Meadow Theory

Author’s note:

The essay below centers mostly on a philosophy of parenting. However, it’s not “about” parenting—rather, parenting is a domain in which the concepts I’d like to convey can be easily represented and productively applied. Parenting is one specific instantiation of the broader set of ideas, just as one might use an essay about football or baseball as a vehicle for the concepts of friction or acceleration or collision elasticity.

Author’s note II:

Feedback on this draft ran thus:

[this essay seems to be missing something important, namely] stuff that would flip me from the primary activity of “i am trying to download Duncan’s meanings into a sandbox in my mind and then make sense of them in the sandbox before i possibly do the further work of rendering them meaningful and/​or useful to me if i have the tenacity for that”, to the different primary activity of “i am using Duncan’s guidance to recognize in my own head and from my own perspective the possibility of relating to the world in this new way, and i am personally engaged in the process of figuring out when and how and why it might matter to me”

I’m not sure how to get readers to do the latter thing; I think it is the correct thing to do and had sort of implicitly assumed that everybody already would. Since I don’t know what changes to make to this draft to encourage the latter thing for readers like the one above, I’m just bluntly making the recommendation. Hopefully I will get better at helping in the near future.

I. The meadow

Imagine a wide, level, grassy meadow, stretching to the horizon, empty of all obstacles.

This meadow will be our metaphor for the world/​the territory/​reality-as-a-whole, and the picture of it will gradually grow more complex as we step through various thought experiments. But for the moment, it’s just one big blank plane.

Now imagine a blindfolded child, running through the meadow.

This is safe, because the meadow contains nothing besides soft, springy grass. There are no walls or other obstacles to run into. If they trip or tumble, they’ll be fine.

Running will be our metaphor for human activity. It is good to run. More precisely, it’s good to be able to run—to have the freedom to move in whatever direction you please, at whatever speed you please, without fear or constraint. The ideal state is one in which the blindfolded child can run as much as they want to, wherever they want to, whenever they feel like.

(The blindfold is our metaphor for the state of human knowledge—our inability to perceive and comprehend the vast majority of what goes on around us, and our uncertainties and confusions about even the tiny slice we do manage to be aware of.)

II. The post

Now imagine a parent, lounging on the grass, idly supervising the child as they run.

The parent is not blindfolded, because the parent has a wealth of experience and knowledge (relative to the child). They can “see” the world more clearly, so in our metaphor they can straightforwardly see.

(Though they are still unable to see what’s behind them, or what’s very far away, or what they fail to properly attend to, or things that blend in with the grass or the sky.)

In the infinite meadow, the parent’s job is extremely easy. The child is safe; they will run and play and eventually tire and come back to the parent for food and rest and conversation.

But imagine for a moment that the meadow is not empty. Imagine instead that it contains one single post, standing upright in the middle of everything.

Art by Logan Strohl

Running into this post would not be pleasant. If the child does so, they could be seriously hurt, or possibly even killed.

A handful of possible responses to the inclusion of the post:

  • The parent might simply go sit by the post themselves, such that they could shout a warning to the child if they get too close.

  • The parent might accompany the child on their romp around the meadow, intervening if they threaten to get too near to the post.

  • The parent might lead the child away from the post—far, far away, until it ceases to be a salient problem.

  • The parent might encourage the child to slow down and be careful.

...there are certainly others. The parent might attempt to dig up the post, for instance, or surround it with soft objects. Or they might say nothing and “let the kid learn their lesson.”

It seems reasonable, though, to say that the job of the parent is to somehow navigate the child-post interaction. That what a parent is is someone who takes responsibility for figuring out a policy for dealing with this post situation.

(Or what a manager is, or a principal, or a general, or anyone who is in any sort of privileged position in which they have greater power and intel than other beings in their circle of concern.)

III. Expansiveness (and its opposite)

We will return to the parent later. First, though, I want to talk a little bit about the internal experience of the blindfolded child.

I would like to use the word expansiveness (as in “an expansive mood”) to refer to the property referenced above, of having the freedom and confidence to move at any speed, in any direction, following the whim of the moment.

In the infinite, empty meadow, it’s easy to feel expansive. To not-closely-track one’s own position, to not-think-about which way one will move next, to not treat any given impulse as a weighty decision in need of careful evaluation.

But imagine that you are the blindfolded child, and that somehow you suddenly become aware of the existence of the post, without knowing its precise location.

I claim that, in that moment, you would be very likely to undergo a contraction. To make a substantial shift away from freewheeling expansiveness toward anxiety, wariness, a sort of feeling-one’s-way-gingerly-forward. What makes sense in an infinite meadow makes less sense in a meadow known to contain a dangerous obstacle. The fear of possibly slamming into a post has an immediate impact on one’s estimation of how good of an idea “running around” is.

This is bad. Or, more precisely, it is the claim of this theory and this philosophy that this is bad. It is correct to contract in response to danger, but expansiveness, to the greatest degree supported by the environment, is the goal. Contraction is often necessary and justified, but it is always seen as an unfortunate cost. It’s one thing to choose to walk. It’s another thing altogether to be unable to freely run, because of an expectation that doing so will result in pain or injury.

IV. Uncertainty

An interesting thing happens, if you are the blindfolded child and you are standing at the post.

If you have your hands on it, and you know with certainty that it is the only post, then suddenly you’re back to expansiveness. As long as you run away from that spot, you can carry on as you were before... first. As you continue to meander, your uncertainty about the location of the post grows. At first, you are quite certain that it’s a dozen steps behind you—that if you turned around and took a dozen steps back and waved your arms, you’d bump into it.

But after you run for a bit, and then pause, and then get up and run a bit more, and blindly turn and turn and turn again, each time with a little more room for error about just exactly how much, it becomes harder and harder to have any confidence at all that you know where the post is, and could find it again (or avoid it) on purpose. It doesn’t take long at all before your-anticipations-about-its-location have become so diffuse that it might as well be anywhere.

In fact, the above picture is slightly misleading, because it implies that you/​the child/​the blue dot can confidently locate itself within the meadow. From a dot-centric perspective, with the meadow expanding effectively infinitely in every direction, and the dot’s own orientation toward or away from the post becoming increasingly uncertain as time goes on, the t=5 step looks more like:

There is (of course) a range of responses to personal risk. Some people would experience greater contraction than others, for a given likelihood of running into the post; there are some people who aggressively insist on “still running anyway” even when they know there’s a post somewhere out there, and others who creep and crawl and feel their way forward even when the odds of encountering a post are very low.

But it seems to me that the impact of maybe-there’s-a-post is of the same kind in all cases, even though people have different sensitivity to it and respond in different ways.

In “The correct response to uncertainty is not half-speed,” Anna Salamon notes that many humans respond to uncertainty by doing something like averaging across strategies. That is, if they’re not sure whether they’ve already passed their destination, or whether it’s still ahead of them, they will tend to slow down, even though this is a worse strategy in each of the possible worlds.

This rhymes in my mind with the tendency of many humans to treat a 50% chance of guilt (and correspondingly a 50% chance of innocence) as if someone is 100% likely to be sort of guilty. It’s rare to find people for whom split and commit is a natural or well-practiced move; as a class, we are not particularly good at composite strategies, especially if we are not paying close and effortful attention.

And so it seems safe to me to predict that, as one’s uncertainty as to the location of the post “bleeds” further and further out, tainting more and more of the meadow, most people will have a duller, slower, sadder, and more contracted experience in every square meter, and during every time step. Even a very low risk (say, 0.1%) of encountering the post in the next ten steps tends to temper one’s ability to be truly expansive and carefree—it’s rare for people to respond to small risks with a correctly proportionately small behavior change.

(c.f. concepts such as risk aversion and loss aversion)

It’s for this reason that meadow theory puts a premium on identifying the precise location of dangerous obstacles. The more accurately you can pin down exactly where the post is (and continue to track it even as time passes and you move around), the more the existence of the post does not result in general contraction.

V. Parenting under meadow theory

This, then, is the primary responsibility of parents (and managers, principals, generals, etc.) under meadow theory: to help the blindfolded children locate the obstacles within the meadow. If a child knows where the hazards are, they can still be relatively expansive—there’s plenty of joy to be had running full-tilt down hallways, even if the walls mean they can’t turn left or right at will. But as their uncertainty rises, their ability to run freely sharply contracts.

In the latter, more uncertain situation, there is only a narrow band in which it is at all reasonable to run freely, even though the actual hazards are the same in both cases.

This responsibility expresses itself in two broad categories of parental action. The first is providing specific warnings about the locations of specific, known hazards, e.g.:

  • “Look both ways before crossing the street, so you don’t get hit by a car.”

  • “Glass cookware doesn’t look any different when it’s very hot.”

  • “Don’t experiment with highly addictive drugs. You can get yourself caught in a trap that is very hard to escape, and the point-of-no-return is often impossible to identify until you’ve already passed it.”

  • “There are people who have a lot of power within their tiny fiefdoms, and sometimes those people behave really poorly, and it will be tempting to challenge them, and sometimes it’s worth it to challenge them, but you should stop and think for a minute about whether you’re making a powerful enemy and whether that’s the right tradeoff given your goals and needs.”

  • [My psych nurse mother’s extremely blunt and not-at-all-trying-to-be-sympathetic-or-politically-correct heuristic] “Boys: girls with borderline personality disorder will seem extremely interesting and sexy and into you and they will draw you in and then everything will be terrible. Girls: boys with psychopathic tendencies will seem extremely charming and capable and deep and they will draw you in and then everything will be terrible. Each of you needs to learn to recognize these types early, and steer clear despite what your libido is telling you.”

...this list could go on for quite literally thousands of entries; human culture has built up a large repository of informal knowledge about where many hazards lie, and each individual adult human brain contains a significant fraction of that total.

No one human, though, comes anywhere close to carrying a complete copy, and even the sum total of all human wisdom regarding posts-in-the-meadow is a woefully impoverished subset. Known dangers present themselves anew in endlessly varied ways; there are dangers which no one has encountered yet; there are dangers which did not exist until we created them (e.g. the profoundly negative impact of Instagram on the self-esteem of many young girls); there are dangers which we don’t know about because no one has yet survived an encounter and come back to warn the rest of us.

Thus, the second category of parental action is teaching children the general skill of how to recognize a hazard before you smack into it face-first. Of knowing what sorts of signs mean that one should slow down, because one has entered a region that’s less safe than usual.

(Metaphors here are less clean, but one could imagine e.g. human echolocation, which is indeed sensitive enough to allow blind humans to detect and avoid many dangerous objects. Or one could imagine a kind of outside view aggregation—if you are moving through the meadow and suddenly all of the voices fall away behind you, this may be a sign that you are entering dangerous territory which is unpopulated for a reason.)

Some vague gestures in this direction:

  • Literal unexplored territory is often dangerous; if no one’s been there before, no one knows where the posts are, and it may pay to move slowly.

  • If someone could directly and significantly benefit from deceiving you or defrauding you or otherwise imposing costs upon you, there’s a higher chance that this will actually happen, and you should keep your wits about you.

  • If someone is proposing that you give up substantial power, resources, or mobility, this is often a bad sign.

  • If someone is contriving a situation in which they have substantial and open-ended power over someone else, this is often a bad sign, especially if the someone else is supposed to be dependent on or otherwise vulnerable to the more powerful person.

...etc. Each of the items in the above list is a generator for items in the first list; they are generalized inductions from the set of historical observations of posts that help people to focus their attention in places where new posts are more likely.

It’s worth reiterating that these two precepts of meadow theory are different from many other parenting philosophies. Meadow theory says that you, as a parent, should:

  1. Convey to your child the most precise information you can about where the hazards are (and, implicitly, what kind of risks they carry).

  2. Teach your child how to recognize and locate previously unknown hazards, such that they can navigate around them with a minimal sacrifice of meadow-area-in-which-they-can-freely-run.

...and furthermore that this is the end of your responsibility, at least when it comes to the question of defending your child against the environment. Other theories of parenting say that you should e.g. prevent the child from running into a post by any means necessary, including those which are highly costly to you, the child, or your relationship (such as by keeping them confined to a known-safe area of the meadow), or that you should dissuade your child from ever running fast enough that they would be injured by a post, if one were there, regardless of how likely there is to be one.


VI. People-as-posts

Assuming that you are reasonably on board with the claim that uncertainty about the location of the post leads to contraction, it’s worth taking a brief aside to note that other people are also posts.

And indeed, as the above uncertainty principle would predict, it is when other people’s boundaries and norms are clear and unambiguous that they are easiest to interact with.

With total strangers (who in America at least we largely know not to touch) and close friends (whose willingness to be touched is understood in detail), it’s relatively simple to answer questions like “should I give this person a long, lingering hug?” Ditto for contexts where roles and norms are explicit in common knowledge (e.g. a corporate office, or a martial arts dojo, or the Sunday service at a church, or a married couple interacting in the privacy of their own home).

Where it becomes difficult is with people who may want a hug, but who may instead punish you for offering one, or who may be open to hugs in one context and strongly opposed to them in another, for reasons that are not immediately apparent or legible.

In all sorts of domains—money, romance, political or religious affiliation, physical interactions, communication norms, tidiness—it’s when people’s true boundaries are unknown or highly variable that others must contract, cautiously feeling their way forward.

(Or barrel forward blindly and hope for the best.)

The more that each of us knows what the other likes and dislikes, wants and avers, can handle and can’t handle, the more each of us can make frequent and confident motions even fairly close to the line, without having to worry about unknowingly and accidentally crossing it.

The less defined those boundaries are, the more they bleed out and pollute the space between us such that neither of us can feel expansive (if we care about not violating them).

Thus, it is useful and prosocial to make the locations of one’s own boundaries and fences and wants and needs outwardly legible. The more people have to tiptoe carefully around one another’s hidden landmines, the worse things are for everyone. Parents, according to meadow theory (and managers, principals, etc.), should help their children locate and assert their own boundaries, and should model setting and clearly communicating boundaries in a way that the children can see and learn from, not merely because this is good for one’s own sake but because it causes each person to accidentally steal less of the common space via uncertain boundary leak.

Hazards that everyone can “see” are much easier to avoid, and much easier to pass by closely, at speed, without danger.

VII. Interlude: Logan Strohl on courage

(Compiled from multiple sources)

One thing that’s really wrong with our subculture (by which I don’t mean “rationalists” so much as “leftists”, or possibly even “contemporary Western society”) is that we think we’re supposed to feel safe.

We notice that in contexts with a whole lot of safety, where we do not feel afraid, it is much easier to be full, vibrant people, to expand, to act freely in accordance with our values. This is an accurate observation, and working to create and maintain such contexts for ourselves and others makes a lot of sense.

But overall, no matter what we do, we are not safe. This world is a dangerous and terrifying place. Even rich white men get cancer. The other monkeys are watching and judging. Many of us are aware of the possibility that everyone and everything we care about might be re-purposed atom-by-atom if AGI takes off.

There are no safe spaces. If I prefer to believe true things, it is actually appropriate to be at least a little afraid basically all of the time.

Because we think we’re supposed to feel safe, and because we’ve noticed ourselves expanding in safer contexts, we treat fear as an enemy holding us back. And since we’re usually afraid, we’re usually held back. We feel that our only options, if we want to act on the world, are either 1) to eliminate fear by creating and maintaining safety (or by simply ignoring dangers altogether), or 2) to spend most of our energy trying to shove down our fear or to fight against it.

We, as a culture, have forgotten what courage is. What it’s for, and what it’s like, and how to have it.


Often, I’m drawn to do something, and also I’m afraid to do it. One really fruitful action I’ve been taking when I notice the attracted/​repelled phenomenological cluster is to ask first what I’m afraid of and what my fear is trying to protect, and then what I’m eager to do and what my eagerness is trying to protect.

Surprisingly often, my fear and eagerness are trying to protect the exact same thing, and the tension I experience comes from a disagreement about the appropriate strategy.

For example, I’ve at times been both repelled and attracted to hiking in the desert. My fear is of rattlesnakes, and my fear of rattlesnakes is trying to protect free and powerful movement (I want to keep the ability to run and hike and climb, which rattlesnake bites threaten). My eagerness for hiking in the desert also protects free and powerful movement (which is not so available when I stay inside all the time).

Another example, one I’ve run into multiple times, is the simultaneous attraction and aversion to the idea of talking to someone, especially someone I don’t know very well. I’m afraid of talking to them, or of having a “real conversation”, and the fear is trying to protect a fulfilling relationship that might exist in the future. The fear recognizes that a conversation-gone-wrong could cut off the possibility. My eagerness to talk to the person is also trying to protect a fulfilling relationship that might exist in the future, and it recognizes that a conversation is just about the only means by which the possibility could be instantiated.

[edit: the following is definitely not the full story, it’s more like musings that contain accurate observations incorrectly interpreted]

Perhaps caution is the attempted preservation of possible value in more distant imagined futures, while bravery is the attempted instantiation of near-present value. Caution and bravery are often championing the same values while suggesting different strategies. A strategy that favors caution by default without taking bravery seriously is called “cowardice”, and looks like unwillingness to act. A strategy that favors bravery by default without taking caution seriously is called “recklessness”, and looks like unwillingness to think.

Which means, perhaps counter-intuitively, that courage tends to require patience. Courage is the strategy that takes caution seriously and, having done so, chooses the attempted instantiation of near-present value. It doesn’t require the indefinite, endlessly deferred patience of cowardice. But it does require perseverance, the willingness to go on working at a problem despite confusion and frustration. Sometimes it even requires the tolerance of ongoing discomfort that it takes to passively allow thoughts and observations to arrange themselves by “sleeping on it”. It’s not always obvious what you’re afraid of, what the fear is trying to protect, or what the world would look like if caution were the appropriate strategy. I think you have to load all of that somehow to be courageous.

In these situations where hesitance and eagerness champion the same value, you can’t just be like, “well, which do I care about more?” because you’re weighing one thing against that exact same thing. They really take a lot of patience.


Fear is not an enemy. It reminds us that what we value is in jeopardy, and it’s one half of the scarab beetle. The other half is awareness of our values themselves, that spark of recognition of something excellent and beautiful in the world. When we hold both halves at the same time, fitting them one into the other, the beetle comes to life, and flies directly toward the Cave Of Wonders. We move through danger toward what we care about. And that is courage.

Courage is the capacity to act from your values in the presence of fear. It is expanding yourself when you are not safe. It is feeling fear, listening to its accurate descriptions of real risks, and fighting with your whole heart for all of what you care about.

And something I personally have been missing this whole time is that courage—real courage—feels good. Not necessarily soft or warm or pleasant, but wonderful nonetheless.

It is not at all the thing where you shove down your fear and try to power through with your hands over your eyes until you’re allowed to stop. And it’s nothing like the frantic nausea of pretending you’re safe when you aren’t. It’s electrifying. It’s suddenly arriving at the place where you’ve been standing but were too asleep to notice. It’s the opposite of giving up.

We’re not supposed to feel safe. We aren’t safe. We are in danger, and we’re supposed to feel courage.

VIII. Summary

  • Reality is well-modeled by a meadow, through which we wish to have the affordance to run freely.

  • There are hazards in the meadow, which some of us (such as parents) see more clearly than others of us (such as children).

  • Uncertainty about the location of the hazards is corrosive to the ideal condition of expansiveness. (Actual injury from colliding with a hazard is also corrosive, both directly and in the way it changes one’s future behavior.)

  • Increasing certainty about the location of the hazards restores the ideal condition of expansiveness, at least to the greatest extent allowed by the environment.

  • Among the hazards that cause people to contract in a bad way are other humans, and other humans can also have an ill-defined “location” in that it can be very unclear at what point they will “hit back,” the way a post hits back when you run into it at high speeds.

  • The primary responsibility of cooperative individuals (such as parents helping their children but also colleagues and allies helping one another) is to cause one another’s maps of the meadow to become more and more accurate, such that everyone knows where the hazards are.

  • A secondary responsibility of cooperative individuals is to help one another to develop the ability to perceive and map previously unknown hazards (since we do not have anything even remotely close to a complete understanding of where all the hazards lie).

I claim this is an extremely useful way to think about parenting, as well as things like difficult interpersonal conversations, or project management, or teaching/​pedagogy, or directing literal troops in a literal war.

I don’t think the above represents the sum total of useful actions to take within the meadow. There are other things that people can do which are helpful, such as bulldozing hazards to make the meadow safer for everyone, or building small closed-in enclaves for meadow-runners who are unable to perceive or comprehend certain hazards.

But the first priority should be to help people precisely locate known hazards, such that they can be reasonably confident about where those known hazards aren’t, and the second priority should be to translate the hazard-locating skill. Everything else seems to me to be dependent on something unsustainable (e.g. “I’ll sacrifice my own priorities entirely and devote my time and attention to the task of keeping you safe myself”) or unrealistically optimistic (e.g. “we’ll make the entire meadow entirely safe”).