Chapman isn’t really part of the rationalist community, and was working in parallel on something different that has now intersected because both Chapman and Eliezer are trying to reach similar audiences.
Probably the strangest thing about Chapman’s writing has always been the way that he would rail against “rationalists” and “rationality”, and then, when it was pointed out that his characterization doesn’t match the actual beliefs and behavior of Less Wrong style rationality and its adherents, would respond along the lines of “oh, I didn’t mean those ‘rationalists’ and that ‘rationality’; heck, I don’t even know what those Less Wrong guys believe”. To my knowledge, he’s never made it clear just who he is against, then. Where, in 2024 (or 2014), is Chapman encountering these people who have no connection to Less Wrong, but self-describe as “rationalists”, talk of “rationality” as their philosophy, etc.? Who even are these mythical people?
It seems very weird to write a whole website-book, in the 21st century, arguing against early 20th-centry logical positivism, of all things. Besides, Chapman often writes as if the “rationalists” he takes as foils are, you know, still around!
I’d say they very much are, they just aren’t as prevalent on Less Wrong (and I think there are still plenty of them on LW!). My experience is that you can’t throw a stone without hitting a logical positivist (even if they don’t know that they are, if you talk to them it’s clear those are their beliefs) in any STEM university department, engineering company, etc.
But what are those beliefs exactly? I mean, the actual, historical, early-20th-century positivists had some pretty specific beliefs, some of which were (now-)clearly wrong in their strongest forms… but do very many people believe those strongest forms now? Or are they logical positivists in the same way that Scott Alexander is a logical positivist?
This is why I find David Chapman’s “vagueblogging” so annoying. This whole conversation doesn’t need to be happening; it could all be avoided if he just, like, linked to specific people saying specific things.
Indeed, even just explicitly saying “logical positivists” instead of “rationalists” would make his writing more clear. Why say the latter if what you actually mean is the former…?
Assuming by “the modern rationality movement” you mean the LessWrong-adjacent subculture, some of what they write is unambiguously meta-rational. The center of gravity is more-or-less rationalism as I use the term, but the subculture is not exclusively that.
Among the (arguably) core LW beliefs that he has criticized over the years are Bayesianism as a complete approach to epistemology, utilitarianism as a workable approach to ethics, the map/territory metaphor as a particularly apt way to think about the relationship between belief and reality.
Yes, I’ve seen that quote; but what it means is that Chapman’s use of the terms “rational”, “rationality”, etc., are so different from ours (on LW) that we have to translate anything he writes before we can understand it.
As for the criticized beliefs—well, I also reject utilitarianism as a workable approach to ethics. So do many people here, I think (though probably not most). Bayesianism as complete approach to epistemology seems like at least a bit of a strawman.
The map/territory one is interesting; I can’t easily predict what that criticism consists of. Do you have any links handy, by chance?
Well, I blame Yudkowsky for the terminology issue, he took a term with hundreds of years of history and used it mostly in place of another established term which was traditionally sort of in opposition to the former one, no less (rationalism vs empiricism).
As I understand it, Chapman’s main target audience wasn’t LW, but normal STEM-educated people unsophisticated in the philosophy of science-related issues. Pretty much what Yudkowsky called “traditional rationality”.
I have to agree with @Richard_Kennaway’s evaluation of the essay. Also, Chapman here exhibits his very common tendency to, as far as I can tell, invent strawman “mistakes” that his targets supposedly make, in order to then knock them down. For example:
Taking maps as prototypes gives the mistaken impression that simply correcting factual errors, or improving quantitative accuracy, is the whole task of rationality.
Maybe someone somewhere has made this sort of mistake at some point, but I can’t recall ever encountering such a person. And to claim that such a mistake arises, specifically, from the map-territory metaphor, seems to me to be entirely groundless.
But of course that’s fine; if I haven’t encountered a thing, it does not follow that the thing doesn’t exist. And surely Chapman has examples to point to, of people making this sort of error…? I mean, I haven’t found any examples, at least not in this essay, but he has them somewhere… right?
Maybe someone somewhere has made this sort of mistake at some point, but I can’t recall ever encountering such a person. And to claim that such a mistake arises, specifically, from the map-territory metaphor, seems to me to be entirely groundless.
I think you should seriously consider you live in a bubble where you are less likely to encounter the vast valley of half-baked rationality. I regularly meet and engage with people who make exactly this class of errors, especially in practice, even if they say they understand in theory that this is not the whole task of LW-style rationality.
Every example Chapman gives there to illustrate the supposed deficiencies of “the map is not the territory” is of actual maps of actual territories, showing many different ways in which an actual map can fail to correspond to the actual territory, and corresponding situations of metaphorical maps of metaphorical territories. The metaphor passes with flying colours even as Chapman claims to have knocked it down.
To me, the main deficiency is that it doesn’t make the possibility, indeed, the eventual inevitability of ontological remodeling explicit. The map is a definite concept, everybody knows what maps look like, that you can always compare them etc. But you can’t readily compare Newtonian and quantum mechanics, they mostly aren’t even speaking about the same things.
As I understand it, Chapman is promoting some form of Buddhism. (I think he might even be a leader of some small sect? Not sure.) The bottom line is already written; now he is adding the previous lines to make it seem like this is something that a sufficiently smart modern thinker would discover independently.
Here he is using an ancient Dark Arts technique, which in our culture is known as Hegel’s dialectic, but it was already used by Buddha—to win a debate, create two opposed strawmen, classify all your competitors as belonging to one or the other, and then you are the only smart person in the room who can transcend the strawmen and find the golden middle way of “it is actually the reasonable parts of this, plus the reasonable parts of that, minus all the unreasonable parts”. Congratulations, you win!
Buddha classified his philosophical/religious competitors into two groups, and Chapman translated one of those words as “rationalists”. (The reference to early 20th-century logical positivism is just another nice trick, where Chapman is promoting an ancient belief, but he is rebranding it as a cool modern perspective, as opposed to the outdated and therefore low-status ideas of positivism.)
I don’t have the energy to get into it in depth, but I think you’re being pretty uncharitable here and it feels to me like you’re trying to weaponize rationalist applause lights. Some quick thoughts on what I think is insufficient about your comment:
You claim he already wrote the bottom line, but you provide no evidence to substantiate that.
You claim that Hegelian dialectic is a dark art technique with no justification.
You makes some claims about what’s written in Buddhists texts, but offer no reference to the specific arguments that were made to justify the claim that he set up strawmen, which would also require proving that they were strawmen at the time, not just now with 2500 years of philosophical progress.
You offer what I guess I can best interpret as an attempt to dunk on Chapman for promoting “ancient” ideas, as if ancient ideas were inherently bad (lots of math is just as old and we still use it every day, so being old is obviously not the problem; would you dunk on someone for promoting the “ancient” belief in the Pythagorean theorem?).
You claim he already wrote the bottom line, but you provide no evidence to substantiate that.
Prediction: No matter how many books or web articles Chapman writes, their conclusions will always support Buddhism. He will not conclude anything fundamentally incompatible with Buddhism.
You claim that Hegelian dialectic is a dark art technique with no justification.
Yes, and I have just explained how exactly it works. Once you see the pattern, it is obvious. Let me show you how it works in practice:
“There are people who believe that Hegelian dialectic is a useful method to transcend our limited beliefs by transcending the thesis and anti-thesis by creating a new and better syn-thesis.
There are also people who believe that Hegelian dialectic is a dark art technique, where people who disagree with the author’s conclusion are sorted into two opposing groups, and then the author’s solution is presented as the middle way superior to both.
My opinion is that both of these people are correct in some aspect, but wrong in some other aspect. The actual deep understanding of Hegelian dialectic is that in some situations it can be used to transcend two existing contradictory beliefs, while in other situations it can be used as a dark arts technique.”
You makes some claims about what’s written in Buddhists texts, but offer no reference
True; that would be too much work. (Probably enough for someone to write a master’s thesis.)
You offer what I guess I can best interpret as an attempt to dunk on Chapman for promoting “ancient” ideas, as if ancient ideas were inherently bad
I don’t mind the ancient ideas, but the rebranding feels a bit dishonest. To use your analogy, it would be like teaching the Pythagorean theorem under a new name, pretending that it was my invention.
If Chapman said plainly that he was repeating an ancient argument about pre-Buddhist “rationalists”, we could have avoided a confusion. But he made it sound like he was making a fresh observation based on current data.
Notice how Chapman reacts to finding out that there are actual rationalists out there who do not fit his definition of “rationalists”. He simply says “those are not the rationalists I was talking about”. And that’s great! But does this new knowledge make him somehow revise his existing conclusions?
I haven’t read the relevant Chapman stuff, but, to be sure, if we look up Rationalism on Wikipedia, it lists Descartes, Spinoza, Leibniz, and Kant, devoting many paragraphs to their views. Only in one sentence at the end does it mention Less Wrong: “Outside of academic philosophy, some participants in the internet communities surrounding Less Wrong and Slate Star Codex have described themselves as “rationalists.”″ It doesn’t even mention Yudkowsky, Hanson, or Scott Alexander by name.
The point is, there exists some academic context in which “rationalism” has a preexisting meaning that refers to those 1600s-1800s people, and probably not to Less Wrong people. So, when Chapman writes about “rationalists”, is it possible he’s acting like he’s in that context, and talking about the pre-1900 people?
Doing Google searches with site:meaningness.com, I get “descartes” → 8 results, “spinoza” → 2 results, “leibniz” → 1 result [though it’s about calculus], “kant” → 21 results; meanwhile, “yudkowsky OR eliezer” → 10 results, “less wrong” → 5 results, “hanson” → 3 results… and “scott alexander” OR “slate star codex” OR “astral codex ten” → 31 results!
Hmmph. I guess he talks about both. It would require actually reading the blog to judge what he means when he says “rationalist” and whether he’s consistent about it. I’ll let Viliam report on this.
… And I see a further result, from what seems to be a book Chapman wrote, bold added by me:
The book uses “rationality” to refer to systematic, formal methods for thinking and acting; not in the broader sense of “any sensible way of thinking or acting,” as opposed to irrationality.
“Rationalism” refers to any belief system that makes exaggerated claims about the power of rationality, usually involving a formal guarantee of correctness.
Oh, dear. You define a term like that based on whether the claims are exaggerated vs accurate? That seems like a recipe for generating arguments about whether something qualifies as “rationalism”. (If Descartes writes an essay about the power of rationality, and some of the claims are exaggerated while others are correct, does that mean the essay is partly rationalism and partly not? And, obviously, if we disagree about something’s correctness, then that means we’d disagree about whether it’s “rationalism”. I have the impression that people often don’t try to label philosophical ideas beyond who wrote them and when, and any voluntary self-labeling the author did; this type of thing is probably why.)
I’ve now updated to find Said and Viliam’s complaints very plausible.
I’ve only read some of his work, but I didn’t walk away with a sense he wants everyone to be Buddhist. My sense was more that he was pushing back against things he didn’t like within Buddhism, including changes it made to become more memetically fit.
Probably the strangest thing about Chapman’s writing has always been the way that he would rail against “rationalists” and “rationality”, and then, when it was pointed out that his characterization doesn’t match the actual beliefs and behavior of Less Wrong style rationality and its adherents, would respond along the lines of “oh, I didn’t mean those ‘rationalists’ and that ‘rationality’; heck, I don’t even know what those Less Wrong guys believe”. To my knowledge, he’s never made it clear just who he is against, then. Where, in 2024 (or 2014), is Chapman encountering these people who have no connection to Less Wrong, but self-describe as “rationalists”, talk of “rationality” as their philosophy, etc.? Who even are these mythical people?
In the sidebar of his website, the section “Positive and Logical” under “Part One: Taking Rationalism Seriously” says
It seems very weird to write a whole website-book, in the 21st century, arguing against early 20th-centry logical positivism, of all things. Besides, Chapman often writes as if the “rationalists” he takes as foils are, you know, still around!
I’d say they very much are, they just aren’t as prevalent on Less Wrong (and I think there are still plenty of them on LW!). My experience is that you can’t throw a stone without hitting a logical positivist (even if they don’t know that they are, if you talk to them it’s clear those are their beliefs) in any STEM university department, engineering company, etc.
But what are those beliefs exactly? I mean, the actual, historical, early-20th-century positivists had some pretty specific beliefs, some of which were (now-)clearly wrong in their strongest forms… but do very many people believe those strongest forms now? Or are they logical positivists in the same way that Scott Alexander is a logical positivist?
This is why I find David Chapman’s “vagueblogging” so annoying. This whole conversation doesn’t need to be happening; it could all be avoided if he just, like, linked to specific people saying specific things.
Indeed, even just explicitly saying “logical positivists” instead of “rationalists” would make his writing more clear. Why say the latter if what you actually mean is the former…?
Here’s Chapman’s characterization of LW:
Among the (arguably) core LW beliefs that he has criticized over the years are Bayesianism as a complete approach to epistemology, utilitarianism as a workable approach to ethics, the map/territory metaphor as a particularly apt way to think about the relationship between belief and reality.
Yes, I’ve seen that quote; but what it means is that Chapman’s use of the terms “rational”, “rationality”, etc., are so different from ours (on LW) that we have to translate anything he writes before we can understand it.
As for the criticized beliefs—well, I also reject utilitarianism as a workable approach to ethics. So do many people here, I think (though probably not most). Bayesianism as complete approach to epistemology seems like at least a bit of a strawman.
The map/territory one is interesting; I can’t easily predict what that criticism consists of. Do you have any links handy, by chance?
Well, I blame Yudkowsky for the terminology issue, he took a term with hundreds of years of history and used it mostly in place of another established term which was traditionally sort of in opposition to the former one, no less (rationalism vs empiricism).
As I understand it, Chapman’s main target audience wasn’t LW, but normal STEM-educated people unsophisticated in the philosophy of science-related issues. Pretty much what Yudkowsky called “traditional rationality”.
The map/territory essay: https://metarationality.com/maps-and-territory
Thanks for the link!
I have to agree with @Richard_Kennaway’s evaluation of the essay. Also, Chapman here exhibits his very common tendency to, as far as I can tell, invent strawman “mistakes” that his targets supposedly make, in order to then knock them down. For example:
Maybe someone somewhere has made this sort of mistake at some point, but I can’t recall ever encountering such a person. And to claim that such a mistake arises, specifically, from the map-territory metaphor, seems to me to be entirely groundless.
But of course that’s fine; if I haven’t encountered a thing, it does not follow that the thing doesn’t exist. And surely Chapman has examples to point to, of people making this sort of error…? I mean, I haven’t found any examples, at least not in this essay, but he has them somewhere… right?
I think you should seriously consider you live in a bubble where you are less likely to encounter the vast valley of half-baked rationality. I regularly meet and engage with people who make exactly this class of errors, especially in practice, even if they say they understand in theory that this is not the whole task of LW-style rationality.
Sure, that’s possible. Do you have any links to examples?
Every example Chapman gives there to illustrate the supposed deficiencies of “the map is not the territory” is of actual maps of actual territories, showing many different ways in which an actual map can fail to correspond to the actual territory, and corresponding situations of metaphorical maps of metaphorical territories. The metaphor passes with flying colours even as Chapman claims to have knocked it down.
To me, the main deficiency is that it doesn’t make the possibility, indeed, the eventual inevitability of ontological remodeling explicit. The map is a definite concept, everybody knows what maps look like, that you can always compare them etc. But you can’t readily compare Newtonian and quantum mechanics, they mostly aren’t even speaking about the same things.
Switching from a flat map drawn on paper (parchment?), to a globe, would be an example of ontological remodeling.
As I understand it, Chapman is promoting some form of Buddhism. (I think he might even be a leader of some small sect? Not sure.) The bottom line is already written; now he is adding the previous lines to make it seem like this is something that a sufficiently smart modern thinker would discover independently.
Here he is using an ancient Dark Arts technique, which in our culture is known as Hegel’s dialectic, but it was already used by Buddha—to win a debate, create two opposed strawmen, classify all your competitors as belonging to one or the other, and then you are the only smart person in the room who can transcend the strawmen and find the golden middle way of “it is actually the reasonable parts of this, plus the reasonable parts of that, minus all the unreasonable parts”. Congratulations, you win!
Buddha classified his philosophical/religious competitors into two groups, and Chapman translated one of those words as “rationalists”. (The reference to early 20th-century logical positivism is just another nice trick, where Chapman is promoting an ancient belief, but he is rebranding it as a cool modern perspective, as opposed to the outdated and therefore low-status ideas of positivism.)
I don’t have the energy to get into it in depth, but I think you’re being pretty uncharitable here and it feels to me like you’re trying to weaponize rationalist applause lights. Some quick thoughts on what I think is insufficient about your comment:
You claim he already wrote the bottom line, but you provide no evidence to substantiate that.
You claim that Hegelian dialectic is a dark art technique with no justification.
You makes some claims about what’s written in Buddhists texts, but offer no reference to the specific arguments that were made to justify the claim that he set up strawmen, which would also require proving that they were strawmen at the time, not just now with 2500 years of philosophical progress.
You offer what I guess I can best interpret as an attempt to dunk on Chapman for promoting “ancient” ideas, as if ancient ideas were inherently bad (lots of math is just as old and we still use it every day, so being old is obviously not the problem; would you dunk on someone for promoting the “ancient” belief in the Pythagorean theorem?).
Prediction: No matter how many books or web articles Chapman writes, their conclusions will always support Buddhism. He will not conclude anything fundamentally incompatible with Buddhism.
Yes, and I have just explained how exactly it works. Once you see the pattern, it is obvious. Let me show you how it works in practice:
“There are people who believe that Hegelian dialectic is a useful method to transcend our limited beliefs by transcending the thesis and anti-thesis by creating a new and better syn-thesis.
There are also people who believe that Hegelian dialectic is a dark art technique, where people who disagree with the author’s conclusion are sorted into two opposing groups, and then the author’s solution is presented as the middle way superior to both.
My opinion is that both of these people are correct in some aspect, but wrong in some other aspect. The actual deep understanding of Hegelian dialectic is that in some situations it can be used to transcend two existing contradictory beliefs, while in other situations it can be used as a dark arts technique.”
True; that would be too much work. (Probably enough for someone to write a master’s thesis.)
I don’t mind the ancient ideas, but the rebranding feels a bit dishonest. To use your analogy, it would be like teaching the Pythagorean theorem under a new name, pretending that it was my invention.
If Chapman said plainly that he was repeating an ancient argument about pre-Buddhist “rationalists”, we could have avoided a confusion. But he made it sound like he was making a fresh observation based on current data.
Notice how Chapman reacts to finding out that there are actual rationalists out there who do not fit his definition of “rationalists”. He simply says “those are not the rationalists I was talking about”. And that’s great! But does this new knowledge make him somehow revise his existing conclusions?
I haven’t read the relevant Chapman stuff, but, to be sure, if we look up Rationalism on Wikipedia, it lists Descartes, Spinoza, Leibniz, and Kant, devoting many paragraphs to their views. Only in one sentence at the end does it mention Less Wrong: “Outside of academic philosophy, some participants in the internet communities surrounding Less Wrong and Slate Star Codex have described themselves as “rationalists.”″ It doesn’t even mention Yudkowsky, Hanson, or Scott Alexander by name.
The point is, there exists some academic context in which “rationalism” has a preexisting meaning that refers to those 1600s-1800s people, and probably not to Less Wrong people. So, when Chapman writes about “rationalists”, is it possible he’s acting like he’s in that context, and talking about the pre-1900 people?
Doing Google searches with site:meaningness.com, I get “descartes” → 8 results, “spinoza” → 2 results, “leibniz” → 1 result [though it’s about calculus], “kant” → 21 results; meanwhile, “yudkowsky OR eliezer” → 10 results, “less wrong” → 5 results, “hanson” → 3 results… and “scott alexander” OR “slate star codex” OR “astral codex ten” → 31 results!
Hmmph. I guess he talks about both. It would require actually reading the blog to judge what he means when he says “rationalist” and whether he’s consistent about it. I’ll let Viliam report on this.
… And I see a further result, from what seems to be a book Chapman wrote, bold added by me:
Oh, dear. You define a term like that based on whether the claims are exaggerated vs accurate? That seems like a recipe for generating arguments about whether something qualifies as “rationalism”. (If Descartes writes an essay about the power of rationality, and some of the claims are exaggerated while others are correct, does that mean the essay is partly rationalism and partly not? And, obviously, if we disagree about something’s correctness, then that means we’d disagree about whether it’s “rationalism”. I have the impression that people often don’t try to label philosophical ideas beyond who wrote them and when, and any voluntary self-labeling the author did; this type of thing is probably why.)
I’ve now updated to find Said and Viliam’s complaints very plausible.
I’ve only read some of his work, but I didn’t walk away with a sense he wants everyone to be Buddhist. My sense was more that he was pushing back against things he didn’t like within Buddhism, including changes it made to become more memetically fit.