Political Labels, Political Identity, and Bias

Bryan Caplan writes:

I'm a libertarian, a natalist, an atheist, a credentialist, an economist, an optimist, a behavioral economist, an elitist, a public choicer, a dualist, a Szaszian, a moral realist, an anti-communist, a pacifist, a hereditarian, a Masonomist, a moral intuitionist, a free-market Keynesian, a deontologist, a modal realist, a Huemerian, a Darwinian, the other kind of libertarian (=a believer in free will), and much more. I could spend hours adding additional labels to the list. So it naturally caught my attention when Will Wilkinson remarked:

People call me libertarian but I don't in part because I'm not one, but mostly because I suspect that accepting any such label dings my IQ about 15 points.

If the IQ ding is additive, my many labels have long since reduced me to the intelligence of a cranberry. And even if the ding isn't additive, I don't have 15 IQ points to spare.

Bryan is too modest. He has at least 17 points to spare.

Seriously, Bryan has sort of wrongly inferred that my aversion to specifically political labels flows from a much more general aversion to naming one's convictions. At the limit, Bryan makes it sound as though I have a beef with the whole idea of self-predication. I don't. I am an Earthling, a chordate, an Iowan, a compatibilist, and I'm not afraid to say so!

I say "sort of wrongly" because I do have a bit of a general aversion to naming one's convictions, which I'll discuss in another post. But I think there's definitely something special about political ideology which tends to make it rather more central to our self-conception than our positions on obscure philosophical questions.

Politics just is coalitional conflict. A political label puts you, like it or not, on a team in a number of disputes in which there are significant real-world stakes. People therefore tend to see their ideological affiliation as constitutive of their identity in a way their opinion about the ontology of mental illness (to use one of Bryan's examples) isn't. People advertise their politics by putting Che Guevara and Murray Rothbard on t-shirts, but they don't much advertise their metaethics with Kant gear. I've never seen a modal realism button. (If there were a combinatorial fictionalism button, I'd wear it!) Other people are thus likely to see our politics as central to our identity, and to see our attributed identity through the prism of their politics. Self-labeling gives others permission to apply to us the label we apply to ourselves, and (here is something I believe!) who we are is to a large extent a complicated product of our reactions to social expectations.

Let me tell a little story. Up until the weeks before I parted ways with Cato, I never felt any overt pressure to toe any sort of party line. But almost as soon as I left, I found that I was noticeably less reflexively defensive about anti-libertarian arguments. I found it easier to the see merit it in them! I feel sure that much of this has to do with the fact that at some level I had recognized that my livelihood depended on staying within the broad bounds of the libertarian reservation, and that this recognition had been exerting a subtle unconscious pressure on my thought. Once I became an independent operator, much of that pressure lifted. And as soon as that pressure lifted, I began to feel much less attached to the libertarian label. And as that sense of attachment waned, I became even less reflexively defensive about anti-libertarian arguments. It became hard for me to avoid the conclusion that my political self-conception had been interfering with my ability to evaluate arguments objectively. I had been letting people on my team get away with bad arguments, and I had been failing to acknowledge the force of arguments against my team's tenets. The fact that everybody else does this, too, doesn't make me feel any better about my own sins against Truth.

So I am resisting a strong sense of ideological identity. If pressed, I'll say I'm an inscrutably idiosyncratic liberal. This has been paying dividends! How do I know? Because I feel very confused. That is, to put it in Mason-speak, the probabilities of truth I assign many of my politically relevant beliefs seem to be drifting downward into the neighborhood just north of .5, which is about where they ought to be.