I noticed this too, and for a long time it bothered me. But I recently read Jonathan Haidt's book The Righteous Mind, and in it he points out that a lot of people appear to hold believes about a wide array of issues (politics, religion, consumer products, and so forth) that they don't really hold based on logic and evidence, but to signal group identification and affiliation.

In addition, he points out that, on a wide array of issues, people tend to have gut, intuition-based reactions first, then look for evidence to support their intuition, while a lot of us assume or want to assume that it works the other way around.

I probably learned something from The Righteous Mind on every page, and I say this about very few books; I also wrote at more length about it here: http://jseliger.com/2012/03/25/jonathan-haidts-the-righteous... .

(BTW, I agree with your basic point and think it's well put.)