Those are good examples, BUT, I wasn't saying that educated westerners don't accept all forms of weirdness, but rather that weirdness alone isn't damning, so dislike for some of your examples could be attributed to something else than conformism (mentally ill people are more likely to be stupid or dangerous, poor speakers are worse at communication, voluntarily unemployed people are more likely to be lazy, promiscuous people may be less likely to be faithful spouses).
That being said, you do make a good argument that weirdness per se will probably get you dark looks, and yeah, parenting and sex are probably areas where weirdness is judged particularly negatively.
Note that Caplan's argument was that intolerance for weirdness reduced innovation, and I'm not sure that our intolerance of weirdness is strong and broad enough to justify that conclusion.
I'm not really convinced either way; maybe a clearer example of beneficial-but-weird would be eating insects.
It's also possible that I'm in an unusually-tolerant-of-weirness environment; I work in the game industry (with a very loose dress code, and people with pretty diverse backgrounds and attitudes, and nobody minds if we play board games at lunch break), a lot of my uncles and aunts and cousins went off to marry (or live with) foreigners, and I don't get any comments when I take my son to the park without my wife. Someone whose family is mostly local and religious, and who works in a bank may get a different impression of how tolerant society is of difference.
To follow up handoflixue's points:
(1) many of the correlations you noted are very weak, relatively speaking (particularly dangerousness and mental illness) some of these labels are social coding, not bare fact (some people who are "voluntarily unemployed" are called housewives. are they lazy?)
(2) why do you care? even if voluntarily unemployed = lazy, what difference does that make in your life? That's the central problem of intolerance - it's mostly concealed, unstated judgment about how society should be.
...It's also possible that I'm in an u
Here is a new post at EconLog in which Bryan Caplan discusses how signalling contributes to the status quo bias.