Being good at empathy means caring about effectiveness

Awhile back, a number of my effective altruist friends were sharing articles online about how empathy is a bad thing–because leads to people not making moral decisions with their heads. This struck me as wrong-headed. I think a better approach is to say that people need to learn to extend their empathy beyond their immediate circle of friends and family, to people they’re unlikely to ever meet in person.

I remember when I was very young, in grade school I think, and seeing a book that talked about extreme poverty. The number of people living on $1 or $2 per day. (I don’t remember what the statistic was then, but in 2013 it was over a billion people living on less than $1.25.) The book was one of those books with very thick (cardboard?) pages, so it had little bags of rice inside showing how much rice $1 or $2 would buy you.

In retrospect, the amounts of rice used might not have been quite accurate (or might have been per-meal rather than per-day), but I remember looking at the bags and thinking there was no way you could live on that little food. Maybe it would just barely keep you alive, but that was it. I could see that living in that kind of poverty wouldn’t be fun, and that at that level, just a little extra money each day could make your life a lot better.

Another example: one of the main reasons I’m vegan is that I’ve made the effort to try and imagine what it would be like to spend your entire life confined in a space too small for you to even turn around (as many farmed animals are). I can see that that would be horrible, not a fate I’d wish on anyone.

This is something I got thinking about again reading a recent post by Robin Hanson about criticisms of effective altruism. He notices that a lot of criticisms apply to charity generally, but only get trotted out when the topic is effective altruism. Therefore, he suspects something else is at work:

I think the key is the empathy signaling function. People who give because of emotional feelings induced by seeing or hearing those in need are seen as having friendlier and less suspect motives, and people who participate in a political process that includes those they help are also seen as treating them more as equals. In contrast, people with an abstract distant less emotional relation to those in need, whom they help directly as opposed to indirectly via politics, are seen as less having a personal-like relation to those they help, and so are more plausibly trying to dominate them, or to achieve some other less relational purpose.

This interpretation, that the main dislike about effective altruists is their less displaying empathy emotions, is also supported by two other criticisms made of Singer’s essay: two people complained that effective altruism relies too much on numbers and other abstractions, and two people complained that it can be very hard to estimate many numbers.

Imagine someone who said they were in love with you, cared about you, and wanted to live with you to help you, but who didn’t seem very emotionally engaged in this. They instead talked a lot about calculations they’d done on how you two could live your lives together well. You might suspect them of having ulterior motives, such as wanting to gain sex, money, or status from you. Maybe the same sort of thing is going on in charity. We want and expect a certain sort of emotional relation to people who help us, and to people who help the same people we help, and people who say they are trying to help but who won’t join in the usual emotions in the usual way may seem suspect. We’d be more likely to find fault with their approach, and to suspect them of bad ulterior motives.

These complaints may, to some extent, point to a way effective altruism could improve its PR. At the same time, many common criticisms of effective altruism look bizarre from the point of view of showing empathy. Critics of effective altruism would never treat a friend or family member the way they treat the people they allegedly want to help through their charitable activities and activism.

Imagine: if your family desperately needed money, would you refuse to try to find a good job, on the grounds that worrying about money is venal? Or if a friend needed help getting treatment for a life-threatening illness, would you avoid trying to find the most effective treatment, because that would be cold and calculating. If a friend got evicted and needed a place to stay, would you refuse to help because you couldn’t also give her a job, or because that would just be ameliorating the injustices of capitalism?

Robert Wiblin just wrote a post titled, Effective altruists love systematic change. But to the extent I’m pessimistic about systematic change, it’s because I see most people unwilling to widen their circle of empathy far enough. They care too little for the plight of distant strangers, and get upset about the Chinese dog meat festival while eating pigs. But if you want systematic change, maybe you should start trying to get people to feel ashamed at those failures of empathy.


10 thoughts on “Being good at empathy means caring about effectiveness

  1. I think you’re typical-minding pretty hard here. Anecdotally people seem to have a baseline for how empathetic /generally/ they are – how much the suffering of a person right in front of them affects them – and I think trying to move this baseline is a waste of time. People who are high-empathy could maybe benefit from trying to expand their empathy to care about more distant people, I can’t really speak to that since I’m not high-empathy. The EAs I know seem to mostly be people with low baselines, and I’m not sure that’s a coincidence – I /didn’t/ get involved in animal causes by thinking about what it would be like to live on a factory farm, I got involved by noticing that suffering is bad and that lots of it happens on factory farms.

    If I were to devote time to trying to be more empathetic, what good things would you expect to happen? Do you think that people who are more empathetic /in practice/ have a wider circle of moral concern, or just that getting people who are highly empathetic to expand their circle of moral concern by empathy is easier than getting such people to expand their circle of moral concern by abstract reasoning? If the latter, I think I agree with you – highly empathetic people, if they expand their circle of moral concern at all, will do it by empathizing more. But less empathetic people will do it by reasoning, and it’s often easier to help people reason than help them empathize, so the less empathetic people are better targets for effective altruism (and circulating memes that emphasize empathy is likely to be more detrimental to recruiting than you expect).

    On the other hand obviously if I’m talking to a empathetic person about EA I will try to talk about it using analogies and examples that activate their empathy, and I don’t think anyone has ever advocated not doing that.


    • This is an interesting point. I haven’t tended to think of myself as unusually high-empathy among EAs, but maybe I am. At any rate, I think there’s a lot of room to improve how EA is presented to people who are higher-empathy.


  2. > I remember when I was very young, in grade school I think, and seeing a book that talked about extreme poverty. The number of people living on $1 or $2 per day. (I don’t remember what the statistic was then, but in 2013 it was over a billion people living on less than $1.25.) The book was one of those books with very thick (cardboard?) pages, so it had little bags of rice inside showing how much rice $1 or $2 would buy you.

    Was it this book by any chance?


  3. I don’t think empathy actually works that way.

    Empathy actually takes a lot of its inputs from interactive features of other people – their body language, how they’re moving, how their face is variously set when they speak, their tone of voice. It’s just much easier to apply it to a person in front of you, than it is to apply it to your model of a person you’ll never meet.

    Besides, an important piece of empathy is your ability to notice your own surprise at someone’s precise behavior, which isn’t even possible when you’re trying to empathize with your mental model of someone.

    And, yeah, to the extent that I have a limited pool of effort, that it hurts to see the (comfortable, rich) people who I’m actually around hurting, even in subtle ways, makes it a lot more motivating to me to try to help *them*, rather than pouring my resources into whatever a utilitarian calculus would say it’s most useful.


  4. “But to the extent I’m pessimistic about systematic change, it’s because I see most people unwilling to widen their circle of empathy far enough.”

    That would have been the first argument I would advance in favor of systematic change. The basic EA proposal of donating 10% of your income requires a lot of empathy from everyone. Systematic change only requires empathy from enough people to vote on it, to coerce everyone else. You’ve also observed before that voting is a “cheap” form of altruism, and that sometimes it’s particularly easy to get people to vote against their own interests.

    Of course, there are major complications in extending this across national boundaries.


  5. A more comprehensive answer:

    I think this conversation will be more productive if we disentangle a few things. One is whether arguments from empathy or from reason are better at convincing people of ideas like effective altruism; that’s an empirical question. The case for arguments from empathy are that they’re clearly better at convincing people of things, in general. The case against is that we don’t have a /comparative/ advantage in arguments from empathy; our stories and case studies won’t be any more heart-stirring than the stories and pictures from the average charity. Whereas we have a strong comparative advantage in rational argument and can thus pick up people persuaded by rational argument pretty easily. (Consider the ways the Raising for Effective Giving people have moved a lot of money with masterfully not-your-typical-charity messaging).

    The other question is whether empathy is a learnable skill and, if so, whether it’s worth learning. Your title phrasing “being good at empathy” suggests to me that you see it as one, and the rest of the title “means caring about effectiveness” suggests that you see it as a skill worth learning; cultivating more empathy will make people more effective. I disagree on both counts. Like I said initially, empatheticness seems pretty innate, and I haven’t heard any accounts of someone /becoming more empathetic/ (as opposed to extending their empathy to more people). Even if it is learnable I’m skeptical that it’s useful at all. I think that if someone is already an EA, in particular, becoming more empathetic will make them sad more often but not more effective in any respect. It might make them better at converting people, but I doubt it is among the five most necessary skills for that.

    I also do not expect that, if empathy is learnable, teaching it to non-EAs would make them more effective or more altruistic.

    There’s a general heuristic that rationalists are weird and undervalue normalness and so should be skeptical of our own assumptions that skills normal people are better at are not useful. I endorse this heuristic and have worked accordingly hard to be as normal as is strategically useful (anyone who knew me in high school could testify that I’ve moved startlingly far in that direction) but I’m unconvinced that empathy is one such skill.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s