Re: Bound unhappiness below (was Re: What if there's an expiration date?)

From: Matt Mahoney (
Date: Thu May 08 2008 - 14:25:48 MDT

--- Stuart Armstrong <> wrote:

> > You are confusing collective utility with ethics.
> Possibly. I'm seeing ethics as the utility you would want for all of
> humanity.
> > We have already seen
> > the dangers of maximizing total human utility vs. average utility.
> (One
> > leads to a very large population, the other very small). Weighting
> > utility on a nonlinear scale to fit your ethical model will just lead
> to
> > other disasters. (I will leave it to others to come up with examples
> if you can't think of any yourself).
> Yes, but so far every model I've seen leads to a disaster of some
> sort. I feel this model avoids a lot of the disasters I've seen,
> without introducing too many new ones. Can you think of particularly
> intolerable outcomes for such a weighting?

You originally proposed giving greater (perhaps infinite) weight to
negative utility than positive utility. By this reasoning, you could
never test any new drugs because of the possibility they may cause harm.
You could not engage in trade because the negative utility of the payer
exceeds the positive utility of the recipient. I know you could make
exceptions in these cases, but there are many other examples.

-- Matt Mahoney,

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:02 MDT