So says Jane Galt, anyway. Here’s what I posted in comments over there in response…

I’m afraid I’ll have to continue to disagree. Political science, like sociology and economics, is a science because it applies the scientific method (a.k.a. empiricism) to the study of politics.

The difference between political science and say physics isn’t so much the method as the measures; physicists have very good measures of the quantities they like to analyze (819,321 leptons here, 1.32121 angstroms there), while political scientists (and sociologists and psychologists and…) have to deal with things that are much harder to quantify. Economics does slightly better because (a) there’s a simplifying assumption that utility is money (which happens to work nicely for most people except monks and tree-huggers) and (b) money is very easy to quantify compared to “affect,” “ideology,” or “partisanship” (things that political scientists measure).

Now, there are people who study politics (and economic behavior and social groups) without the use of the scientific method. They aren’t scientists, and thankfully most of them have gotten it through their skulls that they aren’t and don’t pretend to be. (The question then is what do you do with the people who don’t study a social science using the scientific method, which becomes a thorny political issue.)

More broadly, I think you’re confusing generalizability with empiricism. Yes, sociologist X followed around a single drug dealer; yes Dick Fenno followed around Senator Y. But if enough sociologists follow around enough drug dealers we can test a theory about how drug dealers behave in general.

I’d find the whole discussion laughable if Jane (who usually I find rather thoughtful) weren’t so horribly misguided on what “science” is and isn’t.

Other reaction to Jane’s original post can be found at The Volokh Conspiracy, where Jacob Levy and Dan Drezner have comments. Also, I somehow forgot to link to Kieran Healy, who defends the honor of sociology.