[extropy-chat] Superrationality, and the Seeking of Mutual Accomodations

Jef Allbright jef at jefallbright.net
Wed Dec 20 02:04:18 UTC 2006

Lee Corbin wrote:

> I'm less certain. Sometimes even an individual at war with 
> his own worst personal enemy has an agent (an element of the 
> society of mind) that can look upon the situation with 
> objectivity, and see the trap.  What is needed at that point 
> is nothing less than
>                    S U P E R R A T I O N A L I T Y
> Yes, our old friend.  The pairs of corn-farming tribes, or 
> the squabbling city-states of ancient Greece, or any battling 
> groups that merely drive themselves to exhaustion do need to 
> rise to a higher level and mutually evade the trap, or arms 
> race, into which they've fallen.  (The Greeks are a great 
> example, because if they were not squabbling and divided when 
> facing the Macedonians, then they were squabbling and divided 
> when facing the Romans.  18th century Poland is another 
> choice example that exhibits a tribe who could not for the 
> life of them unite against common enemies.)
> But attaining superrationality is easier said than done, 
> however, which was the point of my previous post!

Thanks for bring it up Lee. It was the elephant in the room, at least
for me.


> After my initial enthusiasm in 1983 for Hofstadter's 
> superrationality wore off after a couple of years, it became 
> obvious what the problem was.  In the non-iterated prisoner's 
> dilemma (NIPD), one can (should) mathematically cooperate 
> only with one's exact duplicate (or mirror image). Any other 
> strategy simply results in a lower payoff.  Period.
> It's a fatal assumption to suppose that your adversary is 
> like yourself.
> Here it seems to me that sensible people (e.g. people on the Extropian
> list) can see that neither extreme is optimal:   it is 
> non-optimal to be so
> resolute in your contests with the Other that negotiation and 
> compromise are impossible (e.g. Moslem extremists), and it is 
> also non-optimal to be naive enough to not see that sometimes 
> battles must be fought to the bitter end (e.g. the West).  

Yes, exactly, and this can be easily seen by realizing that in the real
world we are always playing multiple games simultaneously.  A
one-dimensional strategy is only going to be effective (at best) with a
one-dimensional game.  This is an example of the kind of valuable
principles of effective interaction that come out of this thinking.

And the arrow of progress continues to point *toward* the beacon of

- Jef

More information about the extropy-chat mailing list