It checks out empirically. War is a Nash semi-equilibrium—both sides would be better off if they could coordinate an end to it, but they usually can’t (in the near term). If Stanislav Petrov had been a bit less chill, the Cold War would have ended in 1983 in the “everybody launches all the nukes; cockroaches win” Nash equilibrium. If that’s not arbitrarily bad, it’s plenty bad enough.
I think “everybody launches all nukes” might not be a Nash Equilibrium.
We can argue that once one side launched their nukes the other side does not necessarily have an incentive to retaliate, given they won’t really care whether the enemy got nuked or not after they themselves are nuked, and they probably will have an incentive to not launch the nukes to prevent the “everybody dies” outcome, which can be argued to be negative for someone who is about to die.
It seems to me that both parties to the Cold War favored the defect-defect outcome (launch all the nukes) over the cooperate-defect outcome (we die, they don’t). It’s hard to tell, though, because both sides had an incentive to signal that preference regardless of the truth.
But that’s an extreme case. Any war you choose will have each side choosing between continuing to fight and surrendering. The cooperate-cooperate outcome (making peace in a way that approximates the likely outcome of a war) is probably best for all, but it’s hard to achieve in practice. And it seems to me that at least part of the problem is that, if one side chooses to cooperate (sue for peace and refrain from maximally fighting), they run the risk that the other side will continue to defect (fight) and seize an advantage.
It checks out empirically. War is a Nash semi-equilibrium—both sides would be better off if they could coordinate an end to it, but they usually can’t (in the near term). If Stanislav Petrov had been a bit less chill, the Cold War would have ended in 1983 in the “everybody launches all the nukes; cockroaches win” Nash equilibrium. If that’s not arbitrarily bad, it’s plenty bad enough.
I think “everybody launches all nukes” might not be a Nash Equilibrium.
We can argue that once one side launched their nukes the other side does not necessarily have an incentive to retaliate, given they won’t really care whether the enemy got nuked or not after they themselves are nuked, and they probably will have an incentive to not launch the nukes to prevent the “everybody dies” outcome, which can be argued to be negative for someone who is about to die.
It seems to me that both parties to the Cold War favored the defect-defect outcome (launch all the nukes) over the cooperate-defect outcome (we die, they don’t). It’s hard to tell, though, because both sides had an incentive to signal that preference regardless of the truth.
But that’s an extreme case. Any war you choose will have each side choosing between continuing to fight and surrendering. The cooperate-cooperate outcome (making peace in a way that approximates the likely outcome of a war) is probably best for all, but it’s hard to achieve in practice. And it seems to me that at least part of the problem is that, if one side chooses to cooperate (sue for peace and refrain from maximally fighting), they run the risk that the other side will continue to defect (fight) and seize an advantage.