I moved this essay to my website: "Values Spreading is Often More Important than Extinction Risk." I'm still glad to discuss here, though.

Why not have it on both? You know what people are like with following links

I broadly agree (at least inasmuch as it's worth treating alternatives to X-risk as potentially comparable merit), though I'd suggest 'values spreading' is slightly too limited - you want to be doing (or at least comparing) something that increases the quality of life conditional on human survival. That could be spreading values, or it could be something mimetic/epidemiological, if you suspect for eg some particular happiness-causing behaviour has a basic reproduction number higher than 1.

Stories like this make me wonder if such behaviour isn’t all around us, in quite commonplace forms. Could the person who first came up with the concept of holding a door open for someone be a plausible candidate for highest aggregator of utility in history?

Arepo wrote:Why not have it on both? You know what people are like with following links

Haha, yes, we discussed this before. The reason is because I make edits, and I don't want to fork two branches. (Better to use a spoon.)

Arepo wrote:if you suspect for eg some particular happiness-causing behaviour has a basic reproduction number higher than 1.

The thing about utilitronium is that it's not widely seen as good, unlike holding doors or paying for someone else's coffee. It's possible that the best way to change values is something indirect, sure. That said, I can't think of anything indirect that seems like it would beat direct advocacy per dollar for encouraging utilitronium.

BTW, I'm all in favor of being nice, but this is for other reasons -- it's my personality, I like to be kind, it makes me happy, and it makes others like me in turn. These are orthogonal or instrumental considerations.

Brian Tomasik wrote:

Arepo wrote:if you suspect for eg some particular happiness-causing behaviour has a basic reproduction number higher than 1.

The thing about utilitronium is that it's not widely seen as good, unlike holding doors or paying for someone else's coffee. It's possible that the best way to change values is something indirect, sure. That said, I can't think of anything indirect that seems like it would beat direct advocacy per dollar for encouraging utilitronium.

I would taboo the term utilitronium when suggesting it, maybe except for rationalist audiences that are familiar with it and open to it (?)

We might also taboo utilitarianism when spreading values because it has all these unpleasant connotations, e.g. the repugnant conclusion - or should we call it the repugnant confusion?

This is from a discussion on Overcoming Bias:

komponisto wrote:Aggregate utilitarianism seems utterly insane to me, and essentially nobody acts as if they believe it. It’s one of those ethical theories that seem to imply that no one should ever be allowed to have any fun, ever.

What works slightly better is to represent utility with something that people can empathize with: Human minds doing things that they love, e.g. entertainment, sex etc. Then you only have a few more inferential distances to cross:

- Moore's law implies we could have the computing power to simulate human minds doing these things, on hardware as cheap and small as a USB stick.
- We could make as many of them as we are willing to afford.
- There's really no need for them to suffer.
- Their pleasures can be made better and they can have more interesting experiences, because it is all digital.

Of course, this isn't hedonium/utilitronium, but it is probably closest to what you can get across as a desirable idea for normal people.

Most people reject it anyway, but it could be a positive vision that may achieve some positive reinforcement, especially when it actually becomes feasible.

Hedonic Treader wrote:e.g. the repugnant conclusion - or should we call it the repugnant confusion?

Haha, I love it.

I don't mind using the word "utilitronium" because I'm not actually out to sell the idea myself. I'm more concerned about preventing awful experiences, and in my piece, I'm just explaining that if you also cared a lot about increasing pleasure, you should promote utilitronium (using a better name).

Hedonic Treader wrote:We might also taboo utilitarianism when spreading values because it has all these unpleasant connotations, e.g. the repugnant conclusion - or should we call it the repugnant confusion?

Do you say "repugnant confusion" because you don't accept the conclusion as being repugnant or unintuitive or for some other reason? I'm not sure I've come across many bullet-biters on this point. If there's something good written on it that you're aware of, I would appreciate a pointer!

jason wrote:Do you say "repugnant confusion" because you don't accept the conclusion as being repugnant or unintuitive or for some other reason?

Some people obviously just disagree with the conclusion and the underlying values. But I think others don't model the formal conditions for the conclusion correctly when imagining a scenario. In order for the conclusion to be valid, the additional people need to be happy enough to compensate for all their suffering (also counting externalities). But when people visualize a world where this is true, they instead visualize loads of miserable people stacked on top of each other, with high infant mortality, forced to breathe smog, high crime rates etc. And that is repugnant to them. But those are conditions in which it could easily be better to have less people with less misery. The quote I quoted ("no one should ever be allowed to have any fun, ever") is such a typical misinterpretation. It might be right if he had written "resource-inefficient fun", maybe, but even that wouldn't be correct since unfocused tinkering and exploring (for fun) can have positive unplannable consequences.

If there's something good written on it that you're aware of, I would appreciate a pointer!

The felicifia search should yield many good arguments, also here. You'll also probably find some discussions on Less Wrong and Overcoming Bias, if you're willing to dig and filter through posts and comments.

Thanks for the insight. I hadn't considered the issue related to visualization before.

All moral theories have implications that are at least as problematic as the repugnant conclusion is for total utilitarianism (see e.g. this paper by Blackorby, Bossert and Donaldson). So it is misleading to suggest that total utilitarians face a special problem here, relative to proponents of these other views.

jason wrote:Thanks for the insight. I hadn't considered the issue related to visualization before.

Ironic: On the popular Standford website about the repugnant conclusion, they even visualize the populations erroneously in their image ( at the time I write this post anyway ):



If you pay attention, you can see that A actually has more surface area than Z, thus violating the assumptions for the repugnant conclusion. Z should be depicted with a larger width to reflect a mildly happy but large-enough population.

Okay, they fixed it. Now Z looks more attractive and A less attractive. I think details like this can affect the underlying intuitions.

Pablo Stafforini wrote:All moral theories have implications that are at least as problematic as the repugnant conclusion is for total utilitarianism (see e.g. this paper by Blackorby, Bossert and Donaldson). So it is misleading to suggest that total utilitarians face a special problem here, relative to proponents of these other views.

Pablo, I started reading that a while ago, found it difficult reading for my level of mathematical competence, and was trying to persevere as well as I could when I hit a reference by them to an earlier paper implying that it was actually there that they'd demonstrated the claim that all populations ethics have this problem. What exactly is it that they purport to prove in this paper?