I'm in Seattle at the annual meeting of the American Political Science Association, which is why I haven't been blogging. The roundtable on John Ikenberry's book Liberal Leviathan went well, I thought, and John was a good sport for allowing four critics (William Wohlforth, David Lake, Charles Kupchan, and myself) to direct a lot of good-natured critical fire at him.
I won't try to summarize the discussion (in which many good points were made), but I will say that the back and forth helped crystallize some of my own impressions of the book and its relationship to IR theory. Here I'll just make two quick points.
First, like most liberal IR theorists, John makes much of the fact that the American-led order is "rule-based." Indeed, that is how he characterizes the entire post World War II liberal order: it was a "rule-based" system that worked because the leading power (the United States) agreed to bind itself within a framework of rules and institutions. (Never mind that we mostly made up the rules, and chucked them or ignored them when they got in our way). Thus, for John the current world order is defined by its "rule-based" character; those binding norms and institutions are a central constitutive feature of the current system.
Realists acknowledge that there are rules, but we don't define the system in this way. For starters, defining the system as "rule-based" doesn't make much sense if the leading state(s) can ignore the rules whenever they want to. Instead, realists would say the system is defined by power and by interests, with the latter heavily shaped though not absolutely determined by the former. Powerful states use rules to pursue their interests (a point that Ikenberry concedes), but the critical difference is that powerful states also ignore the rules when they get in the way, and especially in security affairs. For realists, however, it is a mistake to conceive of the entire international system as "rule-based" because any rules that states do create have little binding character and are just instrumental tools that they use mostly to overcome various coordination and credibility issues.
Second, it became clear to me in preparing my own comments that the theory advanced in Liberal Leviathan is not what social scientists would call a "positive" or "explanatory" theory. It does not in fact explain how states behave, because there are just too many ways that behavior of major powers (especially the United States) depart from the book's core arguments. Instead, Liberal Leviathan is a normative or "prescriptive" theory: it prescribes how states should behave if they want to reap the various benefits that John believes can be achieved by maintaining and following a rule-based liberal order. There's nothing wrong with that type of argument, by the way; lots of good political science is essentially normative in character.
Ironically, I agree with a lot of the book's specific prescriptions (though I think he's too optimistic about some of them), and I think the world would be nicer if states acted in the ways he recommends. The problem, as I said on the panel, is that the people in charge of U.S. foreign policy (both Democrats and Republicans) don't seem to agree with him. They devote a lot of lip-service to law and norms and rules and multilateralism, but when push comes to shove (which happens surprisingly often), they go their own way.
And as a last point: Despite my disagreements with Ikenberry's arguments, it is an ambitious and thoughtful book and he deserves abundant credit for putting the argument out there and letting us come after him.
Stephen M. Walt is the Robert and Renée Belfer professor of international relations at Harvard University.