems there.

So, in some way, I’m using the term in a specific way. There’s no better English word for what I’m talking about. But it’s fine with me if people want to use the word in a different way.

Guernica: A lot of people have been discussing empathy recently with regard to developments in virtual reality.

Paul Bloom: I’m extremely interested in that. I’m really interested in the pleasure we get from stories and the pleasure we get from movies, and certainly the pleasure we get from virtual experiences. My complaint is against empathy as a moral guide. But as a source of pleasure, it can’t be beat. The ultimate aim of VR is to put yourself in the shoes of another person, someone climbing a mountain, or scuba diving, or flying. But also maybe fighting a war. And there’s a huge human appetite for this sort of empathic experience. I don’t think we’re very good at it, but it’s just amazing to put yourself in the shoes of Humbert Humbert, or Anna Karenina, or a kid with autism, or a black teenager in the American South. I think a lot of VR might enhance that experience, and I’m all for that.

Guernica: You mentioned you’re a fan of social media. Some people say teenagers are becoming less empathic because they’re so dialed in.

Paul Bloom: I’ve heard there’s been a drop in self-reported empathy in the last 20 years, and I honestly don’t know what to make of it. Here’s what I’ll add: By every objective measure, the period of this drop in empathy has been a drop—at least in the United States—in every form of violent crime, murder, rape, assault, and so on. It has been a period where people have been far nicer to one another in every possible way. I’m not saying it’s because we’re dropping our empathy that we’re nicer to each other, just that the drop doesn’t seem to be causing any harm.

The effects of Twitter and Facebook and all those things on people’s psychologies is a really interesting question to which nobody knows the answer right now. There was a whole lot of literature suggesting that Facebook makes you lonely—then there are recent experiments suggesting that that’s not true. I have two teenage sons, and they’re both surviving, thriving, and having a great time, and they’re always on social media. You know, my younger son told me nobody uses email anymore. I’m this old fogie with my email. I don’t know what I’m supposed to communicate with now—SnapChat?