I agree that a rich person won't tolerate disposable products where more durable versions are available. Durability is a desirable thing, and people who can afford it will pay for it when it's an option.

But imagine a world where washing machines cost as much as they do in our world, but all washing machines inevitably break down after a couple years. Durable machines just aren't available.

Then, in that world, you have to be wealthier to maintain your washing-machine-owning status. People who couldn't afford to repurchase a machine every couple of years would learn to do without. But people who could afford it would consider it an acceptable cost of living in the style to which they have become accustomed.

Did your really need to say that you'd be brief? Wasn't it enough to say that you'd omit needless words? :)

It seems unlikely that joining a specific elite is terminally valuable as such, except to ephemeral subagents that were built for instrumental reasons to pursue it.

It seems quite likely that people seek to join whatever elite they can as a means to some more fundamental ends. Those of us who aren't driven to join the elite are probably satisfying our hunger to pursue those more fundamental ends in other ways.

For example, people might seek elite status in part to win security against bad fortune or against powerful enemies. But it might seem to you that there are other ways to be more secure against these things. It might even seem that being elite would leave you more exposed to such dangers.

For example, if you think that the main danger is unaligned AI, then you won't think of elite status as a safe haven, so you'll be less motivated to seek it. You'll find that sense of security in doing something else that seems to address that danger better.

I've played lot of role-playing games back in my day and often people write all kinds of things as flavour text. And none of it is meant to be taken literally.

This line gave me an important insight into how you were thinking.

The creators were thinking of it as a community trust-building exercise. But you thought that it was intended to be a role-playing game. So, for you, "cooperate" meant "make the game interesting and entertaining for everyone." That paints the risk of taking the site down in a very different light.

And if there was a particular goal, instead of us being supposed to decide for ourselves what the goal was, then maybe it would have made sense to have been clear about it?

But the "role-playing game" glasses that you were wearing would have (understandably) made such a statement look like "flavor text".

Typo: "And that's why the thingies you multiply probabilities by—the thingies that you use to weight uncertain outcomes in your imagination,"

Here, "probabilities" should be "utilities".

Trying the pill still makes you the kind of person who tries pills. Not trying really does avoid that.

You may be interpreting "signalling" in a more specific way than I intended. You might be thinking of the kind of signalling that is largely restricted to status jockeying in zero-sum status games.

But I was using "signaling tool" in a very general sense. I just mean that you can use the signaling tool to convey information, and that you and your intended recipients have common knowledge about what your signal means. In that way, it's basically just a piece of language.

As with any piece of language, the fact that it signals something does place restrictions on what you can do.

For example, you can't yell "FIRE!" unless you are prepared to deal with certain consequences. But if the utterance "FIRE!" had no meaning, you would be freer, in a sense, to say it. If the mood struck you, you could burst out with a loud shout of "FIRE!" without causing a big commotion and making a bunch of people really angry at you.

But you would also lack a convenient tool that reliably brings help when you need it. This is a case where I think that the value of the signal heavily outweighs the restrictions that the signal's existence places on your actions.

Good points.

I'm having a hard time separating this from the 'offense' argument that you're not including.

I agree that part of offense is just "what it feels like on the inside to anticipate diminished status".

Analogously, part of the pain of getting hit by a hammer is just "what it feels like on the inside to get hit by a hammer."

However, in both cases, neither the pain nor the offense is just passive internal information about an objective external state of affairs. They include such information, but they are more than that. In particular, in both cases, they are also what it feels like to execute a program designed by evolution to change the situation.

Pain, for example, is an inducement to stop any additional hammer blows and to see to the wounds already inflicted. More generally, pain is part of an active program that is interacting with the world, planning responses, anticipating reactions to those responses, and so on. And likewise with offense.

The premise of my distinction between "offense" and "diminished status" is this. I maintain that we can conceptually separate the initial and unavoidable diminished status from the potential future diminished status.

The potential future diminished status depends on how the offendee responds. The emotion of offense is heavily wrapped up in this potential future and in what kinds of responses will influence that future. For that reason, offense necessarily involves the kinds of recursive issues that Katja explores.

In the end, these recursive issues will have to be considered. (They are real, so they should be reflected in our theory in the end.) But it seems like it should be possible to see what initial harm, if any, occurs before the recursion kicks in.