Can There Be Rational War?

If you've ever taken a course in international relations, you've probably had to read James Fearon's essay "Rationalist Explanations for War". In it, Fearon discusses whether two rational countries could ever go to war against the other. He says that because war is risky and costly, rational states always have an incentive to find a peaceful solution to any disagreement. If both sides agreed on the relative odds of victory and defeat, they should theoretically divide the issue along those likelihoods. If state A has an 80% chance of winning a war and state B has a 20% chance, state A should logically take 80% of whatever is being disputed and state B should take 20%. This, of course, requires assuming that there is issue divisibility or some secondary trade that states can make in order to effectively create issue divisibility (e.g. if state A and B both claim sovereignty over some patch of ground, state A could pay B some amount of money in order to take control of it.) However, if we concede that most issues can be made divisible through side payments, can war still be entered into by two rational sides?

I had to read Fearon's essay for a class last term and my in-class final essay was basically answering that question. I'm putting it on this blog because it has policy implications in the real world and might illuminate some ways America has and might again in the future end up at war for reasons other than the perfidious scheming of politicians of the other political stripe.

So, can war be entered into rationally? Fearon offers three possible ways war can be entered into rationally, all of which revolve around incentives toward private information. First, states have an obvious incentive to bluff. Picture Athens and Corinth on the verge of war. Let's say Athens is much stronger than Corinth and Corinth knows it. The King of Corinth tells the Athenian assembly that the Gods protect Corinth even though he has no idea whether or not it's true. The Athenians don't even have to entirely believe him, if they have to weigh even the remote possibility that the King is telling the truth then he will tip negotiations in his favor by some indeterminate amount. True, you could argue by pretending that it's stronger than it really is Corinth might make itself seem more of a threat to Athens and therefore bluffing might not be a good idea. However, that ignores two things: one, Corinth could bluff about defensive strength only (i.e. the Gods will strike down those who attack Corinth, but not if Corinth strikes first) and, second, if there's a large disparity in strength Athens will still not fear a Corinthian attack even if Corinth fudges its strength by bluffing.

Fearon's other two explanations are commitment problems and issue indivisibility. He thinks that the latter doesn't really exist since side payments are always possible. I disagree because not all civilizations have mutually recognized media of exchange. What about two oil-rich Muslim countries (one Shia, one Sunni) fighting over control of Mecca? Money doesn't work for the trade because they're both rich, they can't give away citizens or land because to do so would forefit their moral right to govern (to protect their citizens), and control is a zero-sum issue where the two could not share (by definition they can't both be simultaneously in control of the shrine and the other side's presence as a compromise would be blasphemous.) In that sense, issue indivisibility is rare but not impossible.

Commitment problems have two sides. First, a state's first strike advantage gives an incentive to start the fighting even for a side which has a strong chance of losing. Second, if one state is gaining power and the other state is losing it then there's no rational reason for either side to expect that a peace settlement will hold.

In addition to Fearon's possible rationalist explanations for war, there are a couple more. At a basic level, saying war is irrational depends on what you mean by irrational. It's practically impossible to actually know all the factors involved in establishing whether one side or the other will win a war. Who could have known before the American Civil War that the unquantifiably better generalship of the Confederacy would have kept the war going for four years despite all the material advantages of the Union? How could the French have known that the weather would be rainy, turning the field near Agincourt into a muddy mess that caused the great English victory? Since the costs of the war cannot be accurately predicted (and even the parts that can be predicted yield incentives to misrepresent), war can be rational if a country is in a dire enough situation.

A country with a large military could decide a risky war gamble is preferable to the certainty of slow national death. Like any other act of theft, if you get hungry enough, you'll probably start to consider it. North Korea, anyone?

So, if you ask me, yes, war can be theoretically rational. That doesn't mean that any given war is rational, but it should give pacifists food for thought beyond bumper sticker sloganeering.
 

This post is also available on the New Madisonian blog, where I'll be posting exclusively starting in nine days.