Tuesday, October 2, 2012

Disciplining your theory

Daniel Klein has three suggestions on what you should ask yourself (HT - Tyler Cowen):

1. Theory of what?
2. Why should we care?
3. What merit in your explanation?

I think it's a good list.

As Klein does in lots of cases, he tests whether other economists conform to what he thinks is good theorizing (like similar tests of whether economists are good classical liberals). And he comes up with a number: out of 66 articles in the Journal of Economic Theory, only 12% pass his theory test.

This sort of exercise, I think, is very problematic (I haven't read the linked paper in detail, but the Journal of Economic Theory analysis is taken from an earlier EJW paper I have read).

Ultimately, what ends up happening is the answers to these questions are somewhat different depending on the audience - particularly the second and third one. These are subjective. I don't care about everything you care about. To a certain extent, even meritorious scholarship is subjective.

Just like Daniel Klein's previous finding that non-libertarians aren't economically enlightened, the finding that 12% of the publications in the Journal of Economic Theory aren't theories should have sent up a bunch of red flags.

So I'm not sure that's the right direction to take this. But I do think that if you can't convince yourself about the answers to these three questions, you're going to have no chance of convincing anyone else.


  1. Daniel Kuehn: Does Klein's point have anything to do with Thomas Kuhn's philosophy of science? Just curious.

    1. What are you thinking of exactly? I'm not seeing the connection, except in some general sense of scientific theories being valuable that I imagine is true of just about everyone's philosophy of science.

      I never like thinking of Kuhn as a philosopher of science. He gets past that. I like to think of him as a sociologist of science. "Historian of science" if you must (that's probably the most common), but he did an awful lot of social theory for a historian.

  2. I especially don't like #2, although it is rampant at GMU. I care because I'm curious about things. If you are not, go read something else.

    1. What's amazing is that a lot of their stuff wouldn't pass #2 for a general audience, so you'd think the fact that making the effort to write the damn article would be revealed preference enough to pass #2 for everyone. But I guess not.

  3. This from the guy whose argument for the central importance of the Invisible Hand in Smith's work was that, while the idea comes up explicitly only twice, once in WN and once in TMS, in each case they appear in the exact center of the text!!!!!! Where was #3 then, DK?


All anonymous comments will be deleted. Consistent pseudonyms are fine.