Thursday, February 23, 2017

Book Review: Against Empathy

Against Empathy: The Case for Rational Compassion
Paul Bloom
Psychology, philosophy

I think of myself as being not particularly empathic. I suspect that's a big reason why I don't enjoy mainstream literary fiction; I just don't care about the travails of ordinary people in ordinary situations. In consequence, I'm predisposed to like a book a title like this.

Surprise: I liked it. Against Empathy is, as you'd expect, thought-provoking. It raises important questions. Bloom's argument, in a nutshell, is that empathy is an emotion; that, like all emotions, it's subjective; that it misleads us as often as not; and that reasoned kindness is better than instinctive kindness. The empathic response of a KKK member, for example, is to favor a white over a black person. Even for those of us in the non-white-sheet-wearing-classes, empathy warps our decision making; that's why we don't do much to stop things like genocides in Rwanda and famines in Sudan.

On the other hand, I'm not convinced that Against Empathy is as strong or as well-thought-out as Paul Bloom thinks it is. First, he spends too much time qualifying his alleged position. Second, some of his arguments are, shall we say, open to refutation.

To begin with, to make his argument Bloom has to make a very precise definition of what he means by "empathy". He's talking, specifically, about the feeling that something that's happening to someone else is happening to you. If you cringe and whimper when someone else describes his painful root canal, that's Bloomian empathy.

Bloom argues for the distinction by separating empathy from sympathy and/or compassion. If a small child is terrified of the sound of thunder (he would say), I am not myself terrified--I do not feel empathy--but that doesn't mean I can't be sympathetic.

As a person with a scientific background, I applaud the desire to define terms exactly. This particular hair, however, is being split exceedingly finely. Maybe I'm not afraid of the thunder, but that doesn't mean I've never been irrationally terrified of something. The echo of that fear is what I feel, and it's that echo that brings me to feel compassion. The difference--arguably, at least--is of degree rather than of kind.

Then there's the imprecision in Bloom's term "rational compassion." A truly rational person would not be compassionate. To act in a genuinely compassionate manner is to do something that is against your own self-interest. This is never rational. The motivations that Bloom cites for behavior that he admires are, again, just watered-down versions of the empathy he's arguing against.

Finally, there's a double standard going on here. Bloom makes both of these arguments

  • Empathy is something that we humans don't do well. We shouldn't rely on it.
  • Reasoning is something that we humans don't do well. We should strive to do it better.
I'll give Bloom credit for at least recognizing the problem. His final chapter tries to get to grips with it--by, among other things, arguing that our cognitive biases are neither as fixed nor as strong as they've been represented to be. The marvel, he asserts, is that we can think straight if we choose make the effort--and that when it comes to moral issues we must make the effort.

OK. Only . . . why bother? I'm the very last person to cast doubt on the merits of rationalism, but there's nothing that's obviously privileged about moral issues when it comes to thinking clearly. To assert that it's especially important that we overcome our innate illogic in the case of moral and ethical decisions is simply to say that Paul Bloom has emotively assigned a high value to moral and ethical decisions. I have to suppose that that's because he feels that these decisions are especially important (as, to be fair, I feel as well). You see where this is going, don't you?

Closely related:


  1. Pretty sure I see. You need empathy to even want to use reason to want to be rationally compassionate. This kind of brings us back to our Lechter vs. Alda discussion. You could argue that compassion is by its nature not rational.

    I would argue that people need more empathy rather than less. And then follow that up with reason. WIthout empathy, we continue to think in terms of us vs. them. The "other" isn't like us, doesn't have our values. So we can choose to think of "them" as less than. That's how you get hate groups, wars, Trump rallies.

    1. I do argue that compassion by its nature is not rational.

      Paul Bloom would disagree with your more-empathy-rather-than-less argument. He'd say (I think) that empathy is like anger, or any other strong emotion. It's useful in its place. It exists for a reason. A wise person wouldn't eliminate it--but a wise person would also not put it in the driver's seat and let it take over the decision process.

    2. Interesting, I wouldn't call empathy an emotion at all, much less a strong one. Instead, I'd call it simply the _capability_ to imagine the "other" as like yourself. From there can come strong emotions, but it isn't one in itself.

    3. And here is where we enter one of Bloom's definitional quibbles. He's talking very specifically about empathy as the I-feel-your-pain reaction. I bang my finger with a hammer, you wince and cringe: that's Bloomian empathy. What you're describing he would label as "cognitive empathy" or "sympathy".

      I don't say that his distinction isn't valuable. It lets us discuss with more precision whether emotive empathy is a reliable guide to action (no). But it's a fine distinction, and one that I think ultimately doesn't fully account for empathy as a continuum of reactions.