Thu May 1, 2003

Hats off to D-Squared

Hats off to Daniel Davies for having the energy to do what I do not. He wades into the comments section and says some of what needs to be said about this post by Megan McArdle (aka Jane Galt). She’s writing about the supposedly rock-solid scientific rigor of Economics compared to the humanities and allegedly fake social “sciences” like Political Science and Sociology. As the post and its comments thread show, few topics provide so much opportunity for confidently talking out your ass as the question “Are the social sciences really scientific?”

Informed answers to this question are rare. Instead, you tend to get half-baked ideas about predicability and falsifiability as the criteria for science being put into service shoring up one’s allegiance to a chosen tribe. If predictability looks like a shaky foundation (What? You’re telling me bright young economists don’t get hired on the basis of successful predictions?) the ground can be shifted to the existence of “basic shared premises within the field.” If these shared premises begin to look a bit too metaphysical, then we’ll move to a different criterion. Whatever it takes to preserve the phenomena. It’s easy and fun. All you have to do sacrifice is your consistency. For instance, in comments to her post Megan asserts:

Sociology will assume the science mantle when it stops relying on first person subjective evaluations and surveys

which manages to combine ignorance of sociology, philosophy of science and survey methodology in the space of a few words. She then gets called on a related question a little later, when she asserts engineers tend to be politically conservative and humanists socialist loons. Her response?

It’s not a wild-assed generalization—survey after survey illustrates that in general, engineering schools run much more conservative than humanities departments. Or would you care to offer some disconfirming evidence?

Sooo, maybe surveys aren’t so bad, eh? But perhaps those surveys were conducted by University of Chicago economists or something. Meanwhile, Daniel Davies goes on to note that, empirically,

Interestingly enough, in the Soviet Union, the engineering faculties were more reliably Communist that the arts faculties, who were notorious for incubating capitalist counterrevolutionary dissidents.

Whoops! Looks like we need an auxiliary hypothesis to preserve the scope of Megan’s scientific generalization and make it non-disconfirmed. (Or non-disconfirmable. That’s the great thing about auxiliary hypotheses.)

Henry Farrell has a good post about all this, picking up on some other issues. James Joyner also makes some useful points in defence of Political Science but, alas, cannot resist saying he shares with Megan a “general contempt for sociology as a discipline”. Here James pursues the appeasement strategy, feeding the economism crocodile in the hope that it will eat him and his field last. (Of course, Rat Choice theory ate Political Science first.) A pity.

In the comments to James’s post, Megan pops up again with this amusing anecdote:

I recall an argument I was having with my ex-boyfriend at dinner about some sociology professor he’d had who did a study arguing that crack gangs behave like businesses—well, not so much a study [sic] as following around the head of a gang. I was arguing that while that may be true, it’s not a social science because you don’t really know it—all you know is how that gang leader behaved when the sociology professor was in the room. As it happens, my best friend is an economist … [she] turned around and started to go into ways you could try to control for observer bias.

“It was a single sociologist following around a drug dealer,” I said.

“Well,” she said “that’s because sociology is crap.”

Hahaha. Funny she should pick that example. The sociologist and ethnographer Sudhir Venkatesh does precisely this kind of work. He’s now at Columbia, but was a grad student at Chicago and is probably the sociologist the ex-boyfriend ran into. Venkatesh “follows gang members around” and so on. (Nothing to it, really. Try it yourself sometime.) In addition to his book, American Project, he has a couple of co-authored papers about gangs acting like businesses. One is called “Are We a Family or a Business?” History and Disjuncture in the Urban American Street Gang. Another is “The Financial Activities of an Urban Street Gang.” The former appears in Theory and Society, a respected Sociology journal I don’t expect Megan to have heard of. The latter is in vol 115(3) of the Quarterly Journal of Economics, which she may know about. In both cases, Venkatesh’s co-author is Steven Levitt, the economist at the University of Chicago who was awarded the John Bates Clark medal last week.

Looks like it’s time for another auxiliary hypothesis to preserve those generalizations. Possible candidates include

  • Venkatesh is not really a sociologist.
  • Levitt is not really an economist.
  • The article was published in the special non-scientific section of the QJE.
  • OK I’ll grant you that exception but I’m sure you can’t think of any more.
  • Sentences written by Levitt were scientific; those written by Venkatesh were unscientific.
  • [Fill in as appropriate. Whatever it takes.]

Gaah. Don’t know why I bother, really.