Arnold Kling writes An Open Letter to Paul Krugman. Definitely worth reading. He points out that there are two types of argumentation: against the argument, and against the man. He calls them type C (for consequences) and type M (for motivations). He takes Paul Krugman to task for making too many type M arguments. This isn't a new observation. The Latin term "ad-hominem" means: against the man. An honest argumentor (ispell and google insist that it's not a word; if it wasn't before, it is now) will avoid ad-hominem argumentation. Who says something, or why they say it, has nothing to do with the correctness of that said.
His last two paragraphs are especially near and dear to my heart. 50% of what is wrong with economics is not the results of economics, but instead how we *present* the results of economics. When we use type M arguments we give people the idea that economics is not a science; it's just the opinions of people who call themselves economists; that being an economist is just a matter of having studied all the previous opinions; and that therefore nobody's opinions about economics can be wrong (not "are wrong" but "can be wrong") because they're all just opinions.
I knew I didn't like Paul Krugman. Now I know why.
Addendum: I heard Paul being interviewed on by Scott Simon. He used mostly type C arguments until the end where he used a type M argument ("crony capitalism").