The randomized experiment you did is so cool! I love experiments like this -- like the Allais Paradox -- that reveal biases and irrationalities.
But in this case, as a game theorist, I aver that the switchers are correct. This is a classic application of Schelling points. If we all have the altruistic utility function "minimize death" then there are two Nash equilibria: (1) Everyone take the pill, and (2) no one take the pill. If the question is framed such that everyone taking the pill (like when it's just pressing one of two buttons) is focal, that's the equilibrium you expect and the one you rationally adopt yourself. When it's framed the other way, you expect the other equilibrium and rationally don't take the pill.
PS: I just realized that this is also the exact problem you called out in encouraging us to get set up on kbin for comments. The two equilibria are "no one else is going to set up an account so I won't bother to either" and "that's the new Schelling point for Dynomight comments so of course I want to be there too!"
PPS: Taking you up on your self-promotion encouragement, I predict that Dynomight readers will like the Beeminder blog, maybe especially the posts tagged "rationality". (Also I'm excited to see some of you at Less Online!)
On the methodology, can I ask what is the reason to randomise based on the answer to a presumed uncorrelated question with expected even groups (birth month) instead of just assigning survey responders at random to one of the two groups? Is it to control for people responding to the survey multiple times or something like that?
Haha, no good reason. I just did it that way because I couldn't find any way to randomize in cryptpad. So I chose something effectively random but natural enough that you wouldn't suspect I was using it for the purpose of randomization.
Totally agree there's a strong case for switching being the right choice along the lines you argued. Although it's a little funny—switching is the right choice on an individual level because so many people switch on a group level! I think that the "never take the pill" camp might argue their "no one take the pill" equilibrium is "better" since it's compatible with our individual incentives? I can certainly see someone saying "if y'all want to screw around with voluntarily swallowing explosive pills, OK, but don't ask me to admire you for it".
Absolutely, no swallowing seems the superior equilibrium here. But what if we adjust that 50% threshold so that we all survive as long as at least two of us take the pill? If one person dies you're gonna feel pretty rotten, right? You could've single-handedly saved them! And they only took the pill out of exactly that fear.
Or what if we all get a nice prize if enough people take the pill? Now we might want to shout down the otherwise reasonable "don't screw around with voluntarily swallowing explosives" argument.
As I wrote in the "additional wisdom" box on the survey, I got the yellow/green button version of the question, and I was so befuddled as to why anyone would choose the yellow button or even think of such a question that I assumed dynomight had misstated it. I thought maybe they had forgotten to mention that if everyone chose the green button then everyone would die, or something like that. That would probably change my mind. (Adjusting the threshold wouldn't.)
I've been posing this to friends/family and am getting that reaction a lot. How about this variant:
Take the red pill, labeled "look out for number one" with a picture of Voldemort on it, and be guaranteed to be one of the survivors.
Take the blue pill, labeled "zero deaths via cooperation" with a picture of Mr Rogers on it, and survive if and only if enough other people also take the blue pill.
Does that make the all-blue equilibrium focal enough?
Or what if the threshold is 100%, where literally everyone has to take the blue pill for everyone to be saved? And suppose you like this group of people. Taking the red pill might singled-handedly cause everyone else to die.
I guess the point of all of these is just that you really, really want to do the same thing everyone else is doing.
This is pretty central to such rationality fare as Meditations on Moloch and Inadequate Equilibria.
RE using an editor, it's not clear to me that editors are that helpful. I haven't really looked into this, but I was just thinking recently about how this relates to two of my favorite writers, Paul Graham and Scott Alexander.
Paul Graham spends a long time revising his essays an usually has 3 to 6 people review drafts before publishing them.
Scott Alexander does not have any editors, and often doesn't even revise his essays at all before publishing (IIRC).
The lack of an editor doesn't seem to be hurting Scott. On the other hand, almost all "traditional" writers use editors, even the extremely talented ones, so maybe they're onto something.
RE consciousness progress: IMO people in our circles tend to have too much of a "science good, philosophy bad" bias. People are underrating how much progress philosophy has made in the last 100 years, and I expect progress to continue for the next 50 years. I don't think neuroscience can tell us anything about consciousness, and the only thing it's done so far is provide evidence against theories that were already basically proven false by pure a priori philosophy. (If you scan people's brains, you can see that certain brain areas correspond to certain thoughts! This proves that the brain is involved in thinking! But we already knew that! And this doesn't actually provide evidence for materialism/against dualism!)
Aside: In dark mode, the bold font is hard to distinguish from the regular font, so I found it a bit hard to follow the question/answer section of the post.
Re editors: I don't think they're necessarily better for everyone. But Scott Alexander's writing advice seems to boil down to:
Start with a natural verbal ability 5 standard deviations about the population average.
Write down your thoughts exactly as if you were talking all in one go without editing.
Observe that your first draft is flawless and beautiful, done!
This... doesn't seem like an option for me.
(On the other hand, I doubt Paul Graham's process would work for me either. I like his writing a lot, but he tend to write in a "voice of god" and make a lot of pronouncements of Truth, while it's more my style revel in uncertainty, which rules out a lot of clean and tidy narratives.)
Re consciousness and philosophy: I think the outside view is that so far philosophy has actually made quite a lot of progress on philosophy (without philosophy we basically wouldn't even have a language to discuss it!) so why not expect that to continue?
I think that it's dumb to limit it in specific fields; Neuroscience, philosophy and (potentially, though I'm not all that sure of this) software engineering in the form of AI will all make small, measured amounts of innovations. Human knowledge isn't linear; philosophy doesn't exist in a vacuum, and neither do all of the other fields. Progress in cow breeding might, in a convoluted series of ideas & sparks, lead to a breakthrough in neuroscience ten years down the line.
All these words to say I think it's complicated and not an A/B black & white thing, haha.
What problem do you have with Eliminative Materialism? It seems to me the most probably 'true' of all descriptions. That consciousness will fall away like phlogiston etc. Identity theory seems to just want to hold onto our own ignorance instead of replacing it with more robust concepts.
I don't know for sure that the physical world exists. I don't know that anyone else is real. Maybe I'm in a simulation. Or maybe I'm an distributed alien spider intelligence and my whole life is a weird psychedelic trip. Who knows! But if I were to name the single fact that I have the most confidence in, it's that I'm conscious. So in some sense I actually find it hard to come up with a position about anything that's less plausible than eliminative materialism.
Please ignore the survey results and cater to me specifically by writing more book reviews.
Thanks.
(but really your book reviews are so good, don't listen to the 127/966 haters. I was pretty surprised seeing that topping the "write less" chart because your Bourdieu and Parfit posts are some of my favorites)
Aww, thank you. I think I might just do that because I've found writing book reviews to be amazing for improving my understanding of books I really want to understand. Also, for you specifically (and also because I've already started writing it) there's even more Bourdieu coming.
The most popular option is identity theory—which I find notable mostly for somehow managing to be even worse than full-bore eliminative materialism. Since I can’t emotionally accept that, I choose to see this as an endorsement of sensible materialism (“thinking is a physical process that is done by your brainmeat and in principle is fully explained by physics, etc.”) rather than actual psycho-crazy identity theory (“the feeling of seeing a red flower—the feeling itself—is in the same category as a pebble or tornado or whatever”) . I do this even though the phrasing of the question seems unambiguous.
It did not seem unambiguous to me. I voted as if you meant "psycho-crazy identity theory"[1] because I'm familiar enough with identity theory to know that that's what it is, but your description read to me like you meant "sensible materialism".
[1] I would disagree with the characterization of identity theory as "psycho-crazy". I still think it's wrong but it's not crazy. I understand it as an attempted solution to things like the "China brain" thought experiment—it's an attempt to classify as non-conscious things that intuitively seem like they ought not to be conscious.
That is reassuring! Though it is quite tricky to see how to convey exactly what identity theory is in just a few words.
And, OK, I admit that I can't really defend "psycho-crazy" as a fair and neutral description. It does seem completely crazy to me, but then it seems like any explanation of consciousness would be quite crazy, so maybe best not to throw too many stones unless you have a particular view you want to defend. (After all, I'm open to variants of panpsychism, which not a few people find to be on the crazy side!)
I can't remember how I answered that question, but I for sure didn't understand the description (and honestly still don't really), so ascribing this answer's popularity to people misunderstanding it seems reasonable to me!
Given there doesn't seem to be a robust and commonly understood definition of 'conscious' or 'consciousness' it seemed obvious that question would be misunderstood (if it's understandable at all).
a lot of uncertainty in the consciousness questions
so much support for the one option that doesn't even pretend to answer the question (at least not without some sort of panpsychism or similar attached)
I really can't understand the amount of optimism that progress will be made on this question. What possible things could be observed that would help?
Of all the sections, the answers on consciousness are the furthest from my own views. So it's also quite hard for me to see how exactly AI or neuroscience would tell us too much about consciousness. Maybe 100 years from now we have a giant weakly interconnected solar-system spanning AI and we ask it, "are you conscious?" and "do you feel like you have a single self or are you more distributed?"? Would those answers tell us anything? It's tempting to think, "of course those answers have fully materialist answers" so they couldn't tell us anything, but not sure we'd actually react that way in practice...
Say what you will about him (and I have) but I agree with Sam Harris:
Whatever the explanation for consciousness is, it might always seem like a miracle.
And for what it’s worth, I think it always will seem like a miracle.
Agreed, all conceivable explanations seem so strange, should avoid being too critical! Incidentally, I'm a big fan of "Conscious" by Annaka Harris: https://annakaharris.com/conscious/ (At least for consciousness noobs, I think this might be the best book?)
Either you missed this, or you did, in fact, know this reference and were playing dumb in the post to mess with people. If that's the case I make no apologies for ruining your fun. The public must be correctly informed about such matters.
I'm not playing dumb, I'm actually dumb. (I've published a lot of Cumberbatches in the past, none with penguins.) In fact, in the survey wisdom field, a lot of people wrote weird things like "pengeeeens" but I didn't understand what the hell they were talking about, so it wasn't until someone emailed me after this post that I found out about that meme. Personally, I think it makes things much less funny, but I suspect your explanation is factually correct...
This subject probably deserves a deeper analysis: some jokes are funnier when you don't get them.
I have some examples of jokes I didn't get:
This Abstruse Goose comic explaining magnets ends with "fucking magnets, how do they work?" I thought it was funny because it was a complex technical explanation followed by a casual and expletive-laden summary and the contrast was funny. Later I learned it's a song reference, which is less funny.
I saw a comedy sketch on TV featuring Satan and someone called Satan on the phone and he answered, "Hello, this is Beelzebub Satan." I thought, ha, what a ridiculous and silly first name for Satan, that is a funny joke! Later I learned that is an actual name that some people use to refer to Satan.
There's a part of the Weird Al song "Alburquerue" that does like this:
So I got in my car and I drove over to the donut shop
And I walked on up to the guy behind the counter
And he says, "Yeah, what do ya want?"
I said, "You got any glazed donuts?"
He said, "No, we're outta glazed donuts."
I said, "You got any jelly donuts?"
He said, "No, we're outta jelly donuts."
I said, "You got any Bavarian cream-filled donuts?"
He said, "No, we're outta Bavarian cream-filled donuts."
I said, "You got any cinnamon rolls?"
He said, "No, we're outta cinnamon rolls."
I said, "You got any apple fritters?"
He said, "No, we're outta apple fritters."
I said, "You got any bear claws?"
He said, "Wait a minute, I'll go check
I thought the joke was that the narrator was so desperate for food that he was willing to eat a literal claw of a bear, and on top of that, the shopkeeper thought they might actually have one. Later I learned that a bear claw is a type of pastry.
Not me, but this reddit post of a twitter screenshot describing someone's friend's misconception about an ad campaign.
I don't have a super strong opinion on the Aspartame Question, however in the spirit of giving you wisdom (I hope) I do have one objection:
C) The metabolism of aspartame is very well-understood and seems to exclude the possibility of it causing obesity (or basically anything else).
I think this reasoning is leading you astray! It doesn't have to metabolise to specific compounds to have an affect on the body. The most obvious pathway for a causal effect: it activates your tastebuds, which send signals to your brain. I think that is in itself enough of an argument to explain how it could conceivably have causal effects that have nothing to do with what happens to the molecule later in metabolization. If you want a more concrete story for how that could cause obesity, it could be something like that the registration of a sweet taste makes the brain trigger the release of insulin, or some other regulatory mechanism, that messes you up when there's no actual sugar to metabolize or otherwise respond to. And anecdotally this seems to be the case for me, I feel an annoying, if subtle, pang of hunger after consuming diet soda that causes me to want to eat something to make it stop (which is why I quit drinking diet soda).
As for self promotion, hope this is ok: I do custom software / web development for people, if anyone has an idea for something they you want to build and need a contractor, please reach out and we can have a chat!
I agree that the mechanism you're suggesting is (by far!) the most plausible mechanism by which diet soda could potentially cause obesity. And certainly aspartame hits our tastebuds and this affects our brains (that's kinda the whole point) and we certainly don't understand the mechanism of anything that happens after that.
But... it's been tested and doesn't actually seem to happen in practice. http://orcid.org/0000-0003-3525-3473 If you have people drink artificially sweetened drinks vs. nothing, it doesn't seem to do anything. And—most important to me—if you have people substitute sugar-sweetened drinks for artificial sweeteners, this pretty clearly causes weight loss.
It's possible these are wrong, but I think that's the direction the evidence points to now. I wouldn't tell anyone to consume aspartame, but if you're thinking about substituting aspartame for sugar this seems like one of the easiest wins out there.
Re: utilitarianism ...
You could say I was a professional utilitarian for 30 years (https://en.wikipedia.org/wiki/Matt_Ball). But writing my latest (https://www.losingmyreligions.net/ - free version) led to working through problems that had been bubbling up in my consciousness. I now believe I was wrong. But mine certainly isn't a view I see anywhere! (Specifically, the "I Welcome Our Robot Overlords" and "Biting the Philosophical Bullet" chapters.)
Fighting words! I thought—but I think I largely agree with you. I feel like the purpose of ethics is to interpolate/extrapolate from our ethical intuitions, and so I just don't see why I should accept an ethical system that leads to crazy results like the repugnant conclusion. But I'm still very utilitarianism friendly and I think it works pretty well for the kinds of situations we can actually encounter (except perhaps for decisions that affect if other people/animals will exist or not).
Well, if I had to accept one of those, I guess it would be the third? I guess my moral intuition is just strongly that there's some kind of diminishing returns in terms of much better the world is as you add additional people. Adding 1 billion happy people to a world with 100 happy people is much much better than adding 1 billion happy people to a world that already has 100 billion happy people. If there were some sort of proof that accepting any variant of consequentialism would force me to accept the repugnant conclusion, then probably my response to that would be to stop believing in consequentialism. So I think that's always an option?
You mean why is there a different big header image behind the title for each post? The truth is, it's just because I long ago picked a website template that implied you're supposed to do that. I sometimes regret it because it takes a bit of time. But perhaps it makes it easier to distinguish old articles?
No, I mean when I check for new posts and there is one it will have an initial image. If I check back later, the image is almost always different and it typically seems to stay on this second image in perpetuity.
Did you disaggregate the "beat you in a fight" by sex? I'm extroverted, but solidly into "probably not" because I'm a woman and realistic. but depending on if the m/f i/e split has a gap, that could impact results
I'm an E who said I couldn't fight you because I'm a relatively small female and I'm just assuming you aren't. It's the safest assumption for me, in general. Do you have an idea of the femaleness or size of your readership?
I'm not sure about femaleness, though I'd guess it's a minority due to the survey results for other blogs with high overlap in readers, e.g. slatestarcodex.
For the size of the readership, the only hard number I have is that currently there are 7965 "subscribers" on substack. But how many people actually read it? And how many people subscribe via RSS instead? (This question suggests maybe it's about equal.)
Consider the people that neither subscribe via email nor RSS, too. I was part of that group for a solid year or so; I just went to the website once in a while. I think a statistically significant number of people do this, but I have no data to back it up.
Yeah, that's probably a fair number of people. If there's no viral article, the main page of dynomight.net typically gets ~100 visits per day. (At least, excluding people who have browsers that block goatcounter, which is basically anyone who has any kind of adblocker or tracking protection.
OK, I've updated everything to have a white background. (Not entirely sure this is the best way to deal with dark mode, but at least it's readable.) Should be visible within a few minutes.
Add comment