3 ms·
> I think the naive story about MAD comes from applying game theory but assuming your strategy is static throughout the whole "game". Care to elaborate?
by netheril96 10y ago
> I think the naive story about MAD comes from applying game theory but assuming your strategy is static throughout the whole "game".
Care to elaborate?
- captainmuon 10y agoHere I'm using "game theory" in the pop-sci sense, no deeper mathematics. Assume you and your opponent commit to the following strategy: - If attacked by nuclear weapons, retaliate by completely destroying the enemy with nuclear weapons - Do everything you can to preserve second-strike ability Then it follows it is rational to not use nuclear weapons. So far so good and true, but there is one problem with this argument: if your enemy does use nuclear weapons on you for any reason, is it then still rational to stick to your strategy? Before the war, you do anything you can to cement your retaliation strategy, to preserve deterrence, MAD, and ultimately peace. But after the war starts, and once you realize you cannot win anymore, it does not make sense anymore to go through with the second strike. You have nothing to gain. You don't punish your enemy so he doesn't attack you again "next time". If your preparations in cementing your strategy haven't been 100% thorough, and you have a way to stop retaliation, then the logical thing would be to do so and surrender unconditionally (assuming you value preserving life over revenge, when there is nothing to gain militarily anymore). What I meant with "naive" is that the standard story about MAD assumes a steady state - both sides commit to a strategy, there is no time development, from this it follows that you should not use nuclear weapons ("the only winning move is not to play"). But if you allow the strategy to be a function of time, players can change their mind, and this changes the whole situation. You no longer have to ask, "what is rational before playing the game" (=> not playing), but "what is rational at every time t >= 0". I believe then it is no longer about being able to second-strike, but about being able to abort the second-strike. If your enemy is in any way able to abort the second-strike (and you are already in an escalated global war, and don't value sparing enemy lives over winning), then the most rational step at t > 0 becomes to nuke the enemy as soon as possible, so they can no longer achieve their goals. If they are rational, they should immediately proceed to surrender. I believe this "alternative story" is somewhat justified, as at least it was believed at some point by certain decision makers in both west and east militaries - I say that because there were many different nuclear strategies floating around during the cold war, and some were very different than the "Dr. Strangelove" scenario. There were ideas to fight a conventional tank war supported by nukes for example. Another rather spectacular one was (declassified one or two years ago I believe), if the soviets were to declare war and to occupy Berlin, to overkill Berlin with nuclear weapons (-> they can no longer achieve their goal, and you'd take out a significant portion of their tanks)... but of course we were never anywhere near a level of escalation where such a crazy, inhuman plan would have been considered.
- ThrustVectoring 10y ago>if your enemy does use nuclear weapons on you for any reason, is it then still rational to stick to your strategy? Yup. If you're the kind of agent that wouldn't stick to your strategy of retaliation, then that changes the best strategies for other players in a way that harms you. Interestingly, there's a significant amount of "removing the steering wheel" going on in terms of nuclear strategy. I remember reading that British nuclear submarine commanders would have a sealed envelope, to be opened and read only in case nuclear war destroys the British Government. It contains one of four orders - retaliate with nuclear weapons, do not retaliate, try to place yourself under American command, or to use your own judgement. The very interesting part is that it convinces others that nuking London might automatically mean that they get hit back, but does not obligate Britain to actually making nuclear war worse. https://en.wikipedia.org/wiki/Letters_of_last_resort https://en.wikipedia.org/wiki/Letters_of_last_resort