OK I better write something that’s actually got philosophical content after the last two posts. I was kinda shocked by something that came up when I was preparing my seminar class for tomorrow. We’re working through various papers on explanation in economics, and social science more generally, and for this week we’re looking at various papers by Philip Pettit. (It isn’t the central paper we’re looking at, but if you’ve got JSTOR access you can see one of these papers “here”:http://links.jstor.org/sici?sici=0007-0882%28199606%2947%3A2%3C291%3AFEAVS%3E2.0.CO%3B2-9.)
Now the unsurprising thing is that the papers are very good and the conclusions seem sound. The surprising thing is one of the presuppositions.
The topic is the objection to functionalist explanations that they don’t really _explain_ anything because they don’t provide a mechanism whereby the institution with a useful function comes into existence or persists. Philip thinks this is a _prima facie_ problem, and I agree. But he thinks the same problem hurts rational choice explanations, since it doesn’t provide a _plausible_ mechanism. (Rational-choice explanations are backed up by _a_ mechanism, but it doesn’t seem to resemble the decision-making mechanisms real people actually use.)
Philip goes on to argue, rightly I think, that both functionalist and rational choice explanations can properly explain the _resilience_ of certain institutions and practices, even if they don’t explain the introduction or even the persistence of those phenomena. There’s lots here to talk about, but as I said I want to look back at the presuppositions involved.
Why think there is even a problem with the functionalist explanations to start with? And at this point I should note that thinking there’s a problem here isn’t some idiosyncratic feature of Philip Pettit – the problem is reasonably widely discussed. For one thing, functionalist explanations fit the old DN model of explanations that Hempel and Oppenheim advocated. More interestingly, they seem to fit the unificationist model of explanation that Michael Friedman and Phillip Kitcher advocate as a replacement for the DN model.
Why is this interesting? Well for one thing it looks like a clear case where intuitions of the relevant scientists simply don’t match up with the predictions of the unificationists. And given that what they are trying to explain is a social/cognitive phenomenon, what counts as a good explanation in our scientific culture, these intuitions should be important even to naturalists.
For another thing, this does not look like a merely technical problem, in the way that cases like the flagpole case feel a little artificial. (The flagpole problem is that on the DN theory there’s no way to distinguish the good explanation of a flagpole’s shadow length in terms of its height from the bad explanation of a flagpole’s height in terms of its length.) This is a naturally occuring putative explanation that fits the criteria but of the unificationist but doesn’t strike people as explanatory at all.
But something about this puzzles me. The problem for unificationists is so obvious that someone must have commented on this before me. And really I should do a huge lit search now to find out who. Since the Red Sox are rained out and ’24’ has been bumped for some press conference, maybe I’ll make a start on this, but I’m not sure how far I’ll get.