Mutually Assured Destruction

Here's the lowdown on the blow up. If you gave everyone an atomic bomb, the school bombings would be bigger.

We would never do that -- but we (the US) would, as a country quite concerned about our security, build enough atomic weapons to defend ourselves against EVERYONE.

And darn it, those tricky Russians decided that they would do just exactly the same thing!

So we invented a new idea, to justify a situation we really couldn't do anything about: Mutually Assured Destruction. We didn't dare try to bomb Russia, because we knew that Russia would respond quickly and thoroughly, destroying us. We hoped that Russia understood that we could also respond quickly with enough firepower to totally wipe them out. (As it turned out, they did realize this).

Sadly, governments can be just as immature and pathogenic as individuals. And there is no metagovernment yet, so the situation is bad. We really lucked out with Russia. Now we have to hope that all other the military-minded countries in the world either respect the logic of mutually assured destruction, or stay too weak to challenge the US.

The preceding write ups missed a few important points about MAD.

First, for MAD to work, both sides have to have amassed arsenals that are either impervious to retaliatory strikes (such as SLBM's or deeply buried and dispersed ICBMs) or such an overwhelming number of nuclear weapons, a first strike cannot possibly destroy them all.

Secondly, and this follows from the first, each side must have a credible second strike capability. In other words one side must have a reserve of weapons capable of surviving a first strike, and the other side must recognize and understand this fact. This closes the loop, so to speak, on the "mutually assured" part of MAD.

Finally, for MAD to be effective, one must have a credible and well known doctrine for the use of nuclear weapons. The Soviets and the United States, throughout the Cold War, made perfectly clear under what circumstances they would resort to nuclear weapons. Through public documents, highly visible military exercises, nuclear tests, and diplomacy both sides were able to make the threat of their nuclear capabilities credible.

[partially in reply to -brazil-'s writeup on MAD - Mutually Assured Destruction]

It has worked relatively well so far, but do you really want to bet the survival of the human race on the hope that governments always act rational?

You know, that's funny, as far as I can tell MAD is not very rational. Let's see:

Scene: Players A and B both have enough weapons of mass destruction to wipe the other party out (as US(SR)? did have) and the capability to detect the attack by the other party and launch a retaliatory strike before getting hit (as I think they did, at least if War Games had anything to do with reality ;) ). Now, if player A wants to take B out and decides to shoot them with all they got, B will detect this and shoot back. Both end up dead. Not good. Therefore, A won't shoot. Right?

Now, imagine being the leader of B in the situation where A just shot. I'm assuming you're rational (in the game theoretic sense) like you claim to be, and that you respect human life (like you claim to do). Therefore, you should not shoot. Regardless of whether you shoot A back, the only difference will be more dead. You can't save your own people in any case. Revenge? Revenge is only applicable when, after revenge, you'll be dead; threat of retaliation is the tool of coercion, and the actual act of revenge is only carried out so that your future threats will maintain credibility

(for example, I can say "I will beat you to a bloody pulp if you touch my M:tG cards again, punk", but if you do despite my threats, I only carry out my threat because if I didn't, you and the other kids would do it again. Beating people in itself is not nice.)

-- however, in the case of global thermonuclear war, there won't be anyone alive to fear or respect you. Therefore, once the nukes fly, the only rational course is to withheld your bombs and say your prayers. Now, if A is rational, he will realize this, and if he assumed you're rational too, the most rational course for him will be to nuke you back to middle ages (hmm, I didn't know Middle Ages had nuclear winter, lethal radiation levels and horribly deformed mutants...) at his convenience. And if you, B, are rational and assume the A to be rational too, you will do the same... only sooner than him. Iterate. Chaos ensues.

So what is wrong with this picture? Nothing, I think. And this is curiously analogical to Prisoner's Dilemma, though it's not exactly the same. So how to avoid the looming destruction? The key to this game would, it seems to me, to convince your opponent that you in fact would shoot back so (even if you wouldn't); this can happen if you can make him think you're not rational (did US(SR)?'s foreign policy seem terribly rational to you? see also The Eye in the Pyramid, pg. 27) or you don't respect all human life (to comment on this would inevitably hurt someone's feelings). Therefore, as a rational and responsible leader of the free world, you should appear to outside world as somewhat sensible but maybe a bit psychotic. Draw your own conclusions.

Well, you could also be mad yourself. Or honestly think the other party is, if you feel better thinking that way.

So the today's lesson is: rationality, indeed, is fatal.

An afterthought: as a rational and benevolent leader of A, I have already reached this conclusion. So, when I see B doing something that suggests he's not quite rational, I must take into account that it might be just a ruse suggested above. If I believe it is, my rational course is, again to nuke. So, from the viewpoint of people who just want to live, the safest course would be to get mad leaders.