Scott Alexander's correction on ivermectin and its meaning for the rationalists
doyourownresearch.substack.comPursuit of rationality can be paradoxical in a sense, insofar as believing yourself capable attaining rationality is fundamentally irrational. You aren't a Vulcan, you're a Human. To err is human. All humans are irrational some of the time and if you believe yourself capable of being better than that, you're only proving yourself wrong with that very belief.
Rationality done right is the pursuit of an unobtainable goal that yields better-than-average results even as you ultimately fall short of the ideal. So, basically like any other form of self-improvement. When you inevitably hit a setback, reorient and adjust your approach. But don't beat yourself up when you continue to fall short, because you will. We all will.
I believe this is known among some as the concept of striving to be less...wrong...
This is what rationalists like to refer to as a strawman. Show me any example of one claiming to eventually be without err or irrationality. The claims are simply that most people have many implicit biases and illogical means, and that there are some ways of bettering your thinking methods.
I think the idea is that the author of the post we're talking about doesn't consider the idea that they may be internally biased in a way that affects their ability to be rational. From my perspective, it looks like they started from the belief that ivermectin is an generally effective treatment for COVID and that "the data" (that is, a properly weighted and filtered version of the publication record) demonstrates that.
To me that shows a lot of overconfidence. Even as I make statements saying that ivermectin isn't a generally effective treatment for COVID, I fully accept that it's possible my reasoning and interpretation of the evidence could be faulty and that some future study could somehow demonstrate a slam-dunk benefit even for countries with no level of parasitic infections, like the US. If that happened (seems very unlikely, but not impossible) I would change my statement (as would the mainstream medical establishment). The author of the above article doesn't really show that level of epistemic humility.
> Show me any example of one claiming to eventually be without err or irrationality.
Interact with them long enough and you'll come across a whole lot who act like it.
What do you even think I'm strawmanning? That comment is a defense of Scott Alexander and rationalists generally.
I thought you meant to imply that rationalists believe they're perfectly rational. Lots of people do so.
I guess 'rationalism', as a name, calls a certain kind of person out of the ether:
1. They like rationality, and feel it's opposed by 'irrationality'.
2. They want to be part of some 'rational group'.
3. They're ignorant enough not to know the name is already taken.
4. They're ignorant enough not to know that the name means pretty much the opposite of what they believe. (A real rationalist, for instance, probably wouldn't be interested in modern science).
5. They're ignorant enough not to know that 'rational' and 'irrational' are usually demarcate lines of social hierarchy, not lines of theoretical commitment.
This is an attractive pitch, so obviously, loads of people jump on. I think the main thing that's nice about it is that no real work is called for. Every smart white kid from a nice background has been called 'logical' or 'rational' at some point, because (5), so it's a value they identify with. It's a young group, full of energy, because the internet is biased young, and young people go for (2) through (5).
> 3. They're ignorant enough not to know the name is already taken.
Overloads are a thing, you know?
> 4. They're ignorant enough not to know that the name means pretty much the opposite of what they believe.
String of characters doesn't mean anything by itself, it can point to meaning (or several).
> 5. They're ignorant enough not to know that 'rational' and 'irrational' are usually demarcate lines of social hierarchy, not lines of theoretical commitment.
By whom, and why is their use of the word supposed to be the default?
> I think the main thing that's nice about it is that no real work is called for.
Lol no. You might want to look back at your comment. Specifically, "I guess". You've done a lot of judging, without doing a shred of work to verify whether your insults are true.
"They're ignorant enough", repeated several times, despite utter ignorance about people you're talking about.
Anyway. "Rationalist" is aspirational, not a claim of one's own rationality.
Sure, this naming kinda sucks because it's unclear.
Lets say you want to contribute to military theory, you feel like you have some good ideas, but you think tactics means the broad goals you mean to obtain with your 'strategy', which is your low-level techniques you used to achieve these goals. How seriously do you think people in military circles would take you? How seriously do you think they should take you?
That's what the rationalist thing is, more or less. I think you can probably argue that the military guys should take you seriously because you have lots of clever ideas, but you can probably also understand it's kinda ridiculous and bonkers.
Yay gatekeeping!
And look at all these idiots using the word "computer" to mean an electronic computation device! They must be super ignorant to not realize the word was already taken to mean human beings who perform computations as their job!
I would like to be more rational in my thoughts and decisions. I would like it if others were as well. I find that reading articles by so-called "rationalists", and reading debate around them, seems to help me do that. Not surprising, since that is their exact stated goal as well.
As far as what "rational" means:
https://en.wiktionary.org/wiki/rational
1. Capable of reasoning.
2. Logically sound; not contradictory or otherwise absurd.
3. Healthy or balanced intellectually; exhibiting reasonableness.
4-7. [Math, chemistry, and physics stuff]
Hmm, nothing about social hierarchy or about how you're not allowed to use that word if you're not a 17th century French philosopher. What the hell are you on about, and why are you so concerned about gatekeeping this word?
Rationalism != Rational. Putting an ism on the end of the word makes it a new word. Also, computation is still used in the sense you described.
PS: I'm not sure if it's gatekeeping to expect people who are interested in something to know a word you would learn in the first hour(s?) of learning about it. But if so, I am fully behind it.
Polish language and Polish culture, heavily influenced by French Enlightenment, also accepts use of word rationalism from empirical perspective.
Rationalism ( https://sjp.pwn.pl/sjp/racjonalizm;2573228.html ):
3. "a position requiring the observance of restrictive standards of the scientificity of knowledge"
I personally prefer that meaning of the word.
I think the use of "rationalism" follows modern philosophy trends and isn't a reinvention. They are "actually" empiricists as opposed to rationalists, yes, but most of those rationalist ideas[0] moved to continental philosophy and are now part of the modern rationalists' enemies, humanities people who read French theory.
"Rationalism" descends from analytic philosophy/logical positivism/scientific method types, so they're into experiments now. This is good, because their main program of logical positivism doesn't actually work, so it's kind of a problem that they're still trying to do it.
[0] the world doesn't exist, there's only sense data causing you to imagine a shared world, thinking about things logically is better than trying them out
> Rationalism descends from analytic philosophy/logical positivism/scientific method types, so they're into experiments now.
This is false. Rationalism predates analytic philosophy by nearly 300 years. Maybe you mean "rationalism" does.
That is what I meant, yes. I don't believe obviously false things like that, no.
Wait, I'm getting confused between the two senses of 'rationalism' here. My guess is that contemporary 'rationalism' doesn't actually descend from any explicit philosophical tradition, just because if you read an introductory philosophy book, you'd not join a group with that name combined with their ethos.
I'm not that well versed in analytic philosophy - I think it's come a long way since the early days, so now straddles both sides of the rationalist/empiricist divide. At this point, I think it's basically a writing style. You get a lot of analytic philosophers with some pretty wild assertions about reality.
On the other side, somebody like Deleuze is basically a hardcore empiricist, but he's as 'continental' as they come.
> My guess is that contemporary 'rationalism' doesn't actually descend from any explicit philosophical tradition, just because if you read an introductory philosophy book, you'd not join a group with that name combined with their ethos.
It definitely doesn't. The community started in almost complete ignorance of philosophy in general and that philosophy had already covered everything they were trying to.
> I'm not that well versed in analytic philosophy - I think it's come a long way since the early days, so now straddles both sides of the rationalist/empiricist divide.
I'm pretty sure they meant "rationalism" comes from analytic philosophy, which I don't think it does although now at least they've realized it exists. Actual rationalism predates both by approximately three centuries.
> 3. They're ignorant enough not to know the name is already taken.
Imagine the hubris to claim ignorance of another when oneself is ignorant that rationalists already know this and it was a mistaken momentum thing.
Words and meanings change over time and depending on the context. Strangely you claim it's ignorance but then give supporting reasons why someone might label this "rationalism".
Rationalism is an important term in exactly the field they are supposed to be interested in. And they're using it in more or less the opposite of its conventional sense.
I guess it could be just really perverse usage, but I figure it's more like 'objectivism', or 'scientology', words that sound big and impressive to people who are either dumb, or just not very well informed.
You can't really change a term like rationalism, since it's really one of the biggest terms in philosophy, and philosophers are the kind of people who get upset when people use stuff like 'begging the question' wrong, which everybody does. So even if the 'rationalists' ended up having loads of great insights despite their inauspicious start, it would almost certainly never change the meaning of the word.
The meaning of the word can change and it will change sooner or later, in Anglo-sphere or somewhere else.
Historically, the meaning of "rationalism" changed both trough time and cultures.
The way "rationalism" is being used by the lesswrong crowd is currently accepted for example by Polish dictionaries:
Rationalism ( https://sjp.pwn.pl/sjp/racjonalizm;2573228.html ):
3. "a position requiring the observance of restrictive standards of the scientificity of knowledge"
Also from: Williams B., Rationalism, [w:] D.M. Borchert (red.), Encyclopedia of Philosophy, t. VIII, Thomson Gale, 2006, s. 239-247, ISBN 0-02-866072-2.
Rationalism in the enlightenment
The term rationalism is often loosely used to describe an outlook allegedly characteristic of some eighteenth- century thinkers of the Enlightenment, particularly in France, who held an optimistic view of the power of sci- entific inquiry and of education to increase the happiness of humankind and to provide the foundations of a free but harmonious social order. In this connection “ratio- nalistic” is often used as a term of criticism, to suggest a naive or superficial view of human nature that overesti- mates the influence of benevolence and of utilitarian cal- culation and underestimates both the force of destructive impulses in motivation and the importance of such non- rational factors as tradition and faith in the human econ- omy. Jean d’Alembert, Voltaire, and the Marquis de Condorcet, among others, are often cited in this connec- tion. Although there is some truth in these criticisms, the naïveté of these and other Enlightenment writers has often been grossly exaggerated. Also, insofar as “reason” is contrasted with “feeling” or “sentiment,” it is somewhat misleading to describe the Enlightenment writers as rationalistic, for many of them (Denis Diderot, for exam- ple) characteristically emphasized the role of sentiment. Reason was praised in contrast with faith, traditional authority, fanaticism, and superstition. It chiefly repre- sented, therefore, an opposition to traditional Christian- ity.
Here there are two contrasts with the seventeenth- century rationalism of Descartes and others. First, this rationalism is not characteristically antireligious or non- religious; on the contrary, God in some sense, often in a traditional sense, plays a large role in rationalist systems (although Spinoza’s notion of God was extremely unorthodox, and it is notable that the opposition of rea- son and faith is important in his Tractatus Theologico- Politicus). Second, the view of science held by such Enlightenment thinkers as Voltaire was different from that of rationalism, being much more empiricist. The central contrast embodied in the term rationalism as applied to the earlier systems is that of reason versus experience, a contrast that is certainly not present in the Enlightenment praise of the “rational.”
I guess contemporary history of philosophy is kind of a modern product, so it makes sense that it might have been used in a different way back in the day. A lot of historical rationalists considered themselves empiricists, and vice versa. It's also not a term contemporary to like, Descartes (iirc).
I think the current meaning of rationalism (as opposed to empiricism) is vastly more useful than it as yet another 'good-ism', but you're probably right, words can change. It's still stupid that they didn't bother googling the term before calling themselves it.
From the original article they're responding to:
> This is still just a possibility. Maybe I’m over-focusing too hard on a couple positive results and this will all turn out to be nothing. Or who knows, maybe ivermectin does work against COVID a little - although it would have to be very little, fading to not at all in temperate worm-free countries. But this theory feels right to me.
> It feels right to me because it’s the most troll-ish possible solution. Everybody was wrong!
Actually, if it does pan out, score 1 for the conspiracy theorists. If you actually asked them by what mechanism they thought Ivermectin worked, they really had only two answers, and "alleviating strain on your immune system by killing parasites you didn't know you had" was one of them, from very early on.
This ivermectin thing is the premier example of motivated reasoning and wishful thinking in the last few years, something which seems lost on the author.
This article was a bit long-winded, but it makes a very fair point and I don't understand why it was flagged.
it's not complicated: people don't like heresy. notice how there are zero comments on this page that deal with the substance of the article in question, instead choosing to find various surface-level reasons unrelated to the findings in the article as cause to internally dismiss it.
if the content of the article was not heresy and instead came from some trusted scientific authority, this would not be the case.
It doesn't even rise to the level of being wrong. Nobody is even remotely interested in engaging what you call the "substance" of the article, because https://en.wikipedia.org/wiki/Brandolini%27s_law
The people who are working on COVID are too fucking busy to waste their time arguing with kooks who claim they are rational.
Yes. I would have written exactly what you did here, but then I’d probably have to fend off the comments you are referring to.
So I worded it more diplomatically, or ambiguously.
Shame; it’s interesting to get challenged. At least I got to read it before it got flushed.
Tip: if you go to https://news.ycombinator.com/active instead of the front page you can see stories that were flagged. This is how I read HN because the number of interesting, relevant and useful stories that get flagged for ideological reasons is just far too high. This one being a good case in point. It's a pattern: there's a group of people who will reliably flag any article demonstrating that official COVID science is wrong. Doesn't matter what it's about - if it contradicts official narratives then it's probably going to disappear within an hour or two.
One might think a site dedicated to intellectual curiousity would be interested in stories like that, or that at least flagging such stories would be penalized, but no.
Thanks. That's a good idea. I already have showdead activated, so I can at least track them down when they disappear.
Wonder if this phenomenon is discussed with dang et al. I am very motivated to find the truth whereever the search takes me, so encountering this kind of collectivist censorship is very frustrating.
Rationality is only a pathway to reliable information when it makes clear its assumptions (i.e. axioms or postulates) and sticks to logical proof.
Empiricism is only useful when it adequately adopts Occam's Razor to stay focused on the facts and has a clear understanding of the difference between correlation and causality.
Want to disprove a mainstream position? Find a commonly held false assumption or some fact that cannot be explained by the current approaches. Keep it simple and be open to having this point explained with standard methods.
Read T.S. Kuhn if necessary to understand what it typically takes to change a scientific paradigm.
Everything else is just noise with the purpose of making a given community feel enlightened or smarter than the other "fools".
The "Do Your Own Research" title is about where he lost me. Without reading, I'm pretty sure the author didn't perform a double blind study himself, nor is he suggesting that we all perform one.
The rest reads like gibberish.
Researchers pretty regularly do meta-studies and call it research so that seems a poor basis on which to dismiss the article.
what value does this comment have? you dislike the title the author has chosen for his blog, and you are physically unable to read the text below it?
The "rationalist community" is morbidly fascinating in their tendency to be so self-important while also having a deficit of self-awareness. The author of this piece can't understand why an internet-famous blogger/Substack writer that he follows doesn't have time to drop everything and debate the minutia of a blog post he wrong last year on a topic that has long since been settled. Note that Scott did listen to him, work through his reasoning, and update his blog with a note about it, but that didn't satisfy the author:
> Step 6: Semi-Permeable Membranes
> One thing that shocked me was how hard it was to discuss even a simple thing with Scott, even when he knew I could have made a big deal about this without giving him an opportunity to make whatever correction he thought appropriate. It felt like communicating through a straw. I get the sense that Scott is busy. Busy and/or surrounded by people who think the world of him; a community of readers that compliment his writing early and often.
This piece also shares several other characteristics of "rationalist" writings: Unnecessarily long and rambling prose, flowery language and dramatic subsection titles when basic text would suffice, hedging in the middle of the article in case the author turns out to be incorrect, and a relentless insistence that the conversation revolve around their experience and some perceived sleights instead of letting the argument stand alone.
Regardless, this seems like a silly diatribe after the medical community has already investigated the Ivermectin idea to great lengths and at massive scale and concluded that any effects it might have are too small to be worth pursuing. It's weird to see someone writing volumes about re-litigating last years' amateur scientist social media battles.
This author is either obsessed with Ivermectin and the TOGETHER trial or playing a game to pander his Substack to a certain audience who loves this content. His first post was only a month ago but he's already written 11 articles suggesting errors and alluding to conspiracy theories.
> Regardless, this seems like a silly diatribe after the medical community has already investigated the Ivermectin idea to great lengths and at massive scale and concluded that any effects it might have are too small to be worth pursuing. It's weird to see someone writing volumes about re-litigating last years' amateur scientist social media battles.
Many rationalists are driven by contrarianism and are the living embodiments of second-option bias[1] and confirmation bias, so I don't find it weird anymore that they're still relitigating such issues, I just expect things like this from that crowd, now.
I used to think of myself as a rationalist, but after being wrong a few times in some passionate internet arguments I began to think about what it would mean to truly be a rationalist, and given my understanding of the fallibility of the human brain (evidenced by reproduced/reliable psychological studies), the best way to be rational would be to assume that I myself am not 100% rational (in the same sense that nobody can prove themselves to be objective, or truly disprove Descartes' Great Deceiver), and am prone to confirmation bias (and others).
once i accepted that to be more rational I had to accept that I was not completely rational, I was able to reason more probabilistically. This helps, because there is actually no physical system that is truly logical (except for a bit, and making a logical bit is nontrivial), but rather, the only way to understand physical system is to apply probablistic (not logical) reasoning; think of the difference btween a perfect step function/dirac delta function versus a sigmoid.
Thanks for the pointer to second option bias. I can already see I'm going to waste part of the rest of my day finding the part of rationalwiki where it turns out my beliefs were anticipated by Wittgenstein.
I think I went down a somewhat similar route. It's easy to find oneself trying to embody some archetypal champion of "logical" and "rational" thinking when you're a highschool nerd.
In the end I think my issue that led me off it was that it just ended up being a) judgmental and b) like you, finding that I was just not right enough to justify it!
When I ran into more of those types (our types?) Who are a dime a dozen in a college engineering department my goto quip was:
The scientific method is the admission that rationality and logic are limited; they're just the first step — forming a hypothesis. The rest of it is about how your rationality and logic failed you.
I think this is partially why you find many rationalist and skeptic types not actually doing any science. They haven't had the humility beat into them by science yet, haha.
Another thing which helps is to not just automatically have an opinion on everything, even if somewhat informed.
It's ok to say "I don't know, this is very complex, even if I spent many hours reading about this matter". Too many rationalists just go with the option which seems a bit more plausible and then behave as if they have near certainty.
So much this! But there is way too much emphasis on opinions, when frankly, unless someone is deeply invested in the given topic, it is likely worthless. And unfortunately, the less knowledgeable someone is, the more likely he/she has a very strong and vocal opinion on the topic.
I was once attracted to it, and to EA, because of the surface-level values. It seemed like a community that wanted to acknowledge their own biases and work past them, yet in practice, it is a community that uses intellectualism as an aesthetic to confirm their preconceived biases. In its malignant form, you have LW and adjacent communities engaging in scientific racism revival. In the less malignant form, you have people working backwards and pretending that the conclusions they came to were "objective" because of the flowery Bayesian language they dressed their thought experiments in, as if they're constantly doing complex Bayesian inference in their heads. In the end, what was striking, to me, was the lack of humility you'd expect from those who agreed with the LW premise.
Similarly with EA, I liked the idea of optimizing charity for the most good, but in practice, the community seems to have no problem dedicating a ton of money, time and effort to MIRI and adjacent groups and people, because they've managed to use their intellectual aesthetics to spook themselves into believing that science fiction is reality. As a result, very real problems people experience today are discounted in favor of whatever scary future AI meme is spooking the community this month.
It's really kind of funny when you think about it, it's just a shame that they suck up so much oxygen in the room.
> Governments have been caught committing conspiracies in the past, so this means every single conspiracy theory today is true...even the ones that contradict each other.
Well, this is defensible, since if you read this with s/is true/gets an increased probability of truth/ it's just Bayesianism. Sadly, sometimes people are waiting till later to find out which contradictory thing is true, but sometimes they're just bullshitting.
There was a lesswrong thread warning people about Leverage (a 100% literal Bay Area Scientology-like therapy cult, as opposed to the general LW community which is just a self-organized kind-of-a-sex-cult). I thought it was notable because it's written as if actually having an emotion or trying to look like it believed anything would be a deadly sin.
https://www.lesswrong.com/posts/Kz9zMgWB5C27Pmdkh/common-kno...
> I thought it was notable because it's written as if actually having an emotion or trying to look like it believed anything would be a deadly sin.
We’re all sinners.
>debate the minutia of a blog post he wrong last year on a topic that has long since been settled.
Part of the point of the article you're commenting on is that the topic is not, in fact, settled.
> The "rationalist community" is morbidly fascinating in their tendency to be so self-important while also having a deficit of self-awareness
I’m about to start calling them what they are: Sophists.
You know those jokes about "how do you know someone is a vegan/PhD"?
Well, how do you know someone is a rationalist? Don't worry, he will tell you soon that he's "updating".
It's a fascinating community, in some regards. Scott is among the few people where I consciously notice that they are highly intelligent. He's also a decent writer and, despite my lingering fear, has not gone down the Glenn-Greenwald-pipeline into obnoxious contrarianism.
...and then there is the rest of the community. They cargo-cult all the phrases and hobbies and opinions, but it's like when you wore that first suit: it wasn't yours, you didn't have the shoulders, and everyone but you knew it looked ridiculous. Seriously: even the other well-known people from that community (whose names I've since forgotten) are so obviously cosplaying the intellectualism. Even in the natural science I studied, I would not be able to identify the smart and the not-so-smart papers or professors with such ease.