Learning to walk through walls

Hi everyone,

I have a running joke that one of the most useful things I do when coaching or consulting is to say to people “Yes, that does sound like a problem. Have you tried solving it?”¹

Part of why this is a joke is that actually most of the useful work happens prior to the point - the hard part is actually articulating what is going wrong well enough that it seems like a soluble problem - but there is genuinely something useful about this, because often it feels people are looking for permission.

Without the external prompt, solving their problem is not something they noticed that they were allowed to do.

Doing what is done

A while ago I wrote about social reality as a game. The specific prompting event was that I had been invited to come out for something that was happening right there and then, but had just made coffee, and I didn’t have a travel mug to bring my coffee in. What did I do? Well I just brought a normal ceramic mug, obviously.

Walking around outside carrying an ordinary ceramic mug is weird though. There’s not actually anything wrong with it. It’s not really that much worse for this than a travel mug, but it’s in some sense not done.

Here’s a related passage from Terry Pratchett’s “Interesting Times”:²

Cohen walked over to the nearest wall, which had a gorgeous pattern of peacocks, and took out his knife.

“Paper,” he said. “Bloody paper. Paper walls.” He poked his head through. There was a shrill whimper. “Oops, sorry, ma’am. Official wall inspection.” He extracted his head, grinning.

“But you can’t go through walls!” said Six Beneficent Winds.

“Why not?”

“They’re—well, they’re the walls. What would happen if everyone walked through walls? What do you think doors are for?”

“I think they’re for other people,” said Cohen. “Which way’s that throne room?”

“Whut?”

“This is lateral thinking,” explained Mr. Saveloy, as they followed him. “Ghenghiz is quite good at a certain kind of lateral thinking.”

“What a lateral?”

“Er. It’s a kind of muscle, I believe.”

“Thinking with your muscles…Yes. I see,” said Six Beneficent Winds.

Cohen the barbarian and his band of fellow barbarians (and one retired school teacher, an honorary barbarian) are invading the imperial palace. Unfortunately the way in is well guarded, with many troops occupying many twisty passages, and they’d frankly rather not.

Fortunately, they can just walk through the walls. This has not occurred to any of the defenders, because they’re walls, you don’t walk through walls.

The difference between Cohen and the palace guards is that the palace guards are inhabiting a social reality, and Cohen is inhabiting physical reality. The guards are thinking in terms of what is done, while Cohen is thinking in terms of what he can do.

He is, in a very real sense, thinking with his muscles. He is acting as a physical entity in a physical space, and letting that physicality be the primary tool through which he expresses his agency. The guards are living in a world of socially significant actions constrained by rules, while Cohen is living in a world where physical actions are defined only by what happens in the external world when you do them.

In contrast, the guards regard the walls as constraints. This is not because they’re any less aware, in principle, that you can walk through paper walls, but because they are obedient to the relevant rules of social reality.

As I point out in “social reality as a game”, Bernard Suits’ definition of a game captures what it is like to inhabit these constrained social worlds quite well:

To play a game is to attempt to achieve a specific state of affairs [prelusory goal], using only means permitted by rules [lusory means], where the rules prohibit use of more efficient in favour of less efficient means [constitutive rules], and where the rules are accepted just because they make possible such activity [lusory attitude].

I also offer the following simpler and, so to speak, more portable version of the above: playing a game is the voluntary attempt to overcome unnecessary obstacles.

Often social reality isn’t exactly a game in this sense, in that you will be punished for going outside the rules so it’s not undertaken voluntarily³, but in situations like the coffee cup or the thinking of the guards there is no external punishment. There is, at most, your own internalised version of that punishment. The only thing holding you to the constraints of social reality here is that you choose to - a voluntary attempt to overcome unnecessary obstacles.

There’s nothing wrong with playing games, but when your main goal is to achieve some outcome, this is often much harder if you confine yourself to the game’s rules, and you can solve many more problems if you’re willing to break out of the game or if, like Cohen, you were never playing that game in the first place.

Playing restricted games

I am a huge fan of the book “Games: Agency as Art” by C. Thi Nguyen. In it, he argues that games are an art form whose medium is agency - they create art by creating systems of actions that you can perform to play the game. When playing a game, you have a set of circumscribed means of acting (your lusory means - the actions permitted by the rules of the game), and a goal you are trying to achieve (your prelusory goal).

Learning to play a game corresponds to developing the skill of acting well within the rules - of performing a particular agency. Playing the game requires learning to be a particular sort of agent.

My friend Cedric has a great piece, Are you playing to play, or playing to win?. In it, he quotes Lesley (who I also know, although less well): “make sure you’re playing the real game, not some more complicated game you’ve made up for yourself.”

This is a good aspiration, but I think “more complicated” is doing a lot of work in this sentence, and in reality most people playing games are playing simpler but harder games that they’ve made up for themselves (although, typically, that they’ve made up in collaboration with others).

In this piece, Cedric cites the following example:

American Olympian Jimmy Pedro said, of his father: “My dad was a guy who started Judo when he was 19 years old. So he was way behind the eight ball against everyone else; he was trying to go to the Olympics. Gripping was something that was introduced to him in his early 20s. And he realised that if he learned to grip fight properly, he could beat those technicians who had been playing Judo since they were five or six years old. Because it’s hard to get a natural feel for the sport when you start so late — like you’re never going to be as instinctually as good as somebody who starts when they’re a kid. But my dad learnt that gripping was a way to get good fast.”

To expand on that a little, in Judo, grip fighting is what happens before you can throw. The competitor that consistently gets a superior grip determines the pace and shape of the match. Pedro and his father are perhaps most famous for systematising the grip fight — that is, they turned it into something that could be taught to all players, instead of relying on unsystematic tricks, or on intuition acquired through a high volume of sparring.

Cedric laments not encountering this attitude earlier:

I wished I’d learnt this alternative view of competition earlier in my life. As a competitor, I didn’t think very deeply about the constraints that I was facing, and I didn’t have a coherent strategy to win. I simply thought to myself: “Ahh, the Japanese play beautiful Judo. That’s the best form of Judo, and the only form of Judo worth playing. I want to be like that.” And then I refused to look at other styles.

I was, in other words, a scrub.

He quotes David Sirlin on scrubs:

"Scrub" is not a term I made up. It sounds like kind of a harsh term, but it's the one that was already in common usage in games to describe a certain type of player, and it made more sense to me to explain that rather than to coin a new term.

A scrub is not just a bad player. Everyone needs time to learn a game and get to a point where they know what they're doing. The scrub mentality is to be so shackled by self-imposed handicaps as to never have any hope of being truly good at a game. You can practice forever, but if you can't get over these common hangups, in a sense you've lost before you even started. You've lost before you even picked which game to play. You aren't playing to win (emphasis added).

A scrub would disagree with this though. They'd say they are trying very hard. The problem is they are only trying hard within a construct of fictitious rules that prevent them from ever truly competing.

Cedric was being a scrub in that he was trying to learn a particular style of Judo rather than learning to play to win. The existing Judo community were not being scrubs at first, though they would have been if they kept losing to better grip fighters and refused to adopt that style because they thought it was “cheap”.

Grip fighting is unsexy and very far from “beautiful” Judo that others were playing - those others’ own restricted version of the game - so by breaking out of that restricted game and playing in the broader game, Pedro’s father (James Pedro) could gain an advantage that others lacked. It is a Cohen the Barbarian style refusal to do what is done, and ask instead what he could do.

And what he did was invent his own restricted game to play.

Inventing symbolic games

A common feature of game rules is that the actions you can perform within them are discrete⁴. Consider something like a game of chess - you have a fixed set of pieces, that can move to a fixed set of squares. You can’t move your rook half a square forward, there’s no way you can invent a new piece, all you can do is push symbols around in a finite and discrete set of ways.

This isn’t intrinsic to games, mind. Judo is also a game and is one very embedded in physical reality and, while your actions are still circumscribed by rules, those rules define how you move about in physical space, operating in the continuous environment of the real world. There really is a difference between whether your grip is there or a couple centimetres down, and that difference may well make the difference between whether you win or lose.

And then we discretize it anyway.

A particularly interesting feature of Cedric’s description of what the Pedros did was the following:

Pedro and his father are perhaps most famous for systematising the grip fight — that is, they turned it into something that could be taught to all players, instead of relying on unsystematic tricks, or on intuition acquired through a high volume of sparring.

Systematizing something like the grip fight takes a continuous, physical, game and turns into a discrete, symbolic, one, by transforming it into something that you can talk about in words.

Because so much of how we learn skills in professional or competitive contexts is about Learning from others’ stories, our conceptions of skills are often very tightly defined by what we can talk about. We give names to things we do, and then we find ourselves fitting our actions to those names. We name a particular gripping situation or throw or the like, and now the thing we practice is doing that named gripping situation right, not how to grip our opponent in some general sense.

We don’t just practice one of course, because we need to have a variety of options available to us, but we learn enough distinct gripping situations that by practicing all of them to a high standard we are able to play the grip fight well.

This carves our goals within the continuous physical world into discrete steps. Instead of “grip your opponent well” the process becomes:

1. Decide which of the 10 gripping situations you know to get into and handle.

2. Get into that gripping situation.

I want to be clear here, this isn’t bad. It’s an incredibly powerful technique. It gives you a way to swap stories and tactics, and it gives you a very tight feedback loop: You can practice a grip in isolation, separately from the overall practice of fighting. Symbolization of a skill - reducing it to a series of discrete symbols that you can practice individually and then combine in application - allows you to become extremely good at a constrained subset of that skill.

You have restricted the game again, but as long as you can play your restricted game better than the people playing the unrestricted variant of it, you still win, and often you can because it is so much easier to become good at symbolic games.

Expanding libraries of agency

Because I’m a mathematician at heart, I’m automatically suspicious of any attempt to exclude the empty set. As a result, I think it makes sense to regard any attempt to achieve a goal as a game in the trivial sense - it just happens not to have any constitutive rules.⁵

We then impose rules on it ourselves, through a combination of social restrictions on what is done, symbolisation and other social learning about how to do it, and otherwise try to reduce the endless space of possibilities into a more constrained space of possibilities that we can practice enough to learn to be good at.

This works great, right up until the point where we forget that we’ve done it. The problem is not that we are playing restricted games. This is not scrub behaviour. The problem is when find ourselves in situations where we need to break out of the game, and we don’t do that - not because we can’t figure out how, but because we choose not to.

The easiest way to break out of our current games is the stories and tactics approach. Tactics we learn from others are often received in a spirit of “oh yeah good point, I hadn’t thought of that”, but sometimes it goes deeper and someone tells you about something that it literally wouldn’t have occurred to you that you could do.⁶

Another approach is to do what the Pedros did - study some subgame thoroughly enough that you develop a whole new set of tactics for it that were not previously available. The world is infinitely complex, and so most areas are more amenable to reveal interesting things through careful study. The current fastest way to play Tetris was only invented in 2020.

But I think the most common, and most accessible, way is that when you get stuck, take a deep breath and remind yourself that you’re not actually playing a game, you’re an actual real live human inhabiting a real world that actually exists, and maybe you need to deploy some lateral thinking and see if you can walk through walls.

Postscript

Share the news!

If you liked this issue, you are welcome and encouraged to forward it to your friends, or share it on your favourite social media site:

If you’re one of those friends it’s been forwarded to or shared with and are reading this, I hope you enjoyed it, and encourage you to subscribe to get more of it!

Community

If you’d like to hang out with the sort of people who read this sort of piece, you can join us in the Overthinking Everything discord by clicking this invitation link. You can also read more about it in our community guide first if you like. We don’t talk about Judo much, but we do talk about getting better at a wide variety of other things.

Additionally, by being part of the community you get the uh privilege of seeing these posts when they’re much worse, as I tend to post first drafts there. Thanks to everyone there for reading a first draft of this post and helping me improve it.

Cover image

The cover image is a picture of a Judo throw made available by wikimedia user Schnuffel2002.

1. I very rarely say literally that, because it would be rude, but there is often a “Have you tried solving it?” component to what I do say.

2. This is a Rincewind book and thus one of the bad ones, but it’s also a Pratchett book and thus not without its moments of brilliance.

3. Whether playing a game has to be voluntary is somewhat questionable though. The participants in gladiatorial games, for example, are not there voluntarily. Less dramatically, a professional football player may be there voluntarily in one sense, but it’s also their job.

4. As opposed to continuous - e.g. the numbers 0, 1, 2, … are discrete, while 0, 1.5, 2.3, etc. are part of a continuous space because you can always wiggle any one a bit and get another number, there is no clear separation between the objects.

5. Note that this is very different from treating life as a game. Cedric in a follow on post The Dangers of Treating Life as a Game talks about this, and says:

I think it’s fairly obvious that real world games have no clear rulesets, beyond that which you discover to be true for yourself. But the most important limitation of treating life as a game is that all game analogies assume clear win conditions that are shared by all the players. Life has no such thing.

Even within a single life, there is no clear overarching goal (Sorry, Aristotle), but a life contains within it many attempts to achieve some specific goal, and those we may think of as games.

6. To continue the Slay the Spire running example, this is very much how I felt the first time I saw someone block the strength down from a flex pot with an artefact charge, my mind was completely blown. Undervaluing artefact and orange pellets is really common, and it substantially changed my gameplay once I realised this thing I had been completely neglecting.