And we must remember that Eliezer’s CEV depends on the supposition that there is no absolute morality, no basis for calling Dumbledore morally superior to Voldemort, or Gandalf morally superior to Sauron.
Do you expect that Eliezer would agree with what you just wrote? I personally would bet at 99:1 odds that he would disagree strongly with the second half (as the other commenters have pointed out).
If you think that he would disagree strongly, have you considered that it’s more likely you’ve misinterpreted his position (apparently Eliezer admits that the metaethics sequence isn’t his best-written material) than that he’s misinterpreted himself?
I personally would bet at 99:1 odds that he would disagree strongly with the second half (as the other commenters have pointed out).
You can’t have the first half without the second half. Either you are a moral absolutist, or a moral relativist. Moralities are things that can be ordered by an independent observer, or they aren’t. I don’t think there’s any middle ground. Saying “Gandalf is morally superior to Sauron when seen from a perspective similar to that of Gandalf; but Sauron is morally superior to Gandalf when seen from a perspective similar to that of Sauron”, is not saying that Gandalf is morally superior to Sauron. Claiming that they are the same thing is like saying, “I can believe that 3 is greater than 2 without admitting that there is any objective basis for comparing the magnitudes of numbers, and without denying that 2 may be greater than 3 to someone else.”
You can’t have the first half without the second half. Either you are a moral absolutist, or a moral relativist.
See, I’d give 99:1 odds that he’d strongly disagree with this as well (as do I). Now, your position is that you must be one or the other (if you’re to be coherent), but I hope you can admit of the possibility that Eliezer sees that as a false dichotomy. From your perspective, this makes his metaethics a hopeless muddle of absolutism and relativism, but this should give you different predictions about how MoR turns out than would the assumption that he’s a standard moral relativist.
I hope we can at least agree on that much, before we turn to arguing anything else.
I’ve just now given you an example where someone can be an absolutist on which criteria are to be used, but a relativist on the weight assigned to those criteria.
Dividing the world to pure absolutists (who can order each morality on a single axis) and pure relativists (who don’t order any moralities) is a very incomplete model.
Either you are a moral absolutist, or a moral relativist. Moralities are things that can be ordered by an independent observer, or they aren’t. I don’t think there’s any middle ground.
There appear to be different flavors of objective and subjective morality.
Moral Realism: There is objective truth, right, wrong; Moral Universalism: There is a morality that can be applied to a general class—for example all humans; Value Pluralism: There may be multiple equally correct and fundamental value systems that conflict with each other; Moral Consequentialism: Right and wrong can be determined from consequences; Moral Relativism: There is no objective truth—we should be equally tolerant of all systems; Moral Perspectivism: There is no objective truth—but some systems are better than others. Moral Nihilism: Morality is an illusion—nothing is moral or immoral.
This list could go on—but if you find it valuable to split belief systems between moral realism and not moral realism, then I don’t see how I could meaningfully object… :)
Do you expect that Eliezer would agree with what you just wrote? I personally would bet at 99:1 odds that he would disagree strongly with the second half (as the other commenters have pointed out).
If you think that he would disagree strongly, have you considered that it’s more likely you’ve misinterpreted his position (apparently Eliezer admits that the metaethics sequence isn’t his best-written material) than that he’s misinterpreted himself?
You can’t have the first half without the second half. Either you are a moral absolutist, or a moral relativist. Moralities are things that can be ordered by an independent observer, or they aren’t. I don’t think there’s any middle ground. Saying “Gandalf is morally superior to Sauron when seen from a perspective similar to that of Gandalf; but Sauron is morally superior to Gandalf when seen from a perspective similar to that of Sauron”, is not saying that Gandalf is morally superior to Sauron. Claiming that they are the same thing is like saying, “I can believe that 3 is greater than 2 without admitting that there is any objective basis for comparing the magnitudes of numbers, and without denying that 2 may be greater than 3 to someone else.”
See, I’d give 99:1 odds that he’d strongly disagree with this as well (as do I). Now, your position is that you must be one or the other (if you’re to be coherent), but I hope you can admit of the possibility that Eliezer sees that as a false dichotomy. From your perspective, this makes his metaethics a hopeless muddle of absolutism and relativism, but this should give you different predictions about how MoR turns out than would the assumption that he’s a standard moral relativist.
I hope we can at least agree on that much, before we turn to arguing anything else.
If I rated morality as (2, 3) on two separate orthogonal axes of goodness (e.g. freedom and joy) and another morality as (3, 2) on the same two axes, a morality that rates (4, 4) is superior to either, and a morality that rates (1, 1) is inferior to both, while I wouldn’t be able to absolutely “order” versus , unless I appropriately weighted those two values—and those weights I needn’t consider absolute constants of the universe, even if I considered the maximization of those qualities as good absolutely.
Note: The above is meant as an example only, not as a description of my own system of morality.
The problem of comparing things on the different axes—“solved” by converting to units of utility—is the same for absolutists and relativists.
I’ve just now given you an example where someone can be an absolutist on which criteria are to be used, but a relativist on the weight assigned to those criteria.
Dividing the world to pure absolutists (who can order each morality on a single axis) and pure relativists (who don’t order any moralities) is a very incomplete model.
There appear to be different flavors of objective and subjective morality.
Moral Realism: There is objective truth, right, wrong; Moral Universalism: There is a morality that can be applied to a general class—for example all humans; Value Pluralism: There may be multiple equally correct and fundamental value systems that conflict with each other; Moral Consequentialism: Right and wrong can be determined from consequences; Moral Relativism: There is no objective truth—we should be equally tolerant of all systems; Moral Perspectivism: There is no objective truth—but some systems are better than others. Moral Nihilism: Morality is an illusion—nothing is moral or immoral.
This list could go on—but if you find it valuable to split belief systems between moral realism and not moral realism, then I don’t see how I could meaningfully object… :)