Improved model of the social climate where revolutions are much less startable or controllable by good actors. Having spent more time chewing on Nash equilibria, and realizing that the trap is *real* and can't be defied away even if it's very unpleasant.
-
-
Replying to @ESYudkowsky @zackmdavis
To a first approximation, you are always actually trapped. If you weren't actually trapped, you'd be seeing defiant teenagers being outside the trap *successfully* in a useful direction.
3 replies 0 retweets 9 likes -
Replying to @ESYudkowsky @zackmdavis
There really are people who don't vote for major party candidates (hi!). There are significantly fewer people, but they exist, who don't go to doctors, who emigrate from countries whose governments they disapprove of, or who evade taxes, though those are much more costly.
2 replies 0 retweets 4 likes -
Replying to @s_r_constantin @zackmdavis
And your candidates don't get elected. Trump got elected, but he's not a good person's candidate. The equilibrium can't be *successfully* broken in a *useful* direction. So your *best* options are the lesser evil acts inside it. The trap is real and cannot be wished away.
3 replies 0 retweets 21 likes -
Replying to @ESYudkowsky @zackmdavis
It's true that not nearly enough people do the things I mentioned to break the Nash equilibrium. The question is, are people really "trapped" in that they *have* to do the more common thing or else the personal consequences for them are dire? Not always.
2 replies 0 retweets 3 likes -
Replying to @s_r_constantin @zackmdavis
Demanding "personal" consequences is moving the goalposts; the question is whether a defiant action is *best*, not whether it has personal consequences. People who refused to vote for Hillary didn't pay the price, kids in cages did, but that still makes the action nonbest.
3 replies 0 retweets 10 likes -
Replying to @ESYudkowsky @zackmdavis
Your original point was "it would be immoral to vote for Hillary/pay taxes/trade with doctors/etc. if you weren't trapped by a bad Nash equilibrium." The counterargument I made is "you're not trapped, you totally can do those things." You said "but there are worse consequences...
1 reply 0 retweets 2 likes -
if you don't do those things." Okay, but that only holds force if you're following a rule like "you can't cause harm (eg increase probability of kids in cages) unless you can in expectation prevent even greater harm by doing so."
1 reply 0 retweets 1 like -
Basically this is act-utilitarianism, right? Or, like, "following a rigid rule regardless of immediate consequence is only justifiable if you can show that there actually are enough rule-followers to produce the better outcomes that would happen if everyone followed the rule"?
2 replies 0 retweets 1 like -
I think "trapped in a bad Nash equilibrium" implies the reason people do the suboptimal thing is that they are *dissuaded* by the bad consequences of doing some other thing unilaterally. So...we have a 2-party system because there are *too many* people worried about harm?
1 reply 1 retweet 2 likes
That doesn't sound credible to me. There are plenty of callous people in the US! Certainly enough to form a third party. They don't, of course, but not because they're scrupulous utilitarians.
-
-
Replying to @s_r_constantin @zackmdavis
I can't seem to follow your line of argument over Twitter. Are you doing something with deontology and actions being forbidden? My argument is meant to simplify to agents that are just assigning cardinal utilities to everything, and don't distinguish "harms" or whatever.
1 reply 0 retweets 0 likes -
My argument isn't meant to distinguish selfish and altruistic consequences either. A trap is effective when agents, agents no more or less coordinated than they actually are, assign their highest utilities (from whatever source derived) to actions inside the trap.
0 replies 0 retweets 2 likes
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.