After previously snarking at Dan Kahan's snarking, I thought I'd poke around his website's blog a bit more to see what I can learn. Still thinking about it, but there are some useful points.
One that he makes in about every third blog post is that a person's benefit in sticking with the beliefs of one's tribe on communal issues like climate change often exceed the costs to that person of that person being wrong. It's a collective action problem where an individual pointing out that one's tribe is making a huge mistake just causes friction. The individual has a disincentive from even seriously considering whether her tribe is wrong.
The ironic part about this is that I think it makes sense on an intuitive level, but I don't recall Kahan citing evidence that proves it to be true (you can speculate that it explains studies of motivated reasoning, but that doesn't prove the theory). His emphasis on the science of science communication doesn't always follow in practice.
Wonkbook reports a related take on motivated reasoning, which found that partisans severely reduced their motivated reasoning when given a personal incentive to do so. Wonkbook refers to this proving that partisans are just "liars" but I think the psychology might be a little more subtle than that. Kahan's take might be that the cost analysis of thinking for one's self v. believing what the tribe believes is adjusted in the experiment.
Anyway, makes sense to me. I think it might also feed into my pet theory that climate adaptation might be the road to acceptance for climate science, because it's more directly about self-interest.