Comment by yummyfajitas | Hacker News Reader

Comment by yummyfajitas | Hacker News Reader

Parent Full thread

[edit: the deleted comment I responded to was asking why I seem to have "contempt" (possibly the wrong word) for empathy.]

I am opposed to empathy as a moral tool. Empathy has all sorts of known biases - we feel more empathy for people like us, attractive people, etc. In the west, for example, we have less empathy for black people:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3108582/

http://www.plosone.org/article/info:doi/10.1371/journal.pone...

Furthermore, even in the absence of easily predicted biases, emotional reactions are simply nondeterministic. There is no reason to believe they will converge to anything useful.

In contrast, principle-based morality doesn't have this flaw. If you take a principled position that poking people with a needle is bad (to borrow the example of the first paper), then it doesn't matter if you are poking a black or white person - the abstract principle of "don't poke people with a needle" is sufficient.

I realize this idea certainly bothers people who take personal pride in feeling empathy, who make it part of their identity. That doesn't mean I believe it to be wrong.

>In contrast, principle-based morality doesn't have this flaw. If you take a principled position that poking people with a needle is bad (to borrow the example of the first paper), then it doesn't matter if you are poking a black or white person - the abstract principle of "don't poke people with a needle" is sufficient.

I'd rather live in a world where I risk being poked by someone who hates me than in one one were I'm spared not because of empathy or kindness but only because of some abstract "principle" of not poking people.

Empathy being flawed is a startling insight. But why then throw it out and go straight to abstractions?

What are the problems preventing empathy from being as good as the abstractions? Seems to me the whole shame vs guilt debate

I go to abstractions because I'm a product of the enlightenment - I believe that the universe is governed by mathematics and reason, and the best way to understand it is to use those tools. I realize this does make me a bit of a weirdo in our post-enlightenment world.

Using emotions instead of reason suffers all sorts of problems. The most easily characterized, and shocking to the mainstream, are things like empathy being racist and sexist.

But there are deeper problems - emotions are often just plain inconsistent. We will feel different things depending on the framing of a situation, our own mood, relative status of other people involved in the situation, etc. This also makes "cheating" easier. The core problem with empathy is simply that it's an appeal to emotional decisions over reason.

It doesn't make you a weirdo but you are throwing away the baby with the bathwater. I can't tell you how many times I've looked at a proof or a solution to a problem and thought to myself that it just feels wrong and ugly and that there should be a more elegant approach. And sure enough after a bit of digging and thinking there is a better and more elegant approach.

That feeling of ugliness is not formal or logical in any way. It is a visceral and emotional reaction and there is no specific set of formal reasons I can point to that gives rise to it. The history of science and mathematics is full of examples of such emotional insights so seems to me your theory of "enlightenment" is incomplete.

One shouldn't ignore intuition, empathy, or a gut feeling. It's a useful signal. But that's all it is. There are true results that go against emotion, intuition, and just feel wrong. In math, Banach Tarski is a perfect example.

http://en.wikipedia.org/wiki/Banach%E2%80%93Tarski_paradox

If you want to use empathy as a hint to keep digging for a better solution, be my guest. I just object to using empathy as a basis for morality. Similarly, mathematical intuition isn't a substitute for a proof.

>What are the problems preventing empathy from being as good as the abstractions?

Empathy is a product of evolution, but the environment we spent millions of thousands of years evolving in is quite different than the environment in which we now exist. I don't think it's a stretch to suggest that deliberate, logical consideration of the optimal manner to act in a particular situation could produce more effective results than following what's essentially just an artifact of a selection process focused on making monkeys have many healthy, fertile babies.

Reply on news.ycombinator.com