seer comments on Discussion of Slate Star Codex: "Extremism in Thought Experiments is No Vice" - Less Wrong

15 Post author: Artaxerxes 28 March 2015 09:17AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (110)
Tags: star slate codex

Comments (110)

You are viewing a single comment's thread. Show more comments above.

Comment author: gjm 30 March 2015 09:33:36AM 3 points [-]

Does anything need to?

I guess you're worried that if the same argument works in both cases then you might end up obliged to give Joe $1M. But those reasons why you should give Joe the money have exactly parallel reasons why you should keep it, and to zeroth order they all cancel out, so no such obligation.

If you look with a bit more detail, then the reasons might be stronger one way than the other; for instance, if you are quite rich and Joe is quite poor, he might benefit more from the money than you would. We don't generally have norms saying you should give him the money in this case for all sorts of good reasons, but instead we have taxation (compulsory) and charity (optional) which end up having an effect a bit like saying that rich people should give some of their money to much poorer people.

In typical cases, (1) if you give Joe a $1M then your loss will be bigger than Joe's gain, so even aside from other considerations you probably shouldn't, and (2) if you kill Joe then Joe's loss will be bigger than your gain, so even aside from other considerations you probably shouldn't. So the simple-minded "do whatever makes people happiest" principle (a.k.a. total utilitarianism, but you don't have to be a total utilitarian for this to be a reason, as opposed to the only possible reason, for doing something) gives the "right" answers in most cases.

Comment author: seer 31 March 2015 05:07:57AM 5 points [-]

I guess you're worried that if the same argument works in both cases then you might end up obliged to give Joe $1M.

No, I'm claiming neither Kindly nor you actually believe the argument you've given.

So the simple-minded "do whatever makes people happiest" principle (a.k.a. total utilitarianism, but you don't have to be a total utilitarian for this to be a reason, as opposed to the only possible reason, for doing something) gives the "right" answers in most cases.

Except, you're not doing that, i.e., you're not giving all your income to charity. So since you're willing to ignore parts of your ethics when its inconvenient, why not also ignore the parts about not killing Joe when it would be convenient were Joe to die.

Comment author: gjm 31 March 2015 09:01:31AM 4 points [-]

I'm claiming neither Kindly nor you actually believe the argument you've given.

Your overconfidence in your mind-reading abilities is noted.

Except you're not doing that [...]

The fact that someone doesn't act as a perfect utility maximizer doesn't mean that utility gains aren't worth seeking, for them out for others. If you ask "why did you buy that thing?" and I say I bought it because it was half the price of the alternative, am I refuted if you point out that I don't always buy the cheapest things I can?

As I said: a reason, not the only possible reason.

Comment author: seer 01 April 2015 02:59:48AM 4 points [-]

How do you distinguish the part of your ethics that you ignore in practice, e.g., not giving all your money to charity, from the part you insist you and everybody follow, e.g., not killing Joe even though he's being really really annoying.

Comment author: gjm 01 April 2015 10:03:31AM 2 points [-]

Giving all my money to charity isn't a part of my ethics.

Increasing net utility (or something of the kind) is one of the things I care about. So the fact that something increases net utility is a reason to do it, and the fact that something decreases net utility is a reason not to. But net utility isn't the only thing I care about, so a thing that increases net utility isn't necessarily a thing I think I should do.

What I insist on, though, is another matter again. That's a matter of Schelling points and traditions and the like, optimized (inter alia) for being easy to remember and intuitively plausible.

So:

  • Giving $1M to Joe: increases his utility, decreases mine, probably not a win overall in terms of net utility. Fails various other tests too. Not in any sense any sort of moral obligation.
  • Giving $100 to Joe, who is much poorer than me: net utility increase, might be a good thing to do on those terms. Probably reasonable not to do simply on the grounds that I care more about my own utility than that of strangers, that if I'm trying to do maximum good there are others who need the money much more than Joe, etc.
  • Giving $100 to a carefully chosen effective charity: close to the best thing I can do for net utility with the money. I still care more about my own utility than about strangers', though, so not necessarily obligatory even "internally".
  • Giving at least a few percent of one's income to effective charities, provided one is reasonably comfortable financially: almost always a big net utility gain, not too burdensome, has the same form as various traditional practices, easy to remember and to do. I'd be comfortable recommending this as a principle everyone should be following.

The attentive reader will notice that not killing people just for being annoying clearly fits into the same category as the last of those.

Powered by Reddit Powered by Reddit