Author’s note I: This is another “theory of everything” essay, in the same vein as Lies, Damn Lies, and Fabricated Options or Social Dark Matter. The intent is to show how a handful of widely-spaced phenomena are all explained by the same underlying dynamic, such that you can start to reason about them as a class and more easily copy-paste intuitions and solutions from one domain to another.

Author's note II: These days, my thoughts go onto my substack by default, instead of onto LessWrong. Everything I write becomes free after a week or so, but it’s only paid subscriptions that make it possible for me to write. If you find a coffee’s worth of value in this or any of my other work, please consider signing up to support me; every bill I can pay with writing is a bill I don’t have to pay by doing other stuff instead. I also accept and greatly appreciate one-time donations of any size.

The very, very, very short version of this essay would be something like: Many equilibria are maintained via a combination of a force that presses, and a constraint that holds.

A cup stays stationary on a table thanks to gravity pulling it constantly downward and the table limiting how far down it can go. If you remove the constraint of the table, the cup only falls a few feet further and then gets stopped by the floor. But in many analogous situations, there is no second, nearby additional constraint, and removing the “table” means that the “cup” will end up traveling very far indeed.

I.

You’ve likely had the experience, at some point or another, of going to open up the back door of a car or an SUV only for a bunch of objects to come tumbling out.

(Maybe this wasn’t your car. Maybe you’re neat and tidy. 😛)

In their original configuration, the objects were more-or-less stable. They weren’t going anywhere, so long as the door was there to keep everything in place. The door was a control system—objects could be piled on top, or stuffed and wedged into cracks, because there was a container holding them in place. Remove some critical part of that container, and the internal pressure sends everything flying in the obvious way.

II.

Another control system that feels similar to me is the way that many organizations handle their finances. Abstracting away a lot of detail about norms and politeness and reading the room, the basic pattern in many places is:

• Employee: continue asking for more money, both for yourself and for your projects
• Employer: set the ceiling at which you stop giving out money

Companies often have a money guy (or a whole money department) whose job it is to track the inflow and outflow and make sure the numbers are adding up and all of the expenditures are proportionate and reasonable. In such a company, it’s not your job, as a mid-level employee just doing the object-level work, to track the impact of your requests on the overall budget. You’re supposed to advocate for what you and your team need, and trust the container hold you.

(“Always ask for more. They’ll tell you ‘no’ when they need to; don’t leave money on the table.”)

But not all companies work this way. I’ve worked in places where the explicit norm was “figure out what you need, and tell us.”

(This requires a lot of trust, but some places have that much trust.)

In this latter sort of organization, there’s no table to stop the cup from falling toward the floor. There’s nobody whose job it is to say “no,” and if you take someone who’s used to the former type of company and embed them in the latter type without sufficient onboarding and acculturation, what will happen is that they’ll ask for more

(and get it)

and ask for more (and get it)

and ask for more (and get it)

and ask for more (and then eventually someone will finally get around to being like, hey, what the heck is going on, here, and they’ll look into it and discover that their new hire was not, in fact, doing the cooperative high-context move of figuring out their own stopping point but instead has just been draining the company of resources and likely would have kept doing so, forever).

(And then they’ll be let go, for reasons of not being a good culture fit.)

III.

It’s actually fine to just let your elbows relax on the plane or the subway, and press up against the other person’s elbows, as long as they’re relaxing, too. If the other person is holding their elbows in and shying away from contact, the relaxer may misinterpret the lack of pushback as tacit permission and just … keep on expanding.

IV.

Pause for a moment, if you’re willing, and try to think of some other situations or systems that have an analogous thing going on. I’ll drop some more of my own answers below, separated by a few extra bits of text here, in case you’re the sort of person who wishes you could stop yourself from reading ahead and getting spoiled but you have a hard time actually stopping because your eyes just move so gosh-darn fast, buddy, this is your spoiler warning, stop now if you want to, there’s a limit to how much buffer text I’m willing to put here to save you for yourself and we’re pretty rapidly approaching it, okay, you’re on your own, hope you’re happy with your choices:

• One roommate just leaves their dishes in the sink, and it’s never a problem (in one sense) because the other roommate is less tolerant of mess and clutter and always cleans them up.
• A sort of “cash bucket” mentality, for things like a personal fun budget, where you can spend all you like because that specific pot of money is finite and at some point it runs out and then you just stop spending. (Or something like I don’t eat sweets, but when Grandma sends me cookies at Christmas I can just eat them until they’re all gone.)
• Couples falling into a dynamic where one of them provides a sort of stabilizing energy, such that the other ends up falling into the habit of being wild or explosive (because they trust the first to hold space for them).
• It’s a bit of an inversion/reflection of the dynamic, but: working under deadline pressure. Not needing to motivate yourself or prioritize but instead letting the immediate crisis spur you to action and then turning to the next-most-on-fire-thing after that. The time pressure’s job is to build and build until a given task pushes itself to the front; the person’s job is to sort of inertly sit there and be moved by only sufficiently urgent tasks.
• Apparently there exist certain BDSM or power-exchange relationships in which a “dom” or “top” will create a container inside of which a “brat” does not have to engage in constant self-control and can (at least sometimes) just … let loose, and be themselves, without any internal top-down censorship or curation or holding-oneself-in-a-shape, idk, sounds fake and extremely too good to be true, I don’t see how any human either a) could contain me with sufficient strength and skill that I could in fact trust them enough to genuinely relax or b) would actually want to do that for me, even if they were capable, but … email me I guess?

The crucial insight (according to me) is something like “once the container breaks, or the stopping force is removed, something bad will happen.” Sometimes the pushing force has its own brakes built in, or some other larger container exists outside of the first, but not always. Sometimes, the pushing force just drives to infinity once nothing is holding it back.

V.

Okay, so. “The cup falls, without the table (and this is usually bad).” Call that Puzzle Piece #1.

Puzzle Piece #2 is the fact that we are made up of a whole bunch of pushing forces.

Humans evolved in an environment, and that environment provided a bunch of constraints (tables). Evolution, in the process of “““designing””” us, didn’t do the equivalent of building cups that have their own method of propulsion and a set of positioning tools that would steer them toward some ideal position and then stay there.

It just … put the cup on the table. Evolution made a bunch of things that are equivalent, in the analogy, to gravity—forces that simply push and keep pushing.

For example: our immune systems can kick into a higher gear during times of acute threat, but they never really go offline. To a first approximation, your immune system is always on alert, always fighting—so much so that if the environment is too clean, it will cast around until it finds some random thing to fight (like peanuts, or grass, or your own cells).

(I’m blurring out some important details, here, but the thrust of the claim is true.)

In the ancestral environment, there was never any risk of running out of pathogens, so immune systems that came with a “relax, there’s no threat” setting didn’t have any reason to be selected-for.

(They’re also more complex to begin with, so it’s possible they never arose in the first place. And even if one were to discover some fairly simple genetic mechanism for going-a-certain-distance-and-then-stopping, for every such design there are bound to be many, many, many more just-as-simple designs that just keep going. Even in the world where the two types of designs are on equal footing re: complexity (and again, they aren’t), if the rubric is just “does it keep the cup pressed into place against the table?” then it’s more likely that evolution would have randomly stumbled across, and subsequently kept, one of the no-stopping-condition designs. The table is right there.)

VI.

I want to talk about a few more concrete examples, but at this point I think there’s enough concrete grounding to back out and talk about the bigger picture.

Namely: lots of our tables are disappearing.

(As we fundamentally alter the environment around us and unlock a bunch of new possibilities with technology that our bodies and brains did not evolve to take in stride.)

Lots of our tables are disappearing, and in the subsequent aftermath, we find ourselves with lots of gravity-like drives that are pushing us very far in various directions.

I think those drives often cause us to overshoot the optimal point, and sometimes send us into territory that’s really, really bad.

The obvious example here is obesity. In the ancestral environment, food was often pretty scarce. A lot of it was not particularly nutrient-dense. And it often took a great deal of active effort to go get it, whether that meant foraging for hours, hunting all day, or spending a lot of your time farming.

This meant that, in that environment, a basic and simple drive that was something like “just keep on eating, eat everything you can find, with a preference toward things that are delicious” … actually worked out okay? In that a person could press toward more food and this couldn’t actually lead to an unbounded problem because there just wasn’t that much food available.

(I’m simplifying a little. There is also, for example, the fallback control system of “what you can literally fit into your stomach before you become too full to eat any more.” But it’s clear that, in the modern environment, with modern levels of exertion and modern delicious, calorie-dense foods, that’s not enough to prevent many of us from eating our way into obesity. The table disappeared, and many of us are falling straight through the floor.)

VII.

Another sort of low-stakes example, which was in fact the seed that eventually became this essay, is cleanliness.

I recently posted the following in a back-and-forth on my Discord server:

I think there is in the general human population a mental illness that is similar to the one behind gymnophobia; I think that humans have gotten confused about what constitutes clean, and started aiming for something that is more like sterile, and this is actually quite bad for us both physiologically and psychologically.

Like, I think people are reinforcing, in one another and in society in general, an overcorrection on hygiene that is actually creating new problems.

I think the monkeys are teaching each other that a certain kind of too clean is good, and that they are just mistaken as a matter of plain fact.

(here I have said a thing very recklessly and quickly, and offhandedly, and not really explained what I mean or justified. feel free to double click.)

People did, in fact, double click, and I’m going to reprise most of what I said in my response:

Okay, so, there are like two #facts that I’m keeping firmly in mind as I think about things.

One is that Jewish folk back in the day had, like, meaningfully lower rates of illness and death, compared to other Europeans, because of their habits of ever actually washing themselves at all. So, I’m thinking “full-on filth is bad.”

Another is that we have (if I understand correctly) fairly strong evidence that immune systems go haywire if presented with too little to fight and that a lot of allergies and other autoimmune disorders are downstream of things being too clean. So, I'm thinking “full-on hospital-style cleanliness is bad (if you don’t have hospital-shaped problems that make it necessary).”

Then my mind goes to the Jews-in-Europe thing and I’m like, okay, but that wasn’t the natural/ancestral environment! Like, the other-Europeans getting sick and dying a bunch was downstream of concentrated populations and poor sanitation (even on the level of, like, villages rather than full-blown cities).

And then I pop back to, like, okay, what actually causes disease, and it’s parasites and germs and pollutants and stuff. So the idea is to maximize something like ... proper contact with the world? ... while minimizing contact with the highest-risk sources of danger, such as fecal matter or stagnant water.

re: proper contact with the world, I think it’s probably really bad, actually, for most people, to be isolated from the physical environment, for a complicated web of reasons. They’ve done studies on touch, and they’ve done studies on interpersonal connection, and they’ve done studies on how much play children get, and they’ve done studies on outdoor time in general, and none of these have (to my knowledge) directly asked the question “is it bad to isolate your sensory system from the environment? Is it bad to have your sensory inputs be under your direct control, and all predictable, and all minimized or optimized to be pleasant?” and my sense is that the answer to that question seems ... pretty obviously “yes,” like I wouldn’t be shocked to be convinced otherwise but the prior seems to be yes.

I can’t see how it would be sane or reasonable to expect “following the optimization chain for raw sensory comfort” to end up in a healthy place, given that following the optimization chain for e.g. pleasurable tastes brought us to obesity and following the optimization chain for nice sensations in the brain brought us to crack cocaine. I feel like I see a general pattern of, like, it’s generally good to push 20% in the direction that our nerves want us to go, but if you push 40% things start to go haywire and unfortunately if you push 60% (or if you push too far during formative years) you end up in a place where your own nervous system is like ... driving you further into the darkness?

Something something, people’s feet are atrophied and then the doctor gives them orthotics that atrophy their feet even further, resulting in a prescription for even more orthotics. I feel like I see a pattern where touch-starved people find touch uncomfortable and challenging and because our society is crazy it doesn’t try to head this problem off at the pass but instead leans into it and those people further isolate themselves in ways that deepen their inability to have a healthy relationship with touch,

and like, sure, fine, there’s no moral component to touch, there’s genuinely no direct “should” around letting people touch you, but also it seems pretty clear to me that these people are fucked up in other ways that descend from this. Like, I have in fact never encountered someone who was like “nobody touch me” who was also fine.

And interpersonal touch is like one small narrow example; my thesis is “we’re built to have the world touch us, a lot, intimately, all over our bodies, in ways that are not necessarily pleasant (like prickly grass and stuff) and our reluctance to let the world touch us, as we gain the power to actually control our sensory inputs, is leading to local comfort at the cost of global health.”

There's a thing here that feels socially risky in the same way as talking about how carrying a lot of fat is unhealthy; I feel like people who disprefer touch might (quite reasonably!) get kinda prickly about how I'm pointing at How They Are and making a claim about it being Generically Bad in a way that doesn’t address the nuances of their particular situation. And just like with people who are carrying too much fat (in some sense) obviously the details do matter, and in particular there’s like … the reality of the situation as it presently exists, like even if I’m correct that someone who is [touch-averse downstream of being touch-starved] is kind of ... damaged, and unhealthy, that doesn’t mean that the correct response is to force them to absorb a bunch of touch that they don’t want and can’t healthily receive, any more than the right response to an overweight person is to make them feel a bunch of negative emotions that don’t even help them change things.

But like if ten thousand median Americans were suddenly dumped into Agor, Agor would be trying to develop some sort of actually effective remedial touch program, to help people who’ve been driven out of the healthy zone in a self-reinforcing way somehow ... find their way back?

(Or I guess maybe transhumanist concerns would take over instead, and we would find ways to fix all the bad stuff that’s downstream of the touch thing and leave the actual preference alone, but like, I suspect that the transhumanist solution is 10-100x harder and is farther away and in fact the correct response is to restore the evolved agent-environment balance rather than trying to build the next-gen equilibrium.)

But yeah, the overall suspicion is like:

• Being dirty in ways that don’t actually cause disease is genuinely good, like not just neutral but actually a critical part of a healthy psychological and biological equilibrium, given what our nervous system is evolved to “expect,” we are a social grooming species and we are way below base rates for what we almost certainly need, like sure there are always going to be people off to one side of the bell curve who genuinely need less touch but the median and modal American is not getting the median and modal correct-amount-of-touch that a human should get, either from other humans or from the grass and the dirt and the mulch and the breeze and the rain and the bugs, when was the last time you went a full week without putting on shoes
• Being clean is pleasant, and healthy to a point, but you can in fact get Too Clean
• Once someone gets accustomed to being Too Clean, they sort of fall down an attractor that makes them inordinately averse to skin sensations and inordinately averse to touch and stuff and this is bad, there is a stable equilibrium at the bottom of the slippery slope where a person is hurting themselves by isolation and also extremely loath to un-isolate

VIII.

This is the part where we talk about hubris.

There’s a well-known principle in design, that you should trust your audience to identify problems, and you should almost never accept their proposed solutions. People are very good at identifying what isn’t working, and very bad at fixing it.

(Or, to quote Henry Ford, “If I had asked people what they wanted, they would have said ‘faster horses.’”)

The overall moral that I’m taking away from these observations is something like my own preferences are often ‘wrong.’

Not wrong in some fundamental moral sense, but wrong in that they will lead me astray, once I exit the context that those preferences evolved to take for granted. Wrong in that indulging my preferences, in this new, unbounded, table-less world, will lead me to some deeply unhealthy, unhappy places.

(A heroin addict’s desire for heroin is miscalibrated, and the median human’s natural response to the superstimuli of modern food and TikTok is barely better.)

In the ancestral environment, if I wanted to wallow in sadness, there’s only so long I could do so before I would have to get up and do some chore associated with keeping myself alive, or join in with the rest of the tribe in some group endeavor.

Today, I can hole up with a month’s worth of food, draw the curtains, and play Moonlight Sonata on repeat for as long as I feel like (which will be longer and longer the more the mood reinforces itself) without ever having to leave a ten-meter radius.

That’s, um. Dangerous. Hopefully it is obvious that that is dangerous. My brain was not built to maintain balance under that sort of pressure. Even one playthrough of Moonlight Sonata has more emotional power than 99.9% of the things I might encounter in the ancestral environment.

I think that, in general, we are not suspicious enough of our own desires and preferences, where those desires and preferences developed in an environment with constraints—constraints that are no longer present. I think most people kinda-sorta get it when it comes to [delicious omnipresent food] + [lack of need to exert ourselves] leading to obesity, but I think these dynamics are present in many more places than that. There are a lot of tables disappearing, these days, and a lot of cups falling very far, very fast.

This is a controversial take. I’ve made specific versions of it, before, and people (pretty reasonably!) got uncomfortable about it.

(How much should we trust our drive toward comfort?)

Earlier this week, I was reading an essay by Aella about her upcoming event, Slutcon, and I encountered this paragraph about courting behaviors in modern society:

The female-dominated alternatives are even worse. Most women don’t know what they want, and if they do they are pretty bad at communicating it, and so their sexual advice tends to be ineffective at best and actively damaging at worst. Heavy focus on safety and deference are there for good reasons, and they need to be respected, but often are deeply unsexy, no matter how much people insist otherwise.

Or, in other words: the courtship you’re asking for, that you want, isn’t the courtship that, on some deeper level, would actually satisfy you.

There’s something pretty risky about pointing out “hey, actually, I think that our baseline standards for something like safety and control may have gotten out of hand,” which is why I’m leaning on Aella for legitimacy instead of just saying it myself.

(And hopefully it goes without saying that the safest quartile might be overdoing it, and the most vulnerable quartile might still be drowning in abuse and disenfranchisement, and advice for the safest quartile is not meant to apply to people in the most vulnerable quartile.)

But I think it’s probably true that most of us have something like an approximately insatiable drive toward more safety and more direct control. That most of us would click “more” rather than “less” every single time we were given the choice (just as most of us would take more comfort, more delicious food, more leisure time, more entertaining entertainment, etc). I think this is true of us because, under the constraints of the ancestral environment, that policy led us to scrape together a barely-tolerable minimum of safety and control, and so the drive got reinforced.

But now the table is gone, and the cup is falling. And while we can and indeed should do better than the standard of our ancestors, I don’t think “infinity” is the correct stopping point. I think that just letting insatiable drives take over is not likely to leave us happy and healthy and satisfied.

(Most people find video games less fun and less fulfilling when they have cheat codes for invulnerability and super speed and infinite ammo and walks-through-walls and and and and and and—)

I think that humans want things to be clean and soft and predictable, and I think that (at least in the more prosperous parts of the world) they’re becoming too clean and too soft and too predictable. Not in a bullshit “you gotta be tough” sort of way, but in a “no, you’re actually giving yourself an autoimmune disorder” sort of way.

I think we want to be safe, and to control the boundaries of our own personal space, and not experience unpleasant surprises and trespasses, and I think that at least some of us are succeeding too hard, just as we succeeded too hard at creating abundant food and removing the need for physical exertion.

That’s all I’m saying, to be clear. I’m not actually staking a claim as to where the correct stopping points are, for all of these various things. But I think very few people are adequately skeptical of their own preferences, and I think very few people are actually looking straight at the problem, and talking openly about it.

(And I think looking straight at it and talking openly about it is a prerequisite for actually figuring out the right answer.)

IX.

Okay, so … what now?

In the short term, and on a personal level, the recommendation is pretty simple: when you notice that you want something—that you’re driving in a direction—pause and ask yourself: in the ancestral environment, would there have been some kind of constraint that limited the expression of this desire? Some table that kept the cup from falling?

And if so: is that constraint now gone, or loosening, such that I have to construct my own stopping point, rather than being able to rely on external factors to keep me in the right zone?

That’s pretty much it. Anywhere that the table has disappeared, make some sort of preliminary decision about how much you think is enough, according to your best ability to guess. My body wants 5000 calories a day, but nutrition science leads me to believe 2000 is more likely to work out, in the long run.

Unfortunately, “simple” doesn’t necessarily mean “easy.” I do think a massive chunk of this problem is solved by simply becoming aware of it (which is why I writ this here essay), but solutions that require noticing a thing, on purpose, are hard for a lot of people. There’s not much I can do about that here, but you might read this sequence of eight mini-essays if you’re interested in leveling up in that domain.

In the meantime, may you only mostly get what you’re wishing for—enough to make you happy, and not enough to drive you nuts.