In a discussion of Paul Krugman and his critics, Noah Smith compares two styles of argumentation:
Way #1 is to put your complete thought process on a page – to lay out both sides of an argument, and explain why you arrived at a conclusion. This is what [Tyler] Cowen calls the “Humean” method, after David Hume. As I [Smith] see it, the Humean method is what you use if you want to get the most out of a discussion with a well-informed but fundamentally disinterested interlocutor. . . .
But not all interlocutors are disinterested. Some have political agendas. Some have strong personal biases. And not all interlocutors are well-informed. . . . In this situation, it may provide the most social benefit to adopt a more Hegelian method of argumentation. . . . two people argue their cases as strongly as possible, and observers can pick and choose the best points of each. This is how our court system works, for example. In the context of econ blogs, using a Hegelian approach means saying “My opponents are going to do everything they can to push their point of view, so I had better do the same in order to balance them out.”
Smith argues that Krugman’s influence on opinion and policy has come about partly because of his [Krugman's] use of the polemical, adversarial style of argument.
I have a couple of thoughts.
First, my impression is that Krugman has always been pretty adversarial in his style. Even before his politicization as a NYT columnist, he was pretty dismissive of those he disagreed with. So, I don’t think it’s a matter of “good Krugman and bad Krugman.” He’s changed his targets more than his style.
Second, the one thing I can’t stand in academic (or other) argument is when somebody makes an attack without identifying the other side. In a blog, it’s appropriate to link to opposing arguments. In print, references will do. In either case, it’s good form to name the people you’re disagreeing with, even if you don’t have the time and space to fully explicate their arguments.
That was one thing that bothered Kaiser and me about Freakonomics: Levitt and his associates have the habit of expressing their view without exploring other possible explanations (most notoriously in the sex-ratio example). It’s not necessary to be completely open-minded but you at least want to give some references to other perspectives.
Krugman and Cowen (the bloggers mentioned by Smith in the linked post above) can both be pretty one-sided at times, but they’re both pretty good at linking to and engaging with the people they disagree with.