A few game theoretic thoughts on Ukraine
Many smart people have decomposed the causes of the current Ukraine crisis. There is the realpolitik view that NATO sort of encouraged it all to happen by expanding eastward, and this is contrasted with the immediate fact that Putin chose to invade. Despite humans ‘studying’ ethics since time immemorial, we still don’t know how to assign blame down the causal tree. In any event, I don’t either, and at this point it doesn’t particularly matter.
What I do think is crazy, is many Americans are now jazzed up, and want to take active military involvement in the Ukraine. The common phrase is asking for a no-fly zone.
I suspect most people don’t know what a no-fly zone is, and don’t realize you actually need to shoot down Russian planes to enforce it. Or maybe some do, and boomers are actually super fucking hardcore from having lived through, and won, the cold war. It is interesting though watching normie-core libs agitate for escalating in a game of nuclear brinksmanship.
But the craziness aside, it’s actually stupid, as is it defeats the purpose of NATO. Specifically: NATO is about securing the safety and interests of its members by communicating a common military alliance, such that it’s common knowledge for all actors. The purpose of military alliances is defeated, if you then go on to defend countries not part of your alliance. This is for two reasons, one obvious, and one insidious.
You want belligerent actors to know (and know you know) who they can, and cannot, attack. Predictability allows rational actors to game situations into the future, and reduces uncertainty. If, for example, we let Ukraine into NATO right now, it would defeat the purpose of the alliance, since external countries are no longer able to reason about our response to their actions, since they’ll never know who we will defend and why. (and why do we care so much about them being able to reason about our response? well, because they have ICBMs).
If countries not part of your alliance think that you might defend them anyway, they may choose to fight in the hopes you help them, rather than surrendering. In Syria we kept intimating that we might help the Free Syrian Army, and then didn’t. I remember reading the Syrian civil war reddit at the time, and there was a ton of hopium that the US would join the efforts on the ground in Syria. We don’t know the counter-factual, but what we do know is that tagging a country along with the hope that we could possibly help them is a bad idea.
If you think the calvary may come, you’re more likely to hold the line. If the cavalry doesn’t come, then you held the line for nothing. I do worry there was an element of that in Ukraine. Up until a week before the invasion, Zelenskyy was affirming Ukraine’s desire to eventually join NATO. This means that Ukraine had some expectation of potential future protection. The implication of this is that they are actually in much more danger today, since if you have a gripe with Ukraine, you should take care of it before they go under NATOs protection. Now, it’s possible Putin’s grand narrative of uniting Russia renders this moot.
But it’s worth paying attention to the rationality of Putin’s actions. As a rational actor, he attempted to take over Ukraine prior to it potentially joining NATO, when the cost is lowest. Once it joined NATO, he couldn’t take it over, and it would be in an alliance with a group he considers his enemy. I’m not really offering my opinion here, this is just the basic model.
Another question I’ve been mulling over: given that Ukraine isn’t in NATO, what’s the right framework to get involved with the War in Ukraine (as opposed to providing aid and admitting refugees)? I don’t think it should be that we deeply care about the Ukranians suffering.
The baseline suffering in the world is very high, and we mostly ignore it. About 13,000 Americans have died of covid in the nine days since this war began. About 92,000 Americans died of drug overdoses in 2021. And unlike covid, these addicts are young. I could go on about wars in Syria, Yemen, etc.
My point here isn’t some logical game where I say bad things happen all the time, so we should let them happen. Nor is it to say we need to isolate our country until we solve every problem at home.
It’s instead to point out we need to calibrate the immense suffering in Ukraine, such that our actions are commensurate with how we respond to other events, the expected benefit, and expected cost. Again, specifically, it means we need to act in our own interests. There is plenty of suffering we cannot prevent in this world, and that we already do not prevent.
I’ll even be antagonistic, and say that part of the reason it feels different is because it’s more visceral to shoot down a Russian bomber, than solve drug addiction. And my point here isn’t to go full Tucker Carlson, I’m instead saying the reasons we should act should be to fulfill our own geopolitical interests and safety.
It could be that there is a great geopolitical argument that shows that maintaining sovereign borders maintains world stability, and that is good for our interests, so we need to get involved to some degree. And to a certain extent I agree, we are providing lethal weapons, so that’s sort of being involved, right? I don’t buy that we should set up no-fly zones, but either way, make the argument for our interests. Not from moralizing about suffering.