Curious what you think of Prometheus Rising in relation to this (if you've read it)? If not: I've only read the first 5 chapters so far, but the exercises he suggests at the end of chapters feel very aligned with what you describe here. Eg in the chapter about "what the thinker thinks, the prover proves" he talks about deeply believing and visualizing that you will find a quarter on the street and seeing how long it takes to do so. Then varying the explanation for this phenomenon you believe in, seeing how that changes the time to get a second quarter etc.
The hostile telepaths thing is an interesting case for me actually. Reading your lw post when it came out has made a significant impact on my perception of this phenomenon, I believe. But it's pretty hard to put into words what exactly changed. I think it started with just noticing self deception in response to a hostile telepath after the fact. Moments where I would recover hidden information after the person left, combined with expectations that that information would be poorly received by said person. After a while I started flagging self-deception happening in the moment. Not that I would uncover the information itself, but a felt sense that it was happening based on a combination of signals. Part of it is a particular flavor or somatic tone, nervous system activation, feeling unsafe and inadequate with the timing suggesting it's related to something the other person did or said. That last bit obscures some of it: I think I experience a 'gnosis' that the change in my system related to a particular expression of the other person, but I'm not 100% sure how I get to that sense. Then there is the part where I can somewhat gauge if there is some information that feels dangerous by reflecting on the thing that just transpired. Often I will naturally start to wonder about how I feel about something, and that process is cut short with a strong sense of fear or danger. Something opaque happens here too, but in practice it feels obvious when I connect with dangerous information by the shape of my experience. It has a particular flow and vibe and tone to it. Sequence of events. Content. All factors into it I think, but I can't really give a formula for it. Another thing that seems to happen reliably is that my image of the other person changes when I'm supressing something. I feel more distant from them, they look less beautiful, more threatening, less nice. I start spontaneously questioning whether I enjoy being with them more.
I'm not thrilled with the vagueness of this description, but it's also not obvious how to improve it. I think very crisply describing exactly what happens is a dead end. Too much experience to keep track of, some of it probably not conscious even in theory.
You know, my parents were reading Prometheus Rising when I was little, but I never picked it up myself. I never actually knew what it was about. I just now looked it up. Now I'm pretty curious! Thanks for bringing it back to my attention.
I relate a lot to what you describe about hostile telepath type self-deception. I've come to explicitly tag when I notice an antimemetic effect, like my thinking becomes extremely thick and I start having a lot of trouble finding clarity and I even lose track of what it was I was trying to say or what the problem was. Thanks for sharing your experience; I find anecdotal stuff like this super interesting and relevant!
Another book that comes to mind is Assessing Adult Attachment by Crittenden. I'm not sure where to place it in this discussion, because it starts from a mostly objective way of working: the therapist or AAI scorer trying to outline a person's attachment strategies as a third party. I don't think they go as far as calling such a scoring objective, but the aim is to have consistency between professionals scoring the same data. On the other hand, a lot of it goes into handling reflexivity more rigorously. Both interpersonal and intrapersonal in the subject. For example, there are passages where they discuss looking for specific (emotional) responses in the assesser, eg when you feel a tendency to help or defend or side with the client which can be indicative of a helplessness strategy being active. Intra personal you have things like watching for micro expressions in speech that indicate distortion of information. Or somatic symptoms (eg cough) that pop up repeatedly in key moments that make talking harder/impossible, a indication that intolerable information might being accessed iirc.
It's actually quite intriguing how subtle the signs get. Even the structure/order in which memories are related can be meaningful. I am recalling my attachment therapist (who worked with Crittenden a lot) telling me that my detailed strictly chronological recounting of a traumatic event is common in people who are in a particular stage of processing that trauma. She couldn't give a deeper explanation than, it's somehow important for you to hold on to the memory like this, but her experience was that it will often change as integration progresses.
Overall things like which information is available, how it is accessed, which types of details are recounted, the types of memory (semantic, episodic etc), whether affect memory is present, which points of view are included/excluded etc are all seen as highly relevant to the underlying processes in the subject that are invisible on the surface. The book is more explicit and rigorous in this than I have been able to absorb, and from what I've heard (from earlier mentioned therapist) most of it is in Crittenden's head. She's only been able to codify parts of it.
I get the sense that there is quite a rich sense of non obvious parameters that contribute to psychological processes related to attachment, but at the same time that the thing is so complex that a model that's both complete and rigorously verifiably correct is quite far away.
> This sounds simple, but it was revolutionary. Alchemists certainly did experiments, but they often interpreted results through pre-existing frameworks... When an experiment didn't work, they might blame [..]. Early chemists began to treat unexpected results as potentially valuable information about how nature actually works, rather than as failures to achieve a predetermined goal.
It feels there is something very deep behind this mindshift. It increases the learning signal density per observation. And the "alchemist"/"scientist" mind stance differentiation seems widespread.
Encountering a bug in your program, answering a question from the textbook incorrectly, or failing to apply self-help advice. Curiosity in learning from negative results brings you much closer to almost any goal in the above endeavors (or "dissolving" a goal). However, the human default seems to ignore large or whole parts of the negative experience.
And this scientific mind stance seems really hard to notice and switch to, if you are in a "alchemist" mode!
1. There is a reflex of feeling bad that covers nodes linked to negative outcomes with 'ugh fields'.
2. And there is resistance towards switching to enjoying leaning towards contact with negative: "but I care about [working program/answering textbook questions right/being a good person/getting gold] - why should I seek leaning towards something that is the OPPOSITE? You want me to stop caring about my GOAL? No, thank you very much!"
Maybe this shift truly happens from some larger viewpoint or worldview?
Yeah, I quite agree with you. I'd love to understand this whole phenomenon better. It's obviously very sensible to focus on solving a problem you have, but at the same time there's something about being "solve this problem" oriented that can actually hinder understanding… which ironically can make some problems harder to solve!
I'm reminded of Joe Hudson's VIEW framework, specifically Wonder (the "W" in "VIEW"). It's remarkable how well it works. And given how remarkably well it works, it's just as remarkable that it's something people have to TRY to do!
I bet this puzzle is related to why the scientific revolution took so long to arise in the first place. People have been curious about things for at least tens of thousands of years. People have been very, very smart for at least that long too — including at modern levels of intelligence! We spread out of Africa something like 50,000 years ago. We started building civilizations about 12,000 years ago. But somehow it wasn't until about 400 years ago that we thought to systematize this style of investigation.
So somehow this mindset shift is a kind of slippery. But apparently not impossibly slippery! We really have nailed down a LOT with objective science.
Which is a big part of why I have hope that we can maybe actually expand the mindset, and the skills involved with cultivating it. It would be so, so good to have mainstream culture based less on forcing certain conclusions and more on being really deeply curious about what the systems we're all embedded in really are and how we might make them good.
In short: I like your points and your question. I have thoughts but no definitive answers — appropriately! :-D
What struck me here is your analogy to chemistry’s transition from alchemy. The point wasn’t that early chemists had a better theory than alchemists — it was that they had a better practice: precise observation, reproducibility, curiosity about failures as well as successes. That really resonates with your suggestion that subjective science has to start from reproducible “state shifts.”
Where I get stuck (and maybe you’re circling this too) is measurement. Chemistry advanced once people could weigh and measure reliably. For subjective science, what’s the equivalent of a scale or a beaker? Is it language, metaphor, physiological readouts, intersubjective agreement, or some yet-to-be-invented practice?
I’d love to see you expand on what a “crucial experiment” might look like here. What’s the equivalent of aqua regia dissolving gold — the phenomenon so striking that it forces everyone to take note, even without a theory?
I agree, subjective measurement is tricky. I could have articulated the problem a bit more clearly in this post. Measurement tends to focus on objectivity, but that's exactly what we have to go beyond when doing a science of subjectivity.
As to how to do that? Well, I have a couple of thoughts.
One is that there's nothing magical about measurement for its own sake. It's a solution to a problem: how do we tell exactly what's going on (e.g. in chemical reactions)? Objective measures were a good solution to that problem in objective science. There's a similar problem in subjective science. If we attend to what solves the problem (rather than asking how to make subjective science valid by force-fitting it to the FORMAT of objective science), then we're way more likely to develop an appropriate answer.
Another is that we already have subjective measures so to speak. This is a bit like saying we're born with objective measures: we have eyes, and hands, and a sense of how heavy or light things are, etc. We built on those to refine objective measures. We have the same thing for subjectivity: how do you know when you're happy, or in pain, or feeling impacted by a story you've read? That's the input we need to refine — which means figuring out WHICH PARTS to refine! (This was totally an issue in objective science: why aren't things like hue or texture considered as important to measure as volume and mass? That arose from refining THEORY as well as measurement precision!)
In particular, I really don't think it makes sense to START by pushing subjective measures into objectivity. Intersubjective agreement is a way of making what people are saying independent of an individual's first-person impressions. But an individual's first-person impressions is precisely what we're studying! Likewise things like physiological readouts. It's extremely understandable that people want to suggest these things as ways of doing precise measurement in subjective science. But I think it's a lot like the story of the guy looking for his car keys where the light is good instead of where he dropped them.
My guess is that there are METHODS of refining subjective precision, and there's some objectivity to which methods work well. Things like concentration meditation seem to help add immense detail for instance. There are also some framing tricks like learning to view cognition in third person. Once multiple people develop their skill with those methods, the can compare notes about what they're observing in their respective subjectivities. That might start to make some general (objective) trends about how subjectivities work more apparent. At that point we'd do a blend of subjective perception and objective measurement to really powerfully refine our models (and individuals' ability to explore their own subjectivities).
A bunch of spiritual traditions basically claim to have done this. Buddhism often makes this claim, kind of saying that that's what the Buddha did to attain enlightenment. But even if so, the scientific approach didn't survive. And I think that approach actually matters more than the specific conclusions.
> I’d love to see you expand on what a “crucial experiment” might look like here. What’s the equivalent of aqua regia dissolving gold — the phenomenon so striking that it forces everyone to take note, even without a theory?
We actually have a bunch of vivid examples of both of these.
There's a pivot point in children's development right around age four when they realize that knowing is a property of PEOPLE, not of the world. And that different people can know different things, including believing some false things. In the psych literature it's usually tagged as the child developing "theory of mind".
Before theory of mind (TOM), there's no reason for a child to lie. They might say things that are false, but they can also see that what they're saying is false, and they don't expect others to believe the false thing they say. They find it funny to point at a blue thing and say "That's red!" The absurdity is hilarious at a certain developmental stage.
But the pre-TOM can't account for a bunch of behavior. As children model social situations, there are confusing things adults do that they just can't make sense of. I'm reminded of a story of a toddler getting very upset with her dad for "intentionally" refusing to give her a sandwich when she was hungry: she felt hungry, and he wasn't giving her the sandwich, which in her world meant that he was knowingly denying her what she wanted. The idea that she could know she's hungry while he doesn't wasn't even conceivable to her.
When TOM appears as a hypothesis for explaining this confusing data, it suggests a kind of crucial experiment: can the child get away with a fib? In the child's pre-TOM world, saying something false is absurd but cannot actually mislead others because those others have access to the same all-pervasive knowing about the world that the child does. But TOM suggests that the child can know something that someone else doesn't, and can mislead that other person about whatever it is, and they'll then act as though the false thing is true. So fibbing becomes a kind of crucial experiment for the child to understand how subjectivities work.
I think the thing with aqua regia is a dramatically different thing than a crucial experiment. It's more like something striking to explain. We have TONS of examples of those in subjectivity — more than I could hope to list, but I can gesture at a bare few:
• psychedelics creating altered states
• profound spiritual experiences
• reports of near-death experiences together with the apparently permanent personality changes those can bring
• falling in love
• some religious ceremonies creating profound subjective experiences of the divine (while others fall flat like a bad middle school play)
• hypnosis
• narratives affecting physiological state seemingly directly ("You can do it! YOU CAN DO IT!!!")
• a news cycle taking over someone's sense of what's real regarding events they have no personal connection to
I could probably keep going for a long while. It feels a bit to me like the subjective equivalent of listing examples of chemical reactions. There's most definitely no lack of them, and some of them are extremely profound!
Curious what you think of Prometheus Rising in relation to this (if you've read it)? If not: I've only read the first 5 chapters so far, but the exercises he suggests at the end of chapters feel very aligned with what you describe here. Eg in the chapter about "what the thinker thinks, the prover proves" he talks about deeply believing and visualizing that you will find a quarter on the street and seeing how long it takes to do so. Then varying the explanation for this phenomenon you believe in, seeing how that changes the time to get a second quarter etc.
The hostile telepaths thing is an interesting case for me actually. Reading your lw post when it came out has made a significant impact on my perception of this phenomenon, I believe. But it's pretty hard to put into words what exactly changed. I think it started with just noticing self deception in response to a hostile telepath after the fact. Moments where I would recover hidden information after the person left, combined with expectations that that information would be poorly received by said person. After a while I started flagging self-deception happening in the moment. Not that I would uncover the information itself, but a felt sense that it was happening based on a combination of signals. Part of it is a particular flavor or somatic tone, nervous system activation, feeling unsafe and inadequate with the timing suggesting it's related to something the other person did or said. That last bit obscures some of it: I think I experience a 'gnosis' that the change in my system related to a particular expression of the other person, but I'm not 100% sure how I get to that sense. Then there is the part where I can somewhat gauge if there is some information that feels dangerous by reflecting on the thing that just transpired. Often I will naturally start to wonder about how I feel about something, and that process is cut short with a strong sense of fear or danger. Something opaque happens here too, but in practice it feels obvious when I connect with dangerous information by the shape of my experience. It has a particular flow and vibe and tone to it. Sequence of events. Content. All factors into it I think, but I can't really give a formula for it. Another thing that seems to happen reliably is that my image of the other person changes when I'm supressing something. I feel more distant from them, they look less beautiful, more threatening, less nice. I start spontaneously questioning whether I enjoy being with them more.
I'm not thrilled with the vagueness of this description, but it's also not obvious how to improve it. I think very crisply describing exactly what happens is a dead end. Too much experience to keep track of, some of it probably not conscious even in theory.
You know, my parents were reading Prometheus Rising when I was little, but I never picked it up myself. I never actually knew what it was about. I just now looked it up. Now I'm pretty curious! Thanks for bringing it back to my attention.
I relate a lot to what you describe about hostile telepath type self-deception. I've come to explicitly tag when I notice an antimemetic effect, like my thinking becomes extremely thick and I start having a lot of trouble finding clarity and I even lose track of what it was I was trying to say or what the problem was. Thanks for sharing your experience; I find anecdotal stuff like this super interesting and relevant!
Another book that comes to mind is Assessing Adult Attachment by Crittenden. I'm not sure where to place it in this discussion, because it starts from a mostly objective way of working: the therapist or AAI scorer trying to outline a person's attachment strategies as a third party. I don't think they go as far as calling such a scoring objective, but the aim is to have consistency between professionals scoring the same data. On the other hand, a lot of it goes into handling reflexivity more rigorously. Both interpersonal and intrapersonal in the subject. For example, there are passages where they discuss looking for specific (emotional) responses in the assesser, eg when you feel a tendency to help or defend or side with the client which can be indicative of a helplessness strategy being active. Intra personal you have things like watching for micro expressions in speech that indicate distortion of information. Or somatic symptoms (eg cough) that pop up repeatedly in key moments that make talking harder/impossible, a indication that intolerable information might being accessed iirc.
It's actually quite intriguing how subtle the signs get. Even the structure/order in which memories are related can be meaningful. I am recalling my attachment therapist (who worked with Crittenden a lot) telling me that my detailed strictly chronological recounting of a traumatic event is common in people who are in a particular stage of processing that trauma. She couldn't give a deeper explanation than, it's somehow important for you to hold on to the memory like this, but her experience was that it will often change as integration progresses.
Overall things like which information is available, how it is accessed, which types of details are recounted, the types of memory (semantic, episodic etc), whether affect memory is present, which points of view are included/excluded etc are all seen as highly relevant to the underlying processes in the subject that are invisible on the surface. The book is more explicit and rigorous in this than I have been able to absorb, and from what I've heard (from earlier mentioned therapist) most of it is in Crittenden's head. She's only been able to codify parts of it.
I get the sense that there is quite a rich sense of non obvious parameters that contribute to psychological processes related to attachment, but at the same time that the thing is so complex that a model that's both complete and rigorously verifiably correct is quite far away.
> This sounds simple, but it was revolutionary. Alchemists certainly did experiments, but they often interpreted results through pre-existing frameworks... When an experiment didn't work, they might blame [..]. Early chemists began to treat unexpected results as potentially valuable information about how nature actually works, rather than as failures to achieve a predetermined goal.
It feels there is something very deep behind this mindshift. It increases the learning signal density per observation. And the "alchemist"/"scientist" mind stance differentiation seems widespread.
Encountering a bug in your program, answering a question from the textbook incorrectly, or failing to apply self-help advice. Curiosity in learning from negative results brings you much closer to almost any goal in the above endeavors (or "dissolving" a goal). However, the human default seems to ignore large or whole parts of the negative experience.
And this scientific mind stance seems really hard to notice and switch to, if you are in a "alchemist" mode!
1. There is a reflex of feeling bad that covers nodes linked to negative outcomes with 'ugh fields'.
2. And there is resistance towards switching to enjoying leaning towards contact with negative: "but I care about [working program/answering textbook questions right/being a good person/getting gold] - why should I seek leaning towards something that is the OPPOSITE? You want me to stop caring about my GOAL? No, thank you very much!"
Maybe this shift truly happens from some larger viewpoint or worldview?
Yeah, I quite agree with you. I'd love to understand this whole phenomenon better. It's obviously very sensible to focus on solving a problem you have, but at the same time there's something about being "solve this problem" oriented that can actually hinder understanding… which ironically can make some problems harder to solve!
I'm reminded of Joe Hudson's VIEW framework, specifically Wonder (the "W" in "VIEW"). It's remarkable how well it works. And given how remarkably well it works, it's just as remarkable that it's something people have to TRY to do!
I bet this puzzle is related to why the scientific revolution took so long to arise in the first place. People have been curious about things for at least tens of thousands of years. People have been very, very smart for at least that long too — including at modern levels of intelligence! We spread out of Africa something like 50,000 years ago. We started building civilizations about 12,000 years ago. But somehow it wasn't until about 400 years ago that we thought to systematize this style of investigation.
So somehow this mindset shift is a kind of slippery. But apparently not impossibly slippery! We really have nailed down a LOT with objective science.
Which is a big part of why I have hope that we can maybe actually expand the mindset, and the skills involved with cultivating it. It would be so, so good to have mainstream culture based less on forcing certain conclusions and more on being really deeply curious about what the systems we're all embedded in really are and how we might make them good.
In short: I like your points and your question. I have thoughts but no definitive answers — appropriately! :-D
What struck me here is your analogy to chemistry’s transition from alchemy. The point wasn’t that early chemists had a better theory than alchemists — it was that they had a better practice: precise observation, reproducibility, curiosity about failures as well as successes. That really resonates with your suggestion that subjective science has to start from reproducible “state shifts.”
Where I get stuck (and maybe you’re circling this too) is measurement. Chemistry advanced once people could weigh and measure reliably. For subjective science, what’s the equivalent of a scale or a beaker? Is it language, metaphor, physiological readouts, intersubjective agreement, or some yet-to-be-invented practice?
I’d love to see you expand on what a “crucial experiment” might look like here. What’s the equivalent of aqua regia dissolving gold — the phenomenon so striking that it forces everyone to take note, even without a theory?
I agree, subjective measurement is tricky. I could have articulated the problem a bit more clearly in this post. Measurement tends to focus on objectivity, but that's exactly what we have to go beyond when doing a science of subjectivity.
As to how to do that? Well, I have a couple of thoughts.
One is that there's nothing magical about measurement for its own sake. It's a solution to a problem: how do we tell exactly what's going on (e.g. in chemical reactions)? Objective measures were a good solution to that problem in objective science. There's a similar problem in subjective science. If we attend to what solves the problem (rather than asking how to make subjective science valid by force-fitting it to the FORMAT of objective science), then we're way more likely to develop an appropriate answer.
Another is that we already have subjective measures so to speak. This is a bit like saying we're born with objective measures: we have eyes, and hands, and a sense of how heavy or light things are, etc. We built on those to refine objective measures. We have the same thing for subjectivity: how do you know when you're happy, or in pain, or feeling impacted by a story you've read? That's the input we need to refine — which means figuring out WHICH PARTS to refine! (This was totally an issue in objective science: why aren't things like hue or texture considered as important to measure as volume and mass? That arose from refining THEORY as well as measurement precision!)
In particular, I really don't think it makes sense to START by pushing subjective measures into objectivity. Intersubjective agreement is a way of making what people are saying independent of an individual's first-person impressions. But an individual's first-person impressions is precisely what we're studying! Likewise things like physiological readouts. It's extremely understandable that people want to suggest these things as ways of doing precise measurement in subjective science. But I think it's a lot like the story of the guy looking for his car keys where the light is good instead of where he dropped them.
My guess is that there are METHODS of refining subjective precision, and there's some objectivity to which methods work well. Things like concentration meditation seem to help add immense detail for instance. There are also some framing tricks like learning to view cognition in third person. Once multiple people develop their skill with those methods, the can compare notes about what they're observing in their respective subjectivities. That might start to make some general (objective) trends about how subjectivities work more apparent. At that point we'd do a blend of subjective perception and objective measurement to really powerfully refine our models (and individuals' ability to explore their own subjectivities).
A bunch of spiritual traditions basically claim to have done this. Buddhism often makes this claim, kind of saying that that's what the Buddha did to attain enlightenment. But even if so, the scientific approach didn't survive. And I think that approach actually matters more than the specific conclusions.
> I’d love to see you expand on what a “crucial experiment” might look like here. What’s the equivalent of aqua regia dissolving gold — the phenomenon so striking that it forces everyone to take note, even without a theory?
We actually have a bunch of vivid examples of both of these.
There's a pivot point in children's development right around age four when they realize that knowing is a property of PEOPLE, not of the world. And that different people can know different things, including believing some false things. In the psych literature it's usually tagged as the child developing "theory of mind".
Before theory of mind (TOM), there's no reason for a child to lie. They might say things that are false, but they can also see that what they're saying is false, and they don't expect others to believe the false thing they say. They find it funny to point at a blue thing and say "That's red!" The absurdity is hilarious at a certain developmental stage.
But the pre-TOM can't account for a bunch of behavior. As children model social situations, there are confusing things adults do that they just can't make sense of. I'm reminded of a story of a toddler getting very upset with her dad for "intentionally" refusing to give her a sandwich when she was hungry: she felt hungry, and he wasn't giving her the sandwich, which in her world meant that he was knowingly denying her what she wanted. The idea that she could know she's hungry while he doesn't wasn't even conceivable to her.
When TOM appears as a hypothesis for explaining this confusing data, it suggests a kind of crucial experiment: can the child get away with a fib? In the child's pre-TOM world, saying something false is absurd but cannot actually mislead others because those others have access to the same all-pervasive knowing about the world that the child does. But TOM suggests that the child can know something that someone else doesn't, and can mislead that other person about whatever it is, and they'll then act as though the false thing is true. So fibbing becomes a kind of crucial experiment for the child to understand how subjectivities work.
I think the thing with aqua regia is a dramatically different thing than a crucial experiment. It's more like something striking to explain. We have TONS of examples of those in subjectivity — more than I could hope to list, but I can gesture at a bare few:
• psychedelics creating altered states
• profound spiritual experiences
• reports of near-death experiences together with the apparently permanent personality changes those can bring
• falling in love
• some religious ceremonies creating profound subjective experiences of the divine (while others fall flat like a bad middle school play)
• hypnosis
• narratives affecting physiological state seemingly directly ("You can do it! YOU CAN DO IT!!!")
• a news cycle taking over someone's sense of what's real regarding events they have no personal connection to
I could probably keep going for a long while. It feels a bit to me like the subjective equivalent of listing examples of chemical reactions. There's most definitely no lack of them, and some of them are extremely profound!