RE: NANO: Institutional Safety

Dan Fabulich (daniel.fabulich@yale.edu)
Tue, 16 Nov 1999 17:59:45 -0500 (EST)

'What is your name?' 'hal@finney.org.' 'IT DOESN'T MATTER WHAT YOUR NAME IS!!!':

> I thought it was generally agreed that survivability was a stabilizing
> trait, not a destabilizing one.

Survivability is different from opportunity to retaliate, which is what you're describing.

Low survivability: When opportunity to retaliate is high, and survivability is low, you get mutually assured destruction. When opportunity to retaliate is low and survivability is low, first strike becomes effective.

High survivability: However, if opportunity to retaliate is high and survivability is high, war may erupt, rather than mutually assured destruction. High survivability with low opportunity to retaliate doesn't make any sense.

Thus, (surprise!) having a low survivability is a necessary but not a sufficient condition for nuclear stability in a MAD world. To get MAD stability you need low survivability AND ample opportunity to retaliate.

-Dan

-unless you love someone-
-nothing else makes any sense-

e.e. cummings