My assumption is that IR is generally a field more akin to philosophy than to, say, country studies (or history, or whatever). We don't look to philosophers to find out what to do about Al Quaida, or Bosnia, or XX, why would we look to IR scholars? (frankly, this strikes me as a bit strange, but it seems to be about right...).Two things leapt to mind for me pretty much right away:
1) looking to IR for specific policy advice would be like looking to physics to find out how to build a bridge. Physics is a self-consistent, deductively rigorous portrait of the physical world that permits us to explain phenomena in a distinct manner. But it doesn't tell you how to build a bridge. For that, you want engineering and architecture, fields that take some of the insights into the world that are provided by physics and combine them with a lot of empirical practice to produce phronesis -- practical wisdom -- which can directly guide action. Theorists rarely develop phronesis, and they rarely even inform it directly. Instead, their influence is at a further remove than that -- although it is there, especially in the long term.
2) what strikes Steve as "strange" strikes me as perfectly comprehensible. What strikes me as more than a bit strange is that anyone would actually look to IR theory to find out what to do about Bosnia or Iraq or anything else. This doesn't mean that I don't think that policymakers should consult theorists and examine social-scientific accounts of things, but it does mean that I am wholly unconvinced that doing so will lead either to policy success or to an unequivocal justification for action. As "the slow boring of hard boards," politics is all about trying to bring incommensurate things together, and proceeds via the temporary forging of "good-enough" settlements. Science isn't like that, theory isn't like that -- and to the extent that IR is a social science, IR isn't like that either.
Engineers have to take physics. Practitioners should have to take theory courses, and politicians should draw on the insights that they have gained from scientific accounts. But in the end, practice necessarily escapes (abstract) theory, social life necessarily escapes our social-scientific accounts of it, and an engineer who pleads that the math worked out properly when the bridge collapses should probably be fired for not having consulted sufficiently with the folks who build actual bridges.
That's my reaction, at any rate.