POLITICAL SCIENCE AND POLITICS: Patrick Ruffini has a long rant in response to the DiIulio story that boils down the basic point that political scientists -- i.e., those with a Ph.D. in poli sci -- don't know squat about politics: "For an academic, it takes time to learn that more than 90% of politics is logistical and operational, that the day-to-day mechanics of government have precious little resemblance to a luncheons at the Brookings Institution, good for the soul as they may be.... You have to endure a few lectures of Poli Sci 1 to appreciate just how truly alien the academic study of politics is when stacked up against how politics and campaigns really work."

Is this fair? A full answer would require a much longer post; the short answer is yes and no. [What do you know about this?--ed. I'm a political scientist who did policy work in the government for a year, courtesy of the Council on Foreign Relations International Affairs Fellowship.] It is undeniably true that political scientists often crash and burn when they enter the policy world -- as seems to be the case with DiIulio. I don't have tenure yet, so there's no way in hell I'm going to name names. Is this because they didn't understand the way politics actually works? It might be safer to say it's because they don't understand the art of management, a point Franklin Foer made two years ago in the New Republic.

That said, Condoleezza Rice, another political scientist, seems to be thriving in this administration, even though many DC insiders predicted she'd get eaten alive by Rumsfield, Powell, Cheney et al. Political scientists per se are not congenitally incapable of prospering in government. And the tools of political science are a vital component of their success. [What about Ruffini's argument that history is more useful?--ed. They're both useful -- but woe is the man that relies only on history as a guide to DC. Many policymakers rely on theory to guide their decision-making, but the theory comes in the form of weak historical analogies that get them into trouble.]

I suspect the real difference between those political scientists that succeed in government and those that fail is that the successes know the limits of their trade. The most useful models of politics -- like the most useful models of any set of complex behaviors -- are abstracted from reality. The most capable political scientists know the proper limits of those models. They recognize that other sets of skills matter, skills that go way beyond social science. What those skills are, I'll get into in the next couple of weeks.

P.S.: This phenomenon is not unique to political scientists by any stretch of the imagination, nor does it apply only to Republican administrations. Joeseph Stiglitz and Laurence Summers were both distinguished economists who took reasonably high offices in the Clinton Administration. In 1992, If you were to predict which of them would do better, it would have been Stiglitz, since he was the more affable of the two of them, and Summers already had some bad blood with Al Gore. But Stiglitz crashed and burned, leading some considerable bitterness, as this Atlantic Monthly piece makes clear. Summers, in contrast, managed to thrive because he learned from his early mistakes, as David Plotz pointed out.