As a liberal and an academic, I concede this round to Tom. If I haven't met the guy who wrote that passage, I've sure met his cousins.

I'm a political scientist, and while our discipline doesn't have that disease quite as bad as some others, there's plenty of it around. Fortunately my graduate dept (MIT) was much less post-modern naval-gazing than average.

On the other hand, political science is also filled with the mirror image of what Tom posted -- trivial, scholastic use of complicated economic modelling. I have seen articles where people write (in math symbols), the partial derivative of x with respect to y is greater than zero, not connected to a larger set of equations but simply to say, "when x increases, so does y". More seriously, people do incredibly advanced math to arrive at conclusions like, "the president will launch a diversionary war if his leadership ability is at least pi/2 times his policy competence, and the expected value of the war result is greater than the expected value of economic performance in the next year, adjusted for the president's risk tolerance".

Now, ask said modeller how one would ever quantify "policy competence" or "leadership ability", and they'll just shrug and say it's not their problem, the important thing is having the _theory_. Bah. There certainly are economists and political scientists who care about developing mathematical models that can actually be measured and tested, but the journals are filled with papers full of painful math used to show something that is trivial, picayune, irrelevant, unmeasurable, or all of the above.

I would point out that while most academics are liberal, I know conservatives who are just as guilty of these crimes. Academic obscurantism is a bipartisan disease.

This cult of irrelevance is one reason I'm happy with my job teaching at a military school -- I've got practically minded colleagues and practically minded students.