Wiki Contributions



(Will you all get this comment as an email?) Looking forward to meeting! I'll bring nametags and a sign up sheet and some extemporaneously-chosen food.


Broken link:

Expected behavior: You can see the comment, a la

(do make sure you hit the '+')

Actual behavior: You can't see the comment on the page unless you click "show more" comments. Click "show more," and the page reloads, and it zooms down to see his comment. Given that{...}/2f14 is a direct link to that comment, it should show that comment.


Some time ago I stopped telling people I'd be somewhere at ish-o'clock. 4PM-ish for example. I really appreciate when people tell me they'll be somewhere at an exact time, and they're there.

I've heard that people are more on-time for a meeting that starts at 4:05 than one at 4:00, and I've used that tactic (though I'd pick the less-obviously-sneaky 4:15).


Yeah--when the person asking the question said, "90 years," and the Turing award winners raised some hands, couldn't they be interpreted to be specifying a wide confidence interval, which is what you should do when you know you don't have domain expertise with which to predict the future?

This intuitively feels epistemologically arrogant, but it succeeds in solving the probability language discrepancy.

In general I support the thought that you avoid a lot of pitfalls if you're really precise and really upfront about what kinds of evidence you'll accept and not. I suspect that that kind of planning is not discussed enough in rationalist-circles, so I appreciate this post! You're upfront about the fact that you'll accept a non-explicit signal. I see nothing wrong with that, given that you're many inferential steps from a shared understanding of probability.


First: Yes I agree that my thing is a different thing, different enough to warrant a new name. And I am sneaking in negative affect.

Yeah, no kidding it’s easier to catch people doing it—because it’s a completely different thing!

Indeed, I am implicitly arguing that we should be focused on faults-we-actually-have[0], not faults-it's-easy-to-see-we-don't. My example of this is the above-linked podcast, where the hosts hem and haw and, after thinking about it, decide they have no sacred cows, and declare that Good (full disclosure: I like the podcast).

"Sacred-cow" as "well-formed proposition about the world you'd choose to be ignorant of" is clearly bad to LWers, so much so that it's non-tribal.

[0] And especially, faults-we-have-in-common-with-non-rationalists! I said, "The advantage of this definition is that it’s easier to catch rationalists and non-rationalists doing it." Said Achmiz gave examples using the word "people," but I intended to group rationalists with non-rationalists.


I sometimes hear rationalist-or-adjacent people say, "I don't have any sacred-cow-type beliefs." This is the perspective of this commenter who says, "lesswrong doesn't scandalize easily." Agreed: rationalists-and-adjacents entertain a wide variety of propositions.

The conventional definition of sacred-cow-belief is: A falsifiable belief about the world you wouldn't want falsified, given the chance. For example: If a theist had the opportunity to open a box to see whether God existed, and refused, and wouldn't let anyone else open the box, that belief is a sacred cow.

A more interesting (to me) definition of sacred cow is: a belief that causes you to not notice mistakes you make. The advantage of this definition is that it's easier to catch rationalists and non-rationalists doing it. Rationalists are much better than average at evaluating well-formed propositions, so they won't be baited by a passionately ignorant person talking about homeopathy.

But there are pre-propositional beliefs[1]. Here are some examples:

Talking about it makes it better.
I'm the man of the house, and the man of the house should be strong.
I'm unworthy.
I can tolerate anything but the outgroup.
Worrying solves nothing.
People enjoy my presence.
People need structure.

You walk into a conversation and start talking and miss what your interlocutor was talking about because you believe things like these.

An example rant: I am damn-near unwilling to falsify my belief that "people need structure." When they ask me for help what am I supposed to say? Anything I suggest is additional structure. Were I to suggest taking a break, that's still a change, and change is structure. Maybe change itself is a problem, but that's beside the point; my interlocutor is asking for advice. That means change, and change entails a change in structure. You might as well make it good structure.

What I just said is intended to show that sacred cows are terse beliefs that constrain decisions in social interaction, and offer easy post hoc rationalization. Say I find out my belief is false. Now, I've been making mistakes all my life. I blundered into conversations adding structure. Without that sacred cow, I don't know what to say.

[1] It's probably a spectrum. All those examples can be made falsifiable. They all have in common vagueness, which begets unfalsifiability.

Note: This is a shorter version of a comment I wrote here, written as a comment on an episode of The Bayesian Conspiracy, a podcast I like.

(as Elizier says, it is dangerous to be half a rationalist, link, there's a better link somewhere, but I can't find it)

This might be it:


And you do not warn them to scrutinize arguments they agree with just as hard as they scrutinize incongruent arguments for flaws.  So they have acquired a great repertoire of flaws of which to accuse only arguments and arguers who they don't like.  This, I suspect, is one of the primary ways that smart people end up stupid. 

(it also mentions that it's dangerous to be half a rationalist.


I'm going to write soon about how I don't care about existential risk, and how I can't figure out why. Am I not a good rationalist? Why can't I seem to care?

In one compound sentence: Personal demons made me a rationalist; personal demons decide what I think/feel is important.

I'm still angsty!

Load More