Omelas is a cognitive exploit, yes. That's really the point - it forces people to consider how appropriate their heuristics really are. Some people would make Omelas if they could; some wouldn't, for the sake of the one child. A firm preference for either possibility can be controversial, partially because there are good reasons for both states and partially because different ethical heuristics get levered in different directions. (A heuristic that compares the number of people helped vs. the number hurt will pull one way; a heuristic that says "no torture" will pull the other way).

Aside from its obvious artificiality, and despite the fact that all our instincts cry out against it, it's not at all clear to me that there are any really good reasons to reject the Omelasian solution.

There's a real world analogue to Omelas. The UK (like other countries) has a child protection system, intended to minimize abuse & neglect of children. The state workers (health visitors, social workers, police officers, hospital staff, etc.) who play roles in the system can try to intervene when the apparent risk of harm to a child reaches some threshold.

If the threshold is too low, the system gratuitously interferes with families' autonomy, preventing them from living their lives in peace. If the threshold is too high, the system fails to do its job of preventing guardians from tormenting or killing their children. Realistically, a trade-off is inevitable, and under any politically feasible threshold "some children will die to preserve the freedom of others", as the sociologist Robert Dingwall put it. So the UK's child protection system takes the Omelasian route.

The real life situation is less black & white than Omleas, but it looks like the same basic trade-off in a non-artificial setting. I wonder whether people's intuitions about Omelas align with their intuitions about the real life child protection trade-off (and indeed whether both align with society's revealed collective preference).

Personally, on reading the story, I decided immediately that not only would I not walk away from Omelas (which solves nothing anyway,) I was fully in favor of the building of Omelases, provided that even more efficient methods of producing prosperity were not forthcoming.

The prevention of dust specks may vanish into nothing in my intuitive utility calculations, but it immediately hits me that a single tortured child is peanuts besides the cumulative mass of suffering that goes on in our world all the time. With a hundred thousand dollars or so to the right charity, you could probably prevent a lot more disutility than that of the tortured child. If for that money I could either save one child from that fate, or create a city like Omelas minus the sacrifice, then it seems obvious to me that creating the city is a better bargain.

That's true, but I think that human values are so complex that any attempt to compress morality into one sentence is pretty much obligated to be vague.

Vague or flawed. Given those options, I think I'd prefer vague.

One rather obvious rejoinder is that there are currently hundreds, if not thousands of children who are in the same state as the unfortunate Omelasian right now in real life, so reducing the number to just one child would be a huge improvement. But you are right that even one seems too many.

I agree completely. If I had any idea how Omelas worked, I might be tempted to try seeing if any of those ideas could be used to improve current societies.

A more robust possibility might be to add "equality" to the list of the "good things in life." If you do that then Omelas might be morally suboptimal because the vast inequality between the child and the rest of the inhabitants might overwhelm the achievement of the other positive values. Now, valuing equality for its own sake might add other problems, but these could probably be avoided if you were sufficiently precise and rigorous in defining equality.

Hmmm. To avoid Omelas, equality would have to be fairly heavily weighted; any finite weighting given to equality, however, will simply mean that Omelas is only possible given a sufficiently large population (by balancing the cost of the inequality with the extra happiness of the extra inhabitants).

Personally, I think that valuing equality itself is a good idea, if mixed in with a suitable set of other values; one possible failure mode for overvaluing equality is an equality of wretchedness, a state of "we're all equal because we all have nothing and no hope" (this is counteracted by providing suitable weights for the other good things, like happiness and the freedom to try to achieve goals (but what about the goal of vengeance? For a real slight? An imagined slight?)).

An example of a society failing due to an over-reliance on equality, is France shortly after the French Revolution.

I think the best explanation I've seem is something like the metaethics Eliezer espouses

I think that I should read through that entire sequence in the near future. I'd like to see his take on metaethics.

In other words, morality is like mathematics in some ways, it generates consistent answers(on the topic of people's wellbeing) that are objectively correct.

Huh. I think we're defining 'morality' slightly differently here.

My definition of 'morality' would be 'a set of rules, decided by some system, such that one can feed in a given action and (usually) get out whether that action was a good or a bad action'. Implicit in that definition is the idea that two people may disagree on what those rules actually are - that there might be better or worse moralities, and that therefore the answers given by a randomly chosen morality need not be objectively correct.

To take an example; certain ancient cultures may have had the belief that human sacrifice was necessary, on Midwinter's Day, to persuade summer to come back and let the crops grow. In such a culture, strapping someone down and killing them in a particularly painful way may have been considered the right thing to do; and a member of that society would argue for it on the basis that the tribe needs the crops to grow next year (and, if selected, might even walk up voluntarily to be killed). In his morality, these annual deaths are a good thing, because they make the crops grow; in my morality, these annual deaths are a bad thing, and moreover, they don't make the crops grow.

For what it's worth, I do agree with you that getting the result out of a moral system does not, in and of itself, force an intrinsic motivation to follow that course of action; people can be trained from a young age to feel that motivation, and many people are, but there's really no reason to assume that it is always there.

If there is an objectively correct morality, that can apply to all situations, then I don't know what it is - my current system of morality (based heavily on the biblical principal of 'Love thy neighbour') covers many situations, but is not good at the average villain's sadistic choice (where I can save the lives of group A or group B but not both)

There are some bad things in the dark corners of the human psyche.

That's certainly true, but that simply means that humans are capably of caring about other things besides morality, and these other things that people sometimes care about can be pretty bad.

Hmmm. A lot of the darkness in the human psyche can be explained in this manner; but I'd think that there are other parts which cannot be explained in this way (when a person goes out of his way to hurt someone that he'll likely never see again, for example). A lot of these, I'd think, are attributable to a lack of empathy; a person who sees other people as non-people (or as Not True People, for some self-including definition of True People).

If you're going to do that, why not just directly use happiness and fulfillment?

I cannot create an entire ethical framework, for everyone to follow, on any basis, and expect that it will be able to hold up for the next thousand years. If I try, I will fail, and this is why: because people cheat. Many intelligent agents will poke at the rules, seeking a possible exploit thereof that enhances their success at the possible expense of their neighbours' success. Over the next thousand years, there will be thousands, probably millions, of such intelligent agents hunting for, and attempting to exploit, flaws in the system; people who stick by the letter of the rule, and avoid the spirit of the rule. I cannot create an entire ethical framework, because I cannot outwit thousands or millions of future peoples' attempts to find and exploit gaps and loopholes in my framework.

Hence, the best that I can do is to find a system that has already endured a period of field testing and that hasn't broken yet; and perhaps attempt a small, incremental improvement (no more) in order to test that improvement.

There are many properties homosexuality has but 11-pebble heap preference don't, and vice versa. Why is evolutionary maladaptiveness worth pointing out, is my question.

Well, if moral norms are the Nash equilibria that result from actual historical bargaining situations (that are determined largely by human nature and the ancestral environment), then it seems somewhat relevant. If moral norms are actually imperative sentences uttered by God, then it seems completely irrelevant. Etc...

I suppose whether or not the pebble-sorting metaphor is good depends on which meta-ethical theory is true. In other words, I'm agreeing with PhilGoetz; Example 2 and Example 3 are only in separate classes of meta-wants assuming a (far from universally shared) moral system.

That's why I was careful to refer to them as 11-Pebble and 13-Pebble Favorers. They do value other sizes of pebble heaps, 11 and 13 are just the numbers they do most frequently. Or perhaps 11 and 13 are the heaps they like making in their personal time, but they like larger prime numbers for social pebble-sorting endeavors. The point is, I said they "favored" that size because I wanted to make sure that the ease of sorting the piles didn't seem too relevant, since that would distract away from the central metaphor.