Against Empathy? What Paul Bloom Missed

When I was five or six years old, my father somehow managed to get our two-wheel-drive rental car stuck in the middle of a remote Canadian beach. I’m actually not quite sure why it happened, but I can recall the sudden lurch I felt as we abruptly stopped moving and the shrill squeals of the tires as they spun deeper into the sand. After a dozen or so taps of the accelerator and some cheeky attempts to re-angle the tires, the air inside the car reeked of burning rubber and my father, having finally come to terms with our predicament, taught me some new words that became the cornerstone of my adult vocabulary. My mother, who had been openly critical of the idea of getting anywhere too close to the sand, struggled to determine the appropriate cocktail of emotions to display. She eventually settled on a heady mixture of amusement, frustration, and just a dash of blind rage.

— A review of Against Empathy: The Case for Rational Compassion by Paul Bloom —

This particular tale ends with a nice local Canadian couple towing us off the beach in their four-wheel-drive truck. They’d been watching us, they said with thinly veiled amusement, from a ridge above the beach. When it was clear that our car was irreversibly stuck, they headed down to help. After the tow, they refused to accept monetary compensation and even invited us to a barbecue, which my parents politely declined.

What that couple did for us was a good thing to do, surely a moral thing. They had no overt obligation to help us and yet they did so. Why? I asked my mother and my father this question later that night at the hotel and received two similar, but enlightened and distinct responses. My dad answered first: “You help people, Chris. That’s just what you do. It’s the right thing.” “Yes,” my mother said. “Oh, I just think about how it must have looked to them when you got out of the car. If I saw a family with a little boy or girl in trouble I’d go help immediately. I just couldn’t bear the thought of them being upset or afraid.” I’m paraphrasing, of course, but this was the general idea. My parents had each, individually, reached the same moral conclusion, but one was driven more by empathy than the other. I hadn’t thought about this encounter with kindly folk on a distant shore in many years, but reading Paul Bloom’s book Against Empathy: The Case for Rational Compassion brought it to the front of my mind again.

In his book Bloom, a professor of Psychology and Cognitive Science at Yale University, argues against the usefulness of empathy as it pertains to morality. Bloom begins by clarifying that he is against emotional empathy as a compass for ethical decision making and goes further to suggest that this may even be harmful in our personal lives. What he does advocate for, however, is cognitive empathy which is the ability to understand another perspective without necessarily mirroring another person’s emotional state. When I first read the title of the book, I had an extremely negative visceral reaction — who the hell is against empathy anyway — but Bloom’s position has a great deal of nuance to it that the title doesn’t fully capture and, ultimately, I think the title Bloom (or his editor) went with will sell better than Against Emotional Empathy Insofar as it Pertains to Ethical Decision Making.

I can only assume that you’re still skeptical at this point. Rest assured that I am, too. In fact while I find myself agreeing with what most of Bloom says in his book, I’m not convinced that all of his arguments are correct. The aim of this piece, then, is to first outline Bloom’s major arguments and then dive deeper — just a bit — into some of the issues I have with his claims. The major strength of Bloom’s book is its accessibility; he uses a combination of anecdotes, historical accounts, scientific evidence, and philosophy to make his case. What’s truly unique about psychological phenomenon is that they’re extremely difficult to analyze objectively, unlike results from the hard sciences. To be sure, extrapolating the academic findings of psychological inquiry to make real world inferences is invariably contentious. What’s most fascinating is that in any inquiry of this nature we’re faced with the omnipresent dilemma of higher-order primates: we’ve got to use our own faculties (intellectual, emotional, or otherwise) to examine those very same faculties.

Against Empathy is a multi-disciplined study of empathy (and ethics to a lesser extent), but ultimately is within the realm of what I’d call “scientific philosophy.” Nothing in philosophy is ever really resolved; we still grapple with the same problems today that humans puzzled about millennia ago. Bloom isn’t shy about citing Adam Smith, for instance, when Smith has something interesting to say about empathy or the equivalent term to Smith which, at the time, was sympathy. Likewise, many others have come before Bloom who suggested similar ideas about empathy, and he cites those readily as well. I can’t hope, in the space I have here, to present such a thorough or well-researched approach to the issue, so in this piece I hope to examine the book based on its own merits and speculate as to whether or not empathy — emotional empathy — can play a useful, positive role in modern moral decision making.

While reading I found it necessary to remind myself often that Bloom advocates against emotional empathy and for cognitive empathy. For most people, empathy means both the ability to understand another person and also the ability to feel the emotions of another person. Bloom and others consider these types of empathy to be distinct although I am not entirely convinced that one can exist without the other. For now, however, consider the two as seperate.

In the story about my parents, my father’s position was cognitively empathetic, or perhaps just altruistic. My mother’s position, while it had some of the same elements, was decidedly an emotionally empathetic response to the situation. This is unsurprising to me; she is quite readily the most emotionally empathetic person I know. In her interpretation of the events on the beach, not only was she able to empathetically interpret the feelings of the couple on the ridge, but she was able to envision how they might empathize as well. Empathy within empathy. Of course, who knows why that couple actually did what they did — we never asked — and this is one of Bloom’s arguments against emotional empathy; sometimes we project feelings that may not actually exist and could lead us to make misinformed decisions. Bloom even points out how people tend to project their emotions onto animals with emotional empathy:

“Many people believe that dogs enjoy being hugged, for example, presumably because we enjoy being hugged. But this is probably wrong: Dog experts tell us that dogs don’t naturally enjoy being hugged; they suffer through it.”

Last year while sitting outside at a café, a man on the sidewalk near me begin screaming and beating the woman walking with him. In response, she attempted to put him into a bear hug of some sort, but he broke free and continued his assault: punching, slapping, flailing, and screaming with a renewed intensity. It quickly became clear, to me anyway, that the man was severely autistic, which led me to believe that the woman was probably his aide or friend or family member. I was lucky; I happen to know a number of people who specialize in treatment of autism spectrum disorders which probably clued me in on the situation faster than some of the other people on the terrace. Still, I hadn’t a clue what to do. Others at the café seemed to pick up on the uniqueness of the situation, but we all sat there, and watched the situation unfold. I imagined myself in the woman’s situation: I would want someone to help, but I also knew that this could make the situation much worse. In a physical manifestation of the internal war between my logic and my emotions, I stood up but stayed rooted in place. Shortly thereafter, the woman managed to calm the man down and they walked away. I later confirmed with my professional friends that I had done the right thing: the woman was trained, or at least knew how to deal with such outbursts, and any outside intervention was almost guaranteed to make the situation worse. I — we all — had behaved morally by inaction, but it sure didn’t feel like it.

It’s important to mention that Bloom is a strong advocate for consequentialism as a preferable ethical system. Bloom believes, and I largely agree, that a rational consideration of the consequences of an action will often yield the best moral course. In an extreme example, but one of my favorites, Bloom talks about a difficult decision Churchill had to make regarding the lives of British citizens during WWII:

“In World War II, the British military had cracked the Enigma code and had advance notice of impending German attacks on Coventry. But if they prepared for the attacks, the Germans would have known the code was cracked. So Churchill’s government made the hard choice to let innocent people die in order to retain a military advantage, giving them a better chance of winning the war and saving a greater number of innocent lives.”

But does emotional empathy regularly lead to less desirable moral outcomes than does a rational consideration of available data? Bloom thinks so, and again I find it hard to entirely disagree. Had Churchill decided to prepare for the attack on Coventry, for instance, it’s probably correct to say that the Germans would have at least become more cautious about their communications using the Enigma machine. At worst, perhaps they would have given it up all together and more civilians and combatants would have become the casualties of war. This same thinking is applicable to smaller scale moral quandaries, but these can yield some difficult and uncomfortable considerations.

Probably Bloom’s most emotionally controversial ideas are those involving charity. He parrots Peter Singer’s opinion of the Make-a-Wish foundation, which is that there are probably more worthwhile causes to give to. I want to quote Bloom as completely as possible to make sure his position on this issue is well represented. It may seem like a cold thing to say, but he makes a good point:

“Miles Scott, a five-year-old with leukemia, was helped by the Make-A-Wish Foundation to spend the day as a superhero—Batkid…The Make-A-Wish foundation says that the average cost for making a wish come true is $7,500. The Batkid scenario certainly cost more… Singer tells us that if this same money were used to provide bed nets in areas with malaria, it could save the lives of three children …“It’s obvious, isn’t it, that saving a child’s life is better than fulfilling a child’s wish to be Batkid? If Miles’s parents had been offered that choice—Batkid for a day or a complete cure for their son’s leukemia—they surely would have chosen the cure. When more than one child’s life can be saved, the choice is even clearer.”

As Bloom says repeatedly throughout his book, emotional empathy is a spotlight on individuals. Commercials and mailers for charities for children or the “Arms of an Angel” ASPCA ads are specifically designed to capitalize on the emotional empathy of potential donors by showing heartbreaking pictures and playing emotionally charged music. Bloom himself says that he has been a sponsor of an African child and his family for many years because he just couldn’t say no to a mailer he received. He never suggests that this isn’t a good thing to do, only that it might not be the best thing. For instance, long-term economic development in Africa and in other regions is undoubtedly inhibited by the influx of donations from richer parts of the world; where is the economic incentive to develop infrastructure if food and goods are received for free? What’s worse, in less politically stable regions, those goods may go directly into the hands of corrupt politicians or warlords. There are indeed cases where emotional empathy can lead to less than desirable moral outcomes.

Bloom also argues that he does not believe emotional empathy to be central to the development of morality in humans, or that it is at all necessary for moral action. He suggests that if he saw a little girl drowning near the shore of a lake, he has no need to feel what she is feeling to know that he should save her. He emphasizes the distinction between emotional empathy and compassion and kindness. The academic consensus regarding the evolutionary origins of empathy is that empathy (in general) evolved out of a need for mammalian parental connections, given how mammals tend to raise their young. Mothers especially would need to understand when their offspring was hungry, hurt, sick, or fearful so that they could respond appropriately. I find that intellectually compelling, so I’m inclined to agree.

The Problem is This

In summarizing Bloom’s argument, I’ve tried to be as comprehensive as possible, but I readily admit that I haven’t done it the justice it deserves. The text, while accessible, is very dense with numerous sources on almost every page. Given the information I’ve provided, however, and with some additional information hereafter, I hope to adequately express some of my major objections to Bloom’s position.

I will say again that I largely agree with Bloom about his main thesis, namely that cognitive empathy is probably a superior way to determine the moral worth of an action. I think emotional empathy can be especially dangerous if a person has themselves experienced something traumatic, something that they connect with on a level that others may not be able to. For instance, I once offhandedly mentioned in a group conversation about jogging that it was relatively unsafe for anyone, especially women, to go to certain regions of Washington DC alone at night or in the very early morning. I said this based on recent news of a string of muggings and sexual assaults in those areas. A woman who, unbeknownst to be, was a victim of sexual assault, began to loudly criticize me for this statement, and suggested that I was enabling rape-culture because I wasn’t focusing on the issue, namely that it isn’t right that women can’t feel safe in their own cities. All I had intended to do was urge caution — you certainly wouldn’t see me in those areas around those hours — and I was upset that my remarks had such a negative effect, but also perplexed that such a simple remark (to me) had caused such a negative reaction. I think this is an example of times when emotional empathy can be a detriment; there’s a big difference between sweeping an entire issue under the rug and making a statement about the here and now, about the way things are.

Where I do not agree with Bloom and probably many academics is that I believe that there is nearly always a conjunction of emotional empathy and cognitive empathy at work in moral decisions. I have other objections as well, but many of them follow directly from this objection, so it’d be wisest to start at the beginning.

The Golden Rule

I wrote before of the necessity of empathy in the United States after the 2016 Election. What I most nearly meant when I used that term was cognitive empathy; I don’t think it is necessary for citizens of the US to “feel” the pain of their Democrat or Republican counterparts, but rather to understand those positions intellectually. Nevertheless, I don’t believe even this intellectual understanding happens in a vacuum. Bloom mentions psychopathy as an example of a fringe case where an individual can be cognitively empathetic, but have few feelings of their own. Because of their understanding of other’s emotions — high emotional intelligence — they have the ability to manipulate other people without feeling emotions themselves. But if psychopaths are a rarity and the rest of us are on a spectrum with varying capacities for emotion and logic, then how can we be sure that emotion doesn’t somehow subconsciously factor into cognitive empathy?

I don’t want to misrepresent Bloom’s point here: he suggests that we should force ourselves to behave rationally especially when it comes to maximally affecting large numbers of people — and I agree — but he also suggests that emotional empathy doesn’t entail morality necessarily. He also isn’t suggesting that emotional empathy can’t lead to moral outcomes, only that it generally doesn’t. If it’s true, however, that our emotional empathy does contribute beneficially to our cognitive empathy on a large scale or in ways cognitive empathy cannot, then this would seemingly weaken his point. Here I’m sure Bloom would bring up the following example:

“I told a story earlier from Jonathan Glover about a woman who lived close to a concentration camp and felt empathy for those being tortured. Her response was to ask that the torture be done elsewhere, where it wouldn’t disturb her. This was one of a series of examples meant to show how empathy need not make us good.”

Similarly, if you’ve ever watched the ASPCA commercials backed by the Sara McLaughlin track in a group setting, you’ve undoubtedly heard a chorus of groans or someone shout “Change the channel!” Bloom also suggests that if you’re about to pass a homeless person on the street, you might avoid them intentionally so that you can suppress or avoid an emotionally empathetic response. You might also do this to pushy street representatives of the World Wildlife Fund or another charity. You might even strategically position yourself on the inside of a group of pedestrians to avoid having to interact with anyone. You might do that, but I surely never would.

Where I think Bloom’s perspective in this regard is limited is that he seems to treat moral endpoints of emotional empathy as a binary outcome. For instance, I can’t give money to every homeless person I encounter and maybe that isn’t even a beneficial thing to do, but what I try to do is acknowledge them if they ask me for change or for food. For me, this is a mixture of emotional and cognitive empathy; I’ve weighed my options and I understand what I am capable of giving. More often than not you’ll see people walk by and utterly ignore a homeless person, which I think is dehumanizing; the least anyone can do is give them a shred of dignity by acknowledging that they exist. I’ve been doing this for years, and it was motivated by a single instance where I saw a homeless man ignored entirely by a crowd of well-dressed businessmen. I experienced — or at least I think I did — what that homeless man might have felt, and it wasn’t anything pleasant. When I started acknowledging requests for help even if I had nothing I could give, the response — the facial expressions, the posture, the tone of voice — were all so positive, that — again — I felt emotionally charged by the encounter and continue to do and see the same thing today.

This gets quite messy here, academically. Bloom could argue, and probably justifiably, that you could come to the same conclusion as I did by using cognitive empathy. Again, even if he’d grant me that this is a morally beneficial outcome derived in part using emotional empathy, Bloom would point out that on average empathy is a burden to making intellectually sound moral decisions. Again, I agree. But I’d like to quote something he says in the prologue of his book and discuss the consequences of a world free of emotional empathy (my emphasis added):

“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… 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 what’s not wrong with empathy.”

Yet again, I agree with most of what Bloom says here, but I believe that empathy can be a uniquely powerful motivator for recurrent moral choice. Bloom’s major argument revolves around large-scale policy decisions, but how many of us are policy makers? We can all make smarter decisions that will act as force multipliers for good, but I believe emotional empathy is a crucial component of how we treat the relatively small number of local people we interact with on a given day. Perhaps emotional empathy should be reserved only for small scale encounters, tempered with a healthy dose of cognitive empathy, of course. On a small scale, if we’re motivated by emotional empathy and kindness, on the aggregate — with enough people — those actions can have huge and resounding impacts. Plus, if we consider something like donating money to Make-a-Wish, if we’re to approach that from a position of cognitive empathy, we should be sure to give it a fair shot: what are the externalities that occur when people donate to such a “feel-good” charity or see the results of the charity? Are people nicer to one another for the next twenty-four hours? Will they more readily help others? Will they then more readily use cognitive empathy to make bigger decisions? Perhaps the positive effect of emotional empathy is more impactful than we are presently aware of.

Perhaps if we learn to use our emotional empathy properly it can generate repeat business for the greater good. True remorse, for instance, I think is almost entirely a facet of emotional empathy; if you can feel, in some way, the pain you’ve caused to others, you’re much less likely to re-offend in the same way again. Likewise, emotional empathy could be useful on a large scale if we learned how to extrapolate properly and could apply the feeling and sentiment we derive to areas where we can have a more direct local impact. For instance, if we can to some extent feel the pain of an African child we see in a commercial, surely we can use that empathetic response as motivation to help homeless children within our own communities.

Bloom would suggest that I’m advocating for cognitive empathy here, and he’d be right. What I’m attempting to do, however, is to take a less hard-lined approach to emotional empathy and also to suggest that it may influence our moral decision making in unseen ways even before our thoughts can be effectively processed and analyzed. Bloom does allude to this idea — that we’re reacting and processing our feelings always after they’ve already happened. In my estimation he’s right yet again when he suggests that emotional empathy can be hard on the spirit. We can all try to imagine, using emotional empathy, what it would be like to be forced to feel all of the emotions of everyone around you. It would be absolutely unbearable. Even to a lesser degree, having too much emotional empathy can also be hugely harmful for someone personally. Bloom talks about emotional contagion, where one individual subconsciously or inadvertently adopts the feelings or mood of others by being in proximity with them, interacting with them, or observing them. He gives a specific example of his own experience with this phenomenon:

“More than once I’ve found myself in a dark mood and only later realized that it was because I had been interacting with someone who was depressed… Without an appreciation of the source of one’s suffering, the shared feeling is morally inert.”

This is the only point where Bloom and I wildly disagree. A shared experience of suffering, even if the actual source is not understood, can be invaluable. For those afflicted by depression — like myself, for instance — the simple idea that someone else, somewhere is enduring similar torment is anything but morally inert. The very expression of that suffering itself is, in my opinion, a moral act. Certain musicians and authors have special meaning to me, because not only do I cognitively understand their suffering, I can experience it with them through their words. In a bizarre, macabre way, that type of shared suffering is one of the only things that helps, even if its etiology is unclear or varies wildly from person to person. Bloom, I suspect, does not have depression, so perhaps for him that type of emotional contagion is a burden, but for others it is a paradoxical release. This is a finer point that I think Bloom may miss; consequentialism as a system may not be sufficient to capture the nebulous or intangible good that can come from human connection generated by emotional empathy. In fact, emotional empathy may be the only avenue through which the moral good of shared connections can be established. Let me put it another way: I’d rather speak with a fellow sufferer of depression over one or ten beers than I would sit in an office with someone who merely intellectually understood my position. Maybe I just prefer the beer.

In that same quote, Bloom admits that he was in a dark mood and only later realized it was emotional contagion. This makes me wonder, as I said earlier, if perhaps we are unaware of the benefits emotional empathy conveys upon our decision making processes. Perhaps, as empathy originally evolved for use in small populations, its effects today are magnified wildly. I wonder if we’re not getting the full picture of its role in even the most rational decision making. Our feelings are constantly jockeying for attention from our rationality: you’re hungry, you’re cold, you’re angry, you’re tired. Is it possible that those feelings influence us in subtle ways before they’re actually perceived by our rational brain? Surely. So, I wonder if Bloom isn’t jumping the gun by suggesting that even the most rational use of morally principled cognitive empathy isn’t influenced in some unseen way by emotional empathy.

I’ll share one last anecdote. While strolling through the National Zoo last summer I walked by an outdoor Gibbon enclosure. A middle-aged woman was on all fours in front of the enclosure loudly cooing and talking to one of the Gibbons. At first, I thought she was a zookeeper, but the lack of khaki shorts, a radio, and the very embarrassed looking man standing next to her soon made me change my mind. Curious, I inched closer, pretending to look at the Gibbons, but all of my attention was peripherally focused on the woman. “Look, Jack, he’s smiling at me. Oh, Jack, he knows that we’re both kinds of monkeys!” The Gibbon was obviously baring its teeth at the woman, who mistook it for a smile, and she continued her decidedly one-way conversation for another few minutes until she gave into her husband’s pleas to leave. Right before she left, the Gibbon turned its back to the woman and deposited a nicely sized goodbye gift on the ground for her.

Now if we accept the hard-line differences between cognitive and emotional empathy, this woman — we can assume — wasn’t feeling what the Gibbon felt. She was doing her best to understand its state of mind and wildly misinterpreting it. This is an extreme and ridiculous example, but my point is that even cognitive empathy doesn’t always yield rational results. Bloom talks at length about the rational capacity of human beings and their ability to use cognitive empathy, but I think he too readily assumes the homogeneity of human rationality. Consider political opinions. While everyone is subject to the same general logical rules, people construct their own axioms, so what might follow “logically” for one individual may not make sense to someone else. Where Bloom assumes rationality — or the capacity for it across the board — I remain unconvinced. Even overall, there are some other psychological tests and thought experiments — like the Watson Selection Task or the Monty Hall Problem among many others — that suggest many people do not have a great understanding of numbers, statistics, or logic — all of which are necessary for cognitive empathy to serve as a pathway to morality.

Overall, I agree with Bloom: we should de-emphasize unchecked emotional empathy and instead try to use cognitive empathy as the pathway to moral decision making. My objections aside, I think Bloom is truly onto something. I’ve done my best to faithfully represent his position and make fair criticisms where I saw fit, but it was a constant challenge. I only hope that, if Bloom ever reads this, he can empathize.