Are you the rider or the elephant?
Some recent threads seem to me to be pointing at a really fundamental tension that I don’t know how to articulate in full. But here’s a chunk of it:
When you think of your mind as divided into your System 1 / fast / unconscious / nonverbal / intuitive / emotional processes, which we’ll call the “elephant,” and your System 2 / slow / conscious / verbal / deliberate / analytic processes, which we’ll call the “rider,” which of those do you identify with? That is, when you say “I,” does “I” refer to the elephant or to the rider (or both, or neither, or other)?
We used to talk about straw Vulcan rationality, the kind of rationality that is only about the rider and completely neglects the elephant, and how the kind of rationality that, say, CFAR is interested in is much more about getting the rider and the elephant to communicate with each other and work well together as a team, complementing each other’s strengths and weaknesses.
We say this, but it still seems to me that many people I run into (for example, at introductory CFAR workshops) implicitly identify as their riders and treat their elephants as annoying pets that have to be managed so that they, meaning their riders, can get on with their lives. I think this is… “wrong” would be a type error, and also unkind. But I’m sad about it.
I’ll out myself: I identify mostly with my elephant, and think of my rider as at best a helpful advisor for my elephant.
And I get the sense that all of the disagreements I’ve been navigating recently have been with people who identify as their riders and are deeply suspicious of their elephants, and also (by association) of my elephant; and that this has been the main driving force behind the disagreement.
I don’t really know what to do with this. It seems like the polite thing to do is to mostly only engage people like this rider-to-rider (which is mostly what I’m doing right now, in this post), but there are a lot of important things—even important things for the art of rationality, from my perspective—that I think (and feel!) can only be communicated elephant-to-elephant, and setting aside how feasible it is to do this online, it seems to be unsettled whether people even want elephant-to-elephant communication happening on LW at all.
I’m also just worried about people treating their elephants poorly.
I have a lot more to say about this but I mostly want to open the floor up for discussion.
- Iterated Trust Kickstarters by (20 Apr 2021 3:18 UTC; 74 points)
- Two types of mathematician by (23 Feb 2018 19:26 UTC; 64 points)
- Policy-Based vs Willpower-Based Intentions by (28 Feb 2019 5:17 UTC; 56 points)
- Book review: My Hidden Chimp by (4 Mar 2019 9:55 UTC; 28 points)
- 's comment on Strategies for Personal Growth by (28 Jul 2018 18:52 UTC; 23 points)
- 's comment on Iterated Trust Kickstarters by (20 Apr 2021 5:00 UTC; 13 points)
- 's comment on Pain, fear, sex, and higher order preferences by (22 Feb 2018 12:51 UTC; 9 points)
- 's comment on Yoda Timers 2 by (22 Feb 2018 5:13 UTC; 7 points)
- 's comment on Realism about rationality by (17 Sep 2018 3:56 UTC; 7 points)
- 's comment on Highlights from “Integral Spirituality” by (16 Apr 2019 17:28 UTC; 4 points)
I think the elephant/rider analogy gets the mental imagery backwards, and advocate thinking in terms of monkey(s1)/deliberator(s2) instead. It’s system 1 that provides the motive force and actually does things; system 2 is a procedure implemented by system 1 (if system 1 feels like it).
The discussion of identity feels off to me. The monkey identifies as the monkey (so much as it identifies with anything), the deliberator identifies as the deliberator (so much as it identifies with anything). You seem to be talking about some third thing that gets to choose which of the two it wants to identify as; I’m skeptical. I expect the actual difference between you and other rationalists isn’t mostly about identity. (At a minimum, to make sense of it in my model I need to round it to something other than identification.)
When making decisions, by default the deliberator chooses what it wants and the monkey chooses what it wants, and so we have two optimization processes working slightly at cross-purposes. I think this gives rise to hypocrisy (see this post, Robin’s book) that was evolutionarily adaptive, but that today we can get a Pareto improvement by adopting a better compromise between the two.
I don’t think elephant-to-elephant communication is at odds with other people’s riders identifying with their riders. I think people are more skeptical when the claim is that the rider can’t even understand the nature of what is to be communicated, since they have the (I believe correct) view that the rider is a universal understander-of-things in some strong sense.
I’m not talking about a third thing, although I agree that something is off about my framing and I didn’t quite ask the question I meant to ask. In your framing, I’m talking about something like recognizing that the monkey is the one actually doing things, and using the word “I” to refer to the monkey accordingly.
Can you at least consider the weaker hypothesis that there are things some people know how to communicate elephant-to-elephant but don’t know how to explain to anyone’s riders? (In the same way that for most of human history nobody understood the mechanics of color vision, but everyone could show each other red objects.)
I do think that understanding the mechanics of color vision is not necessary to explain red to the rider. The rider is totally capable of undertanding things like “these things are distinguishable by an attribute that you are not directly aware of, but that is a lot like the difference between green and blue objects to you, and similar to how an eagle is able to see much farther than you can, and similar to how a dog can smell things you cannot” and many similar sentences. I do not think the concept of red has become qualitatively easier to describe with the onset of modern neuroscience (though it has definitely gotten quantitatively easier).
Sorry, I wasn’t clear, the analogy I have in mind for color vision is trying to explain red to someone who lives in a black-and-white-world and doesn’t have any experience of color at all.
I don’t know how much it matters, but I think you’re generalizing from fictional evidence here, in the following sense: If someone truly had no experience of colour, I do not expect that showing them red objects would likely give them much idea of what the experience of seeing red things is like for people who have lived with colour vision all their lives. (Compare those experiments in which cats were raised with no horizontal lines in their environment and grew up insensitive to horizontal features.)
In all likelihood, when I’m talking about ‘identification’ I do not actually exactly mean the elephant vs rider concept.
Instead of a S1 vs S2 division, I’m probably actually talking about a S1-S2 vs S1-S2 division (each side has both). But the associations / cultural understandings are clearer when I call it an S1 vs S2 division.
‘S1’ is the thing that keeps scrolling Facebook even as ‘S2’ is having verbal thoughts like ‘maybe I should stop’, but calling that S2 is not entirely accurate. Actually, something in S1 is causing me to have the thought ‘maybe I should stop’ which is also tied to some more subconscious emotion.
Everything I actually do is tied to S1/elephant in some way.
But for some reason, I have divided the elephant into parts I identify with and other parts I don’t.
I cannot tell if I’ve actually addressed your points or not because I’m having trouble with terms. I felt confused by your second paragraph.
Yeah, maybe a better frame for the thing I want to talk about is disidentifying with parts of yourself that want things that are ego-dystonic.
don’t like it much, seems to divide it between ‘endorsed’ vs ‘unendorsed’ ?
It’s a little different. Endorsing is a conscious activity, but finding something ego-dystonic isn’t.
If consciousness (S2) is really just the forum for disputes between S1 submodules vying for access to the motor neurons, then you’re mostly only going to become conscious of S1 activity when those modules are in conflict. You could even say S2 is “awareness and attempted arbitration of conflicts between S1 subminds” and so it makes sense that the floating point of view in your head would identify with whatever motor program it has chosen as the correct one, even while the body refuses to listen.
Despite originally liking the metaphor, I think the elephant/rider dichotomy has been showing cracks recently. I think a better metaphor for the mind would be something like a large unruly troop of clever monkeys with one bigger, more talkative monkey who is loosely in charge. Call it King Louie and the Apes.
First, it’s increasingly clear that System 1 is best understood as a parallel cluster of neural submodules. A swarm of a congress or a crowd feels much more accurate than … a large slow animal.
Next, I think one loquacious monkey trying to convince fifty other monkeys to steal fire feels much more like the internal experience of trying to decide to do something that you’re ambivalent about.
There’s still the issue that you tend to think that you are King Louie and the monkeys are your wild but potentially useful minions. No, you are the whole troop, Louie plus monkeys, and if you ignore this fact, be amazed at how quickly the monkeys that you’re not paying attention to manage to ransack your ambitious jungle schemes the minute you turn your back on them.
I really do think that the elephant/rider fast/slow System 1/System 2 “dichotomy” is just neither useful nor correct. Or, calling it a dichotomy leads to unhelpful assumptions. System 2 relies on System 1 processes for every microsecond of its activity—in a sense, System 1 provides thoughts, concepts, plans and assessments and System 2 buffers them and coordinates them in a single conscious workspace. By what algorithm does System 2 do this? By ”unconscious, implicit, fast“ algorithms and heuristics provided by System 1. Where else would they come from?
And, since it’s not a post from me if it doesn’t explicitly mention meditation, the complete lack of anything that’s actually “in charge” becomes apparent after relatively little time spent watching your own thoughts objectively.
Yeah, one way I would say it is that System 2 plausibly cleaves reality at the joints (it might literally just be the dorsolateral prefrontal cortex, at least according to Critch) but System 1 doesn’t, it’s a whole bunch of stuff we should carefully disentangle. All models are wrong, etc. Nevertheless I still think I am pointing to a meaningfully distinct cluster of people when I talk about people who identify as their System 2s.
Elephant-free decision making and rider-free communication both sound like hippie dreams to me.
I’m annoyed about the way in which you continue to state all of your positions on these topics so dismissively. I’m happy to wrestle with disagreement, but dismissive disagreement seems to me to be just straight up a bad communication / epistemic norm; I think you are not attempting to inhabit the world in which you’re wrong.
I just tried to pack too much into it. Guess it didn’t work too well.
First of all, you’re referring to Luke’s idea of elephant, but I prefer Robin’s idea of elephant which rings more true to me. In Robin’s idea, the conscious mind isn’t like a person with distinct goals riding atop the elephant. It’s more of a mask that the elephant wears, or a public relations department.
The elephant is responsible for decision-making. If you try to bypass it and put the conscious mind in control, you will likely hurt yourself or deceive yourself. But the idea, though wrong-headed, is alluring and comes up again and again. It’s a “hippie dream”.
The conscious mind is responsible for communication. Everyone is using it this way, and they are right—it’s nature’s way. Trying to bypass it in favor of “elephant-to-elephant communication” is another hippie dream. We should be thankful that it doesn’t happen more often.
PR and sales departments are crucial. We are rightly wary of anyone trying to bypass them and talk to the company directly. If you really want to offer a mutually profitable trade, use the approved channels.
Thanks, I’m much happier with this comment.
I’m happy with this, although I don’t understand the use of the term “hippie dream” to describe it.
I strongly disagree with this. I have seen people engage in elephant-to-elephant communication, and I’ve done a little of it myself, and I think it’s important, although I also want to respect other people’s wishes to not do it. What’s your crux here?
I’m happy with high bandwidth communication when it’s approved by the PR departments of all involved (like exchanging smiles, or teaching someone to play an instrument, or whatever). What I’m not okay with is suppressing your PR department or asking others to do the same. People often claim improvement from such practices but I don’t notice them improving. If anything, they get bad habits, like asking “what do you feel?” instead of reading people. Not to mention that the first effect of lowering criticism is often lowering criticism toward the practice itself.
Compare with something like debating, philosophy, or LW rationality - practices that pit your verbal skills against resistance, making you stronger and more fun to interact with. It’s like the difference between aikido (a martial art where you train only on people who don’t resist) and BJJ (a martial art where every workout involves fighting).
So, I definitely think there’s an uncanny valley of bad authentic relating, analogous to the valley of bad rationality. It sounds like you’ve mostly interacted with people who haven’t gotten out of it; I claim I’ve interacted with people who have, and that I’m mostly out of it myself, and that it has made me stronger and more fun to interact with. (And better at reading people.)
Many people desperately want their PR departments suppressed; that’s why they drink, or do drugs, or go to live concerts, etc. It’s a valuable service you can offer some people to do this for them without using substances.
Not saying you’re wrong, but one reason I’m skeptical about “practices” is that they often seem to make people overvalue them. Have you seen Richard_Kennaway’s comment on the circling thread which compares talking with NVC folks to talking with chatbots? Also see his comment on the kensho thread describing how people’s claimed benefits from meditation are illusory.
To your last point, do you think people drink at parties to suppress their PR departments and get the ugly truth out? To me it looks more like they want a credible excuse to send certain PR messages that aren’t necessarily true, like “I’m fun and free”. That’s also part of nature’s way and I’m happy with it.
My current sense is that Qiaochu has indeed gotten more fun to interact with, and I know of at least two other cases in which I think authentic relating and related practices have helped someone be a lot more fun to be around (from my perspective). I also know of at least two cases where I experience interacting with someone as much worse than I used to, probably as a result of authentic relating. So the overall effect might just be noise, but I do think the effect size is large enough to make that somewhat unlikely.
Thank you! This doesn’t change my mind outright (seen too many examples to the contrary) but it’s a good data point and I’ll remember it.
Went digging, and found it here:
https://www.lesserwrong.com/posts/aFyWFwGWBsP5DZbHF/circling#cgtM3SRHyFwbzBa56
I disagree with the premise of this post.
But I don’t.
So perhaps the better question would be: “Do you think of your mind as divided into …?” (And for me, the answer is “No.”)
When I see people who both do think of their minds in this way, and then also choose to identify with the latter “part” / “aspect” / etc., I am indeed “deeply suspicious”—largely because that seems to me to be a) inventing a distinction that doesn’t exist (or, slightly more accurately, taking an interesting fact about how our minds work and reifying it into some sort of strongly-believed ontology), apparently for the purpose of b) having very strange ideas about oneself. (Needless to say, I look askance at both (a) and (b), separately and especially together.)
So, by way of accepting the invitation for discussion, let me ask:
Why?
Just as importantly, what on earth does this mean in not-metaphorical language?
This is a complicated question for me to answer. It doesn’t really feel like a choice, to me.
One aspect is what Unreal said: many people have conversations with themselves like “I want to do work and I don’t want to watch TV, but oh no, I somehow mysteriously find myself watching TV instead of doing work” and when this happens to me I identify as the part of me that wants to watch TV (or rather, the part of me that wants something, which it is trying to get by watching TV), and regard the part of me that wants to do work with suspicion, because mostly those are Moloch’s preferences, not mine. So, for example, I generally don’t feel guilty about indulging the desire to watch TV.
There are other aspects but I don’t think I can explain them well. You’re asking questions that get to the core of my being in some sense and that’s just not a short conversation.
Relevant SMBC comic.
I can’t tell if you’re intending this as a counterargument or not, but to the extent that you are, this is pica. Listening to yourself is a skill that needs to be trained.
If you’ve been abusing a small child and then one day finally start listening to what it wants, it might say all sorts of crazy shit, but the response to that is not to continue never listening to children.
I was mostly intending it as something funny in a thread-relevant way. I agree that wanting to (in some sense of “wanting”) sit and watch TV all the time might be a pica-like symptom of some more interesting need, and therefore that “the things we feel urged or compelled to do are often obviously a really bad idea” is not a good argument against listening to one’s feelings of need/compulsion. But I feel I should draw attention to the fact that you really did say “for example, I generally don’t feel guilty about indulging the desire to watch TV” rather than, e.g., ”… about finding something to do that’s more satisfying than working”.
(Feeling guilty is probably counterproductive whether one identifies as the wanna-watch-TV agent or the wanna-get-work-done agent or both or neither; I take it the point of your comment wasn’t really about guilt as such.)
Identifying more as “the rider” is often the default in humans, by nature of the fact “the rider” is the part of you that’s conscious. (I think most people identify as both to some degree.) I agree the distinction is kind of weird, but this is what I’ve noticed, in myself.
I lived my life assuming that “me” was the thinking part, the conscious part. So I’d say things like, “I want to go to bed … argh, why am I not going to bed. I should really go to bed.”
Well, “I want to go to bed” was not including the elephant, who was not wanting to go to bed enough that my body didn’t move.
This is already a strong ontology that’s implicit.
Now my ontology is something that seems to bear more weight, at least on my own behaviors, feelings, thoughts, etc. I move through the world better when I include both elephant and rider and recognize the rider is valuable and helpful but plays a smaller role. I don’t convince myself to go to bed by saying “I should go to bed.” I do something else entirely.
I think most people we run into these days identify as the rider but that this isn’t why, and also I suspect that this wasn’t the norm historically. My current suspicion is that identifying as the rider is a learned behavior reinforced by parents, school, work, etc. and that feral humans wouldn’t do it.
What something-else-entirely do you do?
The short answer is Internal Double Crux (a crux-seeking dialogue that can happen verbally or nonverbally between parts in apparent conflict). A similar technique is the Internal Family Systems (IFS) therapy system. Another similar technique is Inner Relationship Focusing (IRF).
There doesn’t seem to be much information about “Internal Double Crux” on the internet—there are various blog posts and things that mention it but nothing that makes it clear how it works. My initial, doubtless point-missing, reaction to the idea is: my Rider knows how to engage in dialogue but my Elephant by definition doesn’t, so surely any “crux-seeking dialogue” I try to engage in is really going to be between different parts/aspects/roles of the Rider; so how can it actually help with cases like the “damn it, I should be going to bed, why aren’t I?” one?
My guess is that the answer is some combination of “if you do it right, you can get the Elephant to pay attention and notice that its interests are being represented adequately by the Rider” and “if you do it right, you can bring the Elephant into the dialogue by involving some appropriate kind of nonverbal internal communication” but it’s far from obvious to me how either of those would be cashed out in practice.
Would you care to say more about the technique?
Ah, so, the dialogue is not actually happening between rider and elephant. (Read my comment response to paul for that explanation.)
Another metaphor that might be more helpful is a pile of monkeys (elephant, S1) and a voicebox (rider, S2).
In IDC, two monkeys are having a disagreement, and I (meta-me) pass the voicebox to each as they talk. Sometimes, the monkey doesn’t want to talk or doesn’t know how to talk, even with the voicebox. In which case, I can use Focusing (extraction or translation of nonverbal content into more consciously understood content / words). The two monkeys relay content. The meta-me directs the conversation and maintains the structure of it, and I will also be trying to gently head for cruxes.
Metaphorically, the meta-me is a third monkey but with eyes (another rider feature) instead of a voicebox.
In more “advanced” IDC, the monkeys can communicate without the voicebox at all. All their communication can theoretically be kept nonverbal. I know less about this and can’t speak to it as well.
Interesting; thanks. I have the same sort of initial reaction to this as before: I don’t know how I would go about “passing the voicebox” to my usually-voiceless internal monkeys, and suspect that in fact there is no connection between them and the voicebox (which is why they are usually voiceless), and more to the point it’s not clear to me how to relay what either monkey has to say to the other monkey. Still, if the technique has been found to work then presumably either my initial reaction is wrong (a shocking concept, to be sure) or else there’s some other way for the dialogue to be helpful (e.g., maybe it turns out that the rider / King Louie is more important than the usual metaphors suggest, and negotiating between different bits/aspects of that is useful sometimes).
Improv comedy often tries to give monkeys the voicebox more directly. If you’ve tried that ever.
This is what Focusing is for.
If that’s the sort of thing Unreal had in mind, then I think I misunderstood the metaphor; to me focusing (as I understand it) isn’t a thing I’d call “passing the voicebox”, but something more like “inferring what they would want to say if there were a way to give them the voicebox”. And, as I said, while I have no trouble with the idea that it’s possible to infer a fair bit about what the monkeys / elephant / System 1 pieces want (note: I am not trying to imply that all of those are equivalent descriptions; just acknowledging that reality probably doesn’t exactly match any of our metaphors), from what I’ve heard so far I’m not seeing how the actual monkeys / elephant / sys1 bits, as opposed to models of them, get involved directly in the dialogue; the technique sounds a bit like holding a mock debate between people trying to represent Israel and Palestine and expecting it to reduce tensions in the Middle East.
(Unreal’s own suggestion of improv comedy seems more likely to give monkeys the voicebox, in the sense of finding ways for their wants, preferences, beliefs, etc., to find a path to influencing what one does. But how you incorporate that into anything that could be called a dialogue and allow for relatively sophisticated logical reasoning like double-cruxing, I still have no inkling.)
To be clear, I’m not saying “this technique probably doesn’t really work”. I’m saying that nothing I’ve heard about it so far enables me to understand how it could, which I suspect means that nothing I’ve heard about it so far is giving me enough description of what it actually is to make sense of it.
I’m confused by your understanding of Focusing. How familiar are you with it in practice?
Or I’m confused about what you’re calling a ‘model of a monkey’.
I can check with the monkey whether my felt sense label is correct. That’s the whole deal of Focusing. The more correct my label is, the more I get a feeling of ‘ahh yes that’ vs ‘ehhh’.
And when I get better at it, I can reach the point where my monkey offers its own labels. Which feels different from when my rider offers labels.
AND, furthermore, I can actually pass my monkey the voicebox. It’s called ‘blending’ in IFS and is actually a common occurrence in humans. When you’re really, really mad, you are blended with a monkey and are speaking from it.
Or when I’m really, really scared, my body will start shaking, and I’ll start repeating the same phrase over and over, “I don’t want to die.” This is also an example of the monkey having the voicebox.
The improv thing is another version, but you’re not necessarily blended with an emotional part.
Double cruxing is hardly sophisticated logical reasoning. It can be more like, “What are you afraid of?” “No rest.” “OK, and what are you afraid of?” “Future pain.” “What is the other one missing?” [sends a felt sense that is what ‘no rest’ feels like accompanied by an inner simulation] The image depicts a specific scenario. Other side [passes a different image that contains a potential alternative] Etc. Etc.
I am not familiar with focusing in practice, and it would be extremely unsurprising if my understanding of it were wrong, and extremely surprising if my understanding of it were not incomplete.
When I called double-cruxing “relatively sophisticated”, the word “relatively” was there precisely because of course it isn’t sophisticated by comparison with, say, proving difficult mathematical theorems—but it seems like it does involve explicit reasoning of the sort that Elephants are not generally supposed to be good at. What you describe seems (though of course this may just be misunderstanding on my part) to be missing something that’s an essential part of double cruxing as distinguished from other forms of dialogue, namely the search for something that if wrong would change your position on the original issue. Are your monkeys sophisticated enough to identify what things have that property?
So does IDC depend on having achieved a certain degree of skill in focusing and IFS? Or does it have its own way of giving voice to (and passing information to) the relevant internal subsystems?
Anyway, let’s return to the original elephant/rider problem we were discussing, which when it happens to me presents to me in these terms: “I” (meaning, so far as is immediately apparent to me, all the bits of me that are consciously present and capable of language; that is, roughly, my Rider) want to go to bed and get some damn sleep for a change, but despite my (apparently, superficially) forming the intention to stand up, turn off the computer, and go to bed, this fails to happen because some other bits of me (roughly, so it would seem, my Elephant) have other preferences. That seems to match well with how you described it.
The IFS model seems to be somewhat different from the rider/elephant model, with a bunch of different subselves that are (in some contexts at least) capable of speech and reasoning and so forth, which seems to make them non-Elephantine. But maybe I’m misunderstanding, and the idea is that they are parts of the Elephant that can, in the right circumstances, steer the rider around and influence its speech and reasoning and whatnot?
A description of what you actually do in the situation you describe where “you” want to go to bed but it Just Doesn’t Happen would, I think, be both interesting and illuminating.
Inferring what the monkeys would want to say is not Focusing. The step of Focusing where you check for fit is directly checking with a monkey. I agree that it’s not at all clear how to explain how to do this to somebody who doesn’t know how.
As a person who gets the metaphor and previously felt as though my mind was split between two parts, maybe I can help address this.
The part people talk about as the rider is the part of your brain that sees itself and constructs memories. Neurologically this is probably centered on the loop between the hippocampus and the prefrontal cortex but I don’t know enough about neuroscience to say that with a lot of confidence, but the point is that it’s the thing that seems to exist between memory formation and the fancy, large-in-humans parts of your brain that deal with language and abstract reasoning. The rider is also the part of the system that gives you complex, multi-layered self experience that we often refer to as some combination of “consciousness”, “self-awareness”, and “qualia”.
The elephant part is basically everything else because it’s the parts of the brain that are not directly connected to this loop where conscious memory formation happens. These are the parts of your brain that do things without involving the rider loop prior to taking action, although after action the rider loop may become aware of the consequences of the elephant’s actions.
Of course this is a metaphor but the point of it is to highlight that there are parts of the brain that are tightly integrated with the memory loop and parts of the brain that are not. The parts inside can feel better seen than the parts outside, and this sets up the opportunity for feeling as though one is split.
Then saying you “identify with your elephant” means that when you turn to the question of “what makes me me?” you think of the parts outside the memory loop as more constituting what differentiates you than the parts inside the memory loop. The alternatives are to identify more with the rider or the parts inside the memory loop, identify with both equally (which may mean not much noticing the distinction), or to not identify anything as much uniquely you.
I’m also bewildered by this idea of knowingly and intentionally identifying with only one conceptual part of your mind in opposition to others. (Or maybe less dualistically, the part that’s connected to your mouth and fingers reporting that it identifies with some parts and not others—how do the repudiated parts feel about it?)
I’m pretty sure I know what OP means by this and I agree that it’s bad, but it also seems like something that generally happens by accident. Why would you deliberately cultivate this kind of disconnect in the opposite direction?
It doesn’t feel deliberate to me, and I think there are genuine asymmetries between the rider and the elephant; among other things, the elephant is much larger, and I think an elephant with no rider is still human and still has moral value (corollary: I think all of the EA work on understanding consciousness in order to understand which things have moral value is misguided, because to me consciousness is not a necessary condition for moral value), but a rider with no elephant is basically nothing.
I don’t claim that the way I’m set up is optimal; probably I have some growth to do in the direction of incorporating the rider into my self-concept. One reason I treat my rider with some suspicion is that it often speaks for Moloch, not for me.
All right. It sounds like I misunderstood some of your previous comments about this.
I think I understand what Qiaochu means except I identify with the rider most of the time. I don’t think it’s a question of knowingly choosing to identify with one part of your mind, versus just automatically feeling like that part is “Actually You.” To give a different example of the same thing: there’s a part of me that is still “high school internet troll” and although I understand his positions and feelings I find it hard to identify as that person at all.
I’d strongly reccomend reading a book called “The Master and his Emissary” by Iain McGilchrist. It won numerous science awards not only for it’s description of the two hemispheres of the brain but of their import and significance.
He has a quick TED talk that serves as a short intro as well.
I know all models are wrong, and some are useful. But in every day life, I find myself thinking of my brain in two parts that are roughly A) my beam of attention and B) my everything else. This doesn’t seem to be totally captured by the other divisions and metaphors mentioned. Both A and B feel like *me*.
My attention (A) feels a lot like a beam of light that can be narrow, wide, dim, or even off. My everything else seems to include my subconscious, the answers that pop into my head without effort, the habits I execute by rote, internal dialogue that chatters away even when I’m not particularly paying attention to it, desires, emotions, and much more.
B churns away whether A is on, off, paying attention to B, or looking in some other random direction. But overall, my decisions and quality of life seem to be better when I use A as much as I possibly can, and when I spend a big chunk of A’s time studying what B is doing.
[I updated this comment to add the rest of this.]
The behavior of scrolling through the internet late at night, and the thoughts that I should probably stop and go to bed, both feel to me like they’re in the same general category. The behavior and the inner dialogue both appear, bubbled up from the mysterious machinery of my brain (B) and presented to my awareness (A) on a platter. (I know awareness is also generated in the same machinery and that it isn’t ultimately an unsolvable mystery how the brain works. I’m only describing how my experience feels.)
Directing my attention seems to be the extent to which any meta level of control is introduced to the picture.
In the above example of scrolling when I have thoughts I should be sleeping, I ultimately decide to go to bed right away if I really focus my attention on how worn down my eyes and body feel and notice how muddy my thoughts are compared to when I’m well rested. For extra ammo, I can focus on how whatever craving is driving the scrolling doesn’t seem to be diminishing from moment to moment and has no hope of being satisfied.
I also identify more with the elephant, which I (probably unhelpfully) think of as the one that ‘actually does maths and physics’, in the sense of gaining insights into problems and building intuitive understanding.
I (also probably unhelpfully) think of the rider as a more of a sort of dull bean counter who verifies the steps in my reasoning are correct afterwards, and ruins my fun for some of my wilder flights of fancy.
I’m slowly learning to like the rider more—it’s doing more than I give it credit for.
Probably some of the issue is trying to fit everything into these two categories. I think Sarah Constantin has convinced me that there are at least three things in the world—flow states, formal step-by-step reasoning and insight. I’ve been unthinkingly lumping flow state in with insight as the good stuff, and leaving the rider with just formal verification. Someone else might lump insight differently.
I associate myself with the unconscious-self more and more (note: an unconscious-self is bigger than an elephant-self because some modules in a brain are deliberate & analytical, but not directly available to the verbal/conscious rider; I very much agree with @moridinamael’s comment above).
Conscious-self seems more like press secretary for more hard-working unconscious-self, who is in charge of most of the decision-making. But, ugh, everyone experienced how «conscious ruled unconscious» (≈ will-power). I think the role of conscious-self in «the use of willpower» is to communicate from long-term modules to short-term modules of unconscious-self.
«Inner Game of Tennis» contains some recommendations on how to augment communication between the modules. I also found TDT-mindset helpful to tell early-evolved modules what later-evolved modules think is worth doing.
For whatever it’s worth, I do not think this has been a crux for me in either of the last two recent discussions, though I wonder whether your model of me suggests that I might be modeling myself badly here.
Yeah, you’re not one of the people I have in mind.
A better image for the conscious/unconscious divide is one I heard from Iain McGhilchrist who describes a large stage that is in darkness, and a small spotlight. The spotlight can move about and bring a small portion into light and detail.
This image tells us that it’s all one thing. When we communicate with each other, we may feel like we are two little riders talking to each other, which on one level we are, but of course we are also transferring a lot of information unconsciously.
It’s easy to feel scared and ignore the fact that most of the stage is dark. But I think the correct way to view our riders is as a servant who can do some simple tasks, but should ultimately get out of the way—and never think that they are running the show.
Do you think this elephant to elephant communication is a similar thing to what Valentine was talking about in Kensho?
Val probably would’ve had an easier time explaining it elephant-to-elephant, although kensho itself is something different. Unfortunately it’s hard to do elephant-to-elephant communication through text except through poetry (in a broad sense), and many LWers are allergic to poetry.
It does seem like Eliezer used a good amount of poetry, though the poetry usually had a different feeling to it, than my experiences of continental philosophy, which is also mostly poetry. I am not sure what the difference between these two types of poetry is, but I have allergic reactions to one of them, and not the other (and I do think both of them are closed to elephant-elephant communication than normal explanations).
I also think fictional storytelling is pretty good at facilitating elephant-to-elephant communication, which is one of the reasons why Eliezer wrote HPMOR, since I think he found that some rationality concepts where hard to communicate abstractly (based on a conversation I had with Brienne about this a while ago). In general, Brienne strikes me as quite good at facilitating elephant-to-elephant communication via text, without triggering most of my poetry allergies.
I’ve kinda stopped understanding what you mean by E2E communication?
Poetry is verbal. Moreover it’s not spontaneous, a single line can have hours and days of conscious work put into it. Moreover it’s often sexual display. Moreover it doesn’t ask you to lower your criticism before reading, in fact it’s famous for working on people who have their shields raised. Moreover it often appeals to the conscious self, e.g. the Sermon on the Mount is amazing poetry but many people say their base instincts stop them from following it fully, and the same holds for LWish poetic rationality. In short, I can’t think of a more fitting task for the “PR department” than poetry.
I’d love to see more good poetry on LW. Successful arguments from beauty are always worthy of attention, because beauty, while not a perfect cue of truth, is still an unreasonably good cue.
Poetry, when done in the way I have in mind, is interpersonal Focusing, which I would describe as E2E via the medium of words (I notice I’m now worried we may be slicing up rider / elephant differently, maybe very differently). You have a felt sense, you write down poetry which is a good Focusing label for that felt sense, and if you’re lucky it will also be a good Focusing label for someone else’s felt sense.
I think we’re talking about the same elephant—the thing that makes you eat one more cookie. But the room is dark and we disagree on what the elephant looks like. I subscribe to Robin’s view of the elephant, it seems simple and consistent.
You, on the other hand, seem to identify the elephant with System 1. That sounds weird to me. For example, you can have a wordless intuition (coming from religion, etc) that something is sinful, but do it anyway. So it’s not a System 1 vs System 2 conflict, but Robin’s account still holds up.
Then you talk about E2E communication, and my confusion grows. People on the kensho and circling threads aren’t objecting to System 1 communication like smiles or dance. They are objecting to mind tweaking and “how do you feel”. The two are different, so let’s not use one to argue for the other.
Then you mention poetry and my confusion grows more. Appealing to people’s existing feelings is a part of poetry, but beauty seems like the more important part, and beauty can argue for new ideas as well. That’s how Eliezer argued for rationality. So saying that LWers are allergic to poetry seems to miss the mark. LWers are happy with beauty, but allergic to woo, and that’s a good thing.
I like this comment; I agree that I’m not using quite the right words and my concepts are slipping around a bit.
I’m trying to pinpoint where you think asking leading questions like “how do you feel” is different from smiles, dance, and poetry. They do seem different, but I’m not sure why.
Smiles and poetry appeal to the PR department. Asking “how do you feel” is a request to bypass the PR department.
Many of my comments in these threads (like the fish comment, or the one about hippie dreams) are trying to argue that no one is entitled to bypass anyone else’s PR department. You’ve got to go through proper channels. If you’re charming, then charm me.
Ah!
You aren’t in fact charmed (or overawed) by people who use feelings-heavy, mystical, or salesy talk — you instead hear it as an explicit/denotative request for you to be charmed, which you think is unjustified. Is that right?
Yes!
Okay, I think that’s a difference between us. I hear that kind of language not as saying something denotatively, but as more like “casting a spell” on the audience. It doesn’t throw up the “error: that doesn’t make sense/seem fair” response because I’m not expecting it to be communication in the first place.
Someone who wants me to relax, say, and is putting verbal and nonverbal optimization pressure into getting me to relax, is going to cause me to relax, just because I want to be compliant in general. For me, only a totally expressionless and artificially dry request would be free of the ”hypnotic” social pressure and would be interpreted as a mere request without the “hoodoo.” I think you probably have a less sensitive “hoodoo-detector” and so you read more things as communication rather than influence.
Ah, you’re totally right in that a lot of poetry is R2E, not E2E. I somehow forgot that facepalms
Similarity to kensho: it is hard to explain in words / is experiential. Don’t think it’s the same beyond that.
E2E obviously happens all the time in normal, in-person conversation? We point to how body language, facial expression, etc. subconsciously affects what’s communicated. Stereotypical flirting is going to be a lot of E2E with some verbal content.
Everyone E2Es; there seems to be a lot communicated below the level of consciousness. My guess is ‘feral humans’ (as Qiaochu calls them) are more in touch with it somehow? I don’t know very much here. I’ve seen E2E happen with no words at all, and it’s confusing to me still what that’s about.
A lot of responses seemed like elephants saying “WTF?”, so maybe not.
Not directly related to topic, but I was interested: does this identification of oneself with either rider or elephant, somehow correlates to MBTI personality types?