Scott is very intentional about not getting into the landmine issues. IMO part of his service to the community is helping define those boundaries for the less... socially aware members of the community (ie, the things you Do Not Discuss).
Link: https://slatestarcodex.com/2017/05/26/the-atomic-bomb-consid...
With the caveat that I'm a Chomskyan free-speech absolutist, and I generally agree with you: it's not hard to apply the same Halting Problem concept to free expression. Such an exploit can manifest many ways: QAnon, Red Guards, literal Nazis, Woke "Neo-Marxists", etc; but regardless, scale the intersection of extremist ideology and human social behavior until they approach infinity, and it's easy to see how they can (and historically have) resulted in a systemic collapse of free expression and free thought. (This was what Popper was trying to capture in his now-oft-quoted Paradox of Tolerance, which is now ironically over-applied as a lever of pre-emptive intolerance of challenges to orthodoxy!)
I do agree that it's both a categorical and strategic mistake to succumb to an epidemiological model of memetic extremism. Even to the extent that model applies, extremist ideas only spread under social preconditions of susceptibility (as described by Hoffer [0]), and I don't think pre-emptive idea suppression is either right, or wise, or helpful. Extremist ideological infection is more a symptom than a root cause.
And yet: we can still recognize that certain taboos might exist for a reason, such as the one against openly voicing "maybe we should just kill the people who disagree with us". James Lindsay (perhaps the second-most infamous opponent of postmodernist thought) has described postmodernism as a "universal solvent", capable of taking apart any idea. It's not that you never use such a cognitive tool; rather, one uses it cautiously and judiciously, when one has the wisdom to wield it properly. Similarly, we need safe spaces for dangerous thoughts, even of the Popperian or Halting Problem variety; yet it may be appropriate to hold social taboos against those ideas being used casually in polite society and the public discourse, lest they dissolve polite society and public discourse themselves.
[0] https://samzdat.com/2017/06/28/without-belief-in-a-god-but-n...
Are you talking about stuff like this? https://slatestarcodex.com/2013/10/20/the-anti-reactionary-f...
I'm no historian, but the stuff he wrote about Catholicism doesn't actually seem right. Argued better than I can in [0].
And some of the comments are not just atrocious, but are not argued against[1]:
> I think there are commonly-known models in all four quadrants. For example:
> a. Widely accepted and good fit for reality: (Law of supply and demand)
• “Supply and demand” is a good first-order model for certain market dynamics, but it only explains… at a guess, ⅓ of the economics I personally interact with.
> b. Socially unacceptable and good fit for reality: (IQ tests as a good proxy for mental ability)
• IQ tests are a reasonable proxy for certain, specific axes of mental ability within a subpopulation; the general “IQ tests are a good proxy for mental ability” claim is blatantly absurd.
> c. Widely accepted and bad fit for reality: (Sexism as main cause of gender wage gap)
• “Sexism is not the main cause of the gender wage gap”… I'm less certain that this is wrong, but I have actually done quite a bit of research on it (including reading what Scott Alexander wrote on the topic!) and this is splitting hairs, to be charitable; I think labelling it “rhetoric” is more accurate. No, several of the systemic injustices aren't due to individual people going “aha! I know what I'm going to do today: not pay my female underlings!”, you're right! But everybody already knows this,[2] and that's not what they mean when they say it's due to sexism. Things like a culture of “you only get a raise if you push for it” (not sure how widespread this is) can contribute to this, and that is, when considered in combination with the rest of everything (e.g. men may be “forthright” and “assertive”, but women are “bossy”), a sexist aspect of the culture.
> d. Socially unacceptable and bad fit for reality (Vaccines cause autism)
(The rest of the comment is mostly okay, apart from those three examples above; I'm cherry-picking to make a point, but I think the point's valid.)
These are just trotted out as “obvious if you're one of us”, when they're probably not even correct. You don't go into a room where people say sensible things and think that your association with those people somehow makes what you say sensible, and you certainly don't sit in the audience of an entry-level lecture and assume that your fellow audience-members are all experts in the field, so why assume that you should take as blind truth things you read in the comments of a blog‽
Also, I'm not certain I agree with the message Scott's trying to convey. This uncertainty correlates with my uncertainty about the historical accuracy of his examples: when he has a point he wants to make and finds some evidence after the fact, it's generally obvious that he's doing so (probably because that skill doesn't get much use).
[0]:https://slatestarcodex.com/2017/10/23/kolmogorov-complicity-...
[1]: https://slatestarcodex.com/2017/10/23/kolmogorov-complicity-...