f

Occasionally thinking out loud.

[f @7ip7ap]

@jensenjeans Thanks, it was a good read.

[f @7ip7ap]

@robinhanson That would also work!

[f @7ip7ap]

@robinhanson how much money would we need to raise to convince you to start writing a book on futarchy? Ballpark.

[f @7ip7ap]

@robinhanson I doubt I'd be able to get people to put together that much.

How far could $50k or $100k go?

[f @7ip7ap]

@tenobrus lol I almost got the same

[f @7ip7ap]

@DellAnnaLuca @Porkchop_EXP You should probably look up adoption studies

[Jack @tracewoodgrains]

A lot of times authors of prominent books experience a sort of reverse causation effect, where they crystallize a sentiment that was already ambient in their milieu and, having made the milieu visible, get credit or blame for causing the milieu.

[f @7ip7ap]

@nosilverv The correlation is positive but I don't think it's particularly strong, and directionally we could use less lookism in most places imo.

People will generally try to convince you of whatever they find most convenient.

[f @7ip7ap]

@mishapathy Not sure about normal, but @NathanpmYoung strikes me as so well-adjusted that he could even be considered excessively so, but it's one of my favorites

[f @7ip7ap]

@reconfigurthing Seems interesting. Followed!

[f @7ip7ap]

I feel quite confident that your model is wrong, and that you wouldn't buy much extra stability if you impartially selected for desirable traits first, and the association between different outward appearances and underlying important traits became very weak.

Or at least, I believe that under those conditions it would be possible and not particularly difficult to implement a system that would be robust and fault-tolerant to that kind of instability trigger.

I think this is a very important crux that would be worth creating civil, noninflammatory discourse over in the right places, in order to get different smart people to chime in and provide arguments and evidence that could update us one way or the other

I think it would be a pretty big deal for you too, and you too would want to know, if there was less reason than you think to believe that you necessarily must sacrifice a lot of stability in exchange for heterogeneity and everything that buys you, no?

[f @7ip7ap]

@AndyMasley would you consider it worth your time to offer AMAs in subreddits overwhelmingly frequented by users who think that the AI water emergency is real?

[f @7ip7ap]

@RokoMijic Whitopia doesnt have the antibodies to be safe from feminists either, if that's your concern. In fact, you'd be safer from that in a muslim country

We were talking specifically about protecting against the destabilizing effects of variation in appearance, though. Different beast

[f @7ip7ap]

I agree high IQ is not enough on its own, but also 98 avg IQ for example means that culture needs to propagate itself through shitty leaky easily hijackable memetically fit heuristics for retards. I think if the average american was as smart as, say, a LW nerd, the situation wouldn't have spiraled into BLM.

But I specifically flagged the OTHER factor with *more importantly*

[f @7ip7ap]

98 average IQ is very much not highly selected for IQ in terms of what could be possible today or soon. Similarly for all other traits.

More importantly, the differences between groups in the West run much deeper than appearance-deep - unlike in the scenario we're discussing. Come on man. That's the entire premise!

[f @7ip7ap]

@RokoMijic I think you agree that "high trust societies" generally do work pretty well?

The part you seem to disagree with is whether they can ever, under any circumstances, be made resilient to physically variable appearances

[f @7ip7ap]

@RokoMijic Lol the average intelligence in the US is mediocre, and there was no deliberate strong selection for the desirable traits I'm talking about. Completely different situation.

[f @7ip7ap]

I think your model is just wrong.

I think that with high enough ambient levels of broadly desirable traits, and in the culture that can emerge under those conditions, playing racial strategies would just be -EV, if it even crossed a few of the less honest members' minds. It couldn't spiral to systemically destabilizing levels.

[f @7ip7ap]

I think there's very desirable prosocial personality traits you'd want to strongly select for in a fit society anyway that include ~honesty and ~impartiality, that would already take care of those concerns.

(I think they correlate with other desirable traits too, and you'd end up picking them up even without going out of your way to)

Stats you have posted yourself also show that whites already show little to no racial ingroup preference, in heterogeneous societies. So it's possible. You can probably select for inclinations compatible with colorblind civnat culture, and unlock a much much wider pool to select from with no destabilizing side effects.

As for intelligence being destabilizing, the empirical evidence seems to show the opposite.

[f @7ip7ap]

Yea that's a crux for me, I think 1) the political stability gains would be small at best once you had already strongly selected for desirable traits, 2) you'd be severely under-optimizing for more important and evo fit traits in the presence of better discrimination criteria, 3) if self-selection played a role, you'd attract lower human capital with these criteria as your manifest

[f @7ip7ap]

Yes, but it won't cut neatly along racial cluster lines, even if desirable combinations of traits were disproportionately present in certain races.
(Not to mention that part of what determines long-term desirability for group membership is plausibly not strictly biological, either)

If you were to assemble your own (Roko's) 50 million people national dream team to associate with, and you could select along the traits you find most terminally important, I think you'd end up with a very substantial fraction of non‐NW-europeans.