3 ms·

I think your characterization is very unfair to the article. Right after what you quoted, the following is stated explicitly: "Of course you may not trust me

I think your characterization is very unfair to the article. Right after what you quoted, the following is stated explicitly: "Of course you may not trust me about any of [the examples you wrote about]. And so my purpose today is not to propose a new litmus test to replace atheism. But I do propose that before you give anyone credit for being a smart, rational skeptic, that you ask them to defend some non-mainstream belief...you ought to have to stick your neck out and say something a little less usual - say where you are not skeptical (and most of your tribemates are) or where you are skeptical (and most of the people in your tribe are not)." I don't see how it can be written any clearer than that. The point of picking those examples is not that they are the perfect tests of skepticism; it's that most of the Less Wrong readership has thought hard about cryonics, many-worlds, nanotech, and AI, and so they are some reasonable examples.

macrael 15y ago
I stand by what I wrote. My main issue is with the Shermer, Penn, and Teller examples. The author uses those to attempt to prove a different point than the rest of the article discusses, and it is a bad point. There's a difference between saying "Someone is probably a 'good' skeptic if they have said something that the mainstream disagrees with" and saying "Someone is probably a 'bad' skeptic (charlatan?) if they have said something the mainstream agrees with that I disagree with." Even if that label is only personal (as you quote, he's not trying to make everyone apply the same labels to those people) it is a bad way to apply it. However, I also find the whole idea of a "skeptic test" a bit disquieting. Paraphrasing what you quoted above: "Don't give someone credit for being a smart, rational skeptic until they have defended some non-mainstream belief, until they have stuck their neck out to say something a little less usual." That's a bit aggressive for my taste.
- gwern 15y ago
  No, the Shermer example is right on. Shermer knows he dishonestly argued against cryonics; he knows that his little frozen-strawberry example is factually incorrect (see his admission in http://web.archive.org/web/20100306205336/http://leitl.org/docs/public_html/tt/msg11904.html http://web.archive.org/web/20100306205336/http://leitl.org/d...) but still published it.

macrael 15y ago

I stand by what I wrote. My main issue is with the Shermer, Penn, and Teller examples. The author uses those to attempt to prove a different point than the rest of the article discusses, and it is a bad point. There's a difference between saying "Someone is probably a 'good' skeptic if they have said something that the mainstream disagrees with" and saying "Someone is probably a 'bad' skeptic (charlatan?) if they have said something the mainstream agrees with that I disagree with." Even if that label is only personal (as you quote, he's not trying to make everyone apply the same labels to those people) it is a bad way to apply it. However, I also find the whole idea of a "skeptic test" a bit disquieting. Paraphrasing what you quoted above: "Don't give someone credit for being a smart, rational skeptic until they have defended some non-mainstream belief, until they have stuck their neck out to say something a little less usual." That's a bit aggressive for my taste.

gwern 15y ago
No, the Shermer example is right on. Shermer knows he dishonestly argued against cryonics; he knows that his little frozen-strawberry example is factually incorrect (see his admission in http://web.archive.org/web/20100306205336/http://leitl.org/docs/public_html/tt/msg11904.html http://web.archive.org/web/20100306205336/http://leitl.org/d...) but still published it.