I want to emphatically endorse what my colleague Daniel Larison says here about the wrongheadedness of thinking about foreign policy in terms of rewarding friends and punishing enemies. It’s wrongheaded for all the reasons that Larison articulates, but it wouldn’t even work as a guide to relations between individuals. One of your neighbors is your friend, so you lend him money on a regular basis and never punish him by demanding he pay you back. Your other neighbor and you have a dispute about where the property line is, so he’s your enemy; you punish him by knocking over his garbage cans and refuse to mediate the property dispute outside of court. Who thinks this is the way to behave?

Of course, the better approach would be to say: the guy who keeps borrowing money and never paying you back seems to be taking advantage of your friendship – maybe he isn’t a friend anymore. (Or maybe he’s going through a really tough time financially, and he needs a more permanent solution than borrowing from you every other week – and if he’s a real friend, you should be able to approach him about the question.) And both you and your neighbor would benefit from settling the property dispute; it behooves you to determine whether a negotiated settlement is possible, and to incur the costs of going to court only if it’s clear it isn’t possible on any terms that you see as reasonable and if there’s a reasonable likelihood of victory in the courts.

The whole paradigm of “reward and punish” is derived from the game theory strategy of “tit for tat” which, indeed, reliably produces the best results in simulations. But those simulations are one-dimensional. The real world isn’t. India and the United States have common interests in fighting Islamist terrorism and in providing a strategic counterweight to China. But India has a fruitful relationship with Iran that they see no reason to sever. Should we “punish” them for that? How would we do that without also “punishing” them for being our allies against the Taliban? Should we have “punished” our ally, France, for not supporting our war in Iraq by not supporting their war in Libya? Or should we have supported our ally Britain for its staunch support in Iraq by joining the very same war against Libya? Should we have rewarded Russia for its support for our war in Afghanistan by dropping our support for Georgian membership in NATO? Or should we have rewarded Georgian support for the Iraq war by pushing harder for their membership in NATO?

That’s why when we talk about relations between states, we talk about common interests – indeed, the cliché line (paraphrasing Lord Palmerston) is that states have no permanent friends, only permanent interests. But even that isn’t true. Interests change with changes in circumstances. A China following an economic policy of autarky will have different interests from one following a policy of export-led development. An America that is a massive importer of oil will have different interests from an America that is a net exporter thereof. A West Germany divided from its Eastern half, and facing the massive Red Army poised to advance westward will have different interests from a united Germany protected from a reduced Russia by a buffer layer two states deep. Change, as Heraclitus said, is the only constant.

And I don’t see how anything is gained by saying that we should pretend nothing ever changes in the world for the sake of a purported friendship.