Why We Think in Fences, Guns and Cats

We don’t think in spreadsheets.
We think in fences.
In guns on walls.
In cats in boxes.
In razors slicing away nonsense.

If you’ve ever heard of:

G. K. Chesterton’s Fence
Anton Chekhov’s Gun
Erwin Schrödinger’s Cat
William of Ockham’s Razor
Murphy’s Law
Cyril Northcote Parkinson’s Law

You already know this.

Notice something odd?
They’re all physical objects.
A fence.
A gun.
A cat.
A razor.

And yet they explain bureaucracy, probability, storytelling, engineering risk and workplace productivity.

Why?

Because humans don’t reason abstractly very well.
We reason with props.

The Brain Likes Objects

Tell someone:

“Do not remove legacy institutional structures without understanding their original purpose.”

Eyes glaze over.

Say instead:

“Don’t tear down a fence until you know why someone built it.”

Instant clarity.

That’s the power of metaphor.

Or take uncertainty.

You could say:

“A system exists in probabilistic superposition until measured.”

Or you could say:

“The cat is both alive and dead until you open the box.”

Guess which one survives in popular culture?

The Six Laws, Side by Side

Here’s what’s interesting.

These laws don’t just work once.
They generalise.
You can swap the objects and the logic still holds.

Here’s the expanded conceptual matrix:

	Fence	Gun	Cat	Razor	Law (Core Meaning)
Murphy’s	If the fence has a weak joint, that’s the joint that fails. In a storm, at night, when the repair crew is off.	If the gun can jam, it will jam at the worst possible moment.	If the box can fail, expect it to fail during the experiment.	If there’s a weak point, that’s where it breaks.	In complex systems, whatever can go wrong eventually will. Plan for the weak joint, not the average case.
Schrödinger’s	Until you check it, the fence is both sturdy and broken.	Until the trigger’s pulled, the gun both fires and doesn’t.	The cat is alive and dead until you open the box.	Until tested, multiple outcomes coexist.	A system stays indeterminate until observed or measured.
Occam’s	If the fence fell, start with “weather,” not “secret sabotage syndicate.”	If the gun fired, prefer the simplest explanation that fits the facts.	If you hear a noise, assume “cat” before “interdimensional portal.”	The best explanation is the one with the fewest extra assumptions.	Prefer explanations that don’t multiply unnecessary assumptions.
Chekhov’s	If you show the fence, someone’s climbing it later.	If there’s a gun on the wall in Act I, it had better go off by Act III.	If the cat gets dramatic lighting, it’s plot-relevant.	If you emphasise a detail, it must pay off.	Don’t introduce elements that don’t matter. Everything shown should serve a purpose.
Parkinson’s	Give yourself a week to remove the fence and it’ll take a week.	Give the story five acts and the gun fires in Act Five.	Give the experiment a month and it runs a month.	Work expands to fill the time available.	Tasks grow to consume the time you give them.
Chesterton’s	Don’t tear down the fence until you know why someone built it.	Don’t remove the gun from the wall until you know why it was hung there.	Don’t throw away the box until you understand what it was protecting.	Don’t simplify the rule until you know what problem it was solving.	Understand before you remove. Old things often exist for a reason.

A Small But Important Note on Schrödinger

The image of the simultaneously alive-and-dead cat is now so embedded in popular culture that it’s easy to forget Schrödinger invented it as a challenge, not a claim.

He considered the idea absurd. He was deliberately highlighting what he saw as a flaw in the Copenhagen interpretation of quantum mechanics: that if quantum superposition applied at the subatomic level, it would produce nonsense conclusions at the everyday level.

The cat became the most memorable thought experiment in modern science precisely because it was meant to be ridiculous.

That irony is worth holding onto. Sometimes the sharpest tools in our thinking kit were originally built to argue the opposite case.

What’s Actually Going On Here?

Each of these laws compresses a deep idea into an image:

Institutional memory → Fence
Narrative causality → Gun
Probabilistic uncertainty → Cat
Trimming assumptions → Razor
System failure → Murphy
Organisational drift → Parkinson

We remember the object.
The object carries the abstraction.
It’s cognitive compression.

Instead of storing a paragraph of theory, the brain stores a prop.

Why Metaphors Travel Further Than Whitepapers

Nobody quotes regulatory frameworks at dinner.

But everyone quotes:

“If anything can go wrong, it will.”
“The simplest explanation is usually the right one.”
“The cat is both alive and dead.”

These survive because they are visual.
You can sketch them.
You can stage them.
You can tell them as stories.

Metaphors are portable. And in the early stages of adoption, portability beats precision. Frameworks earn trust once people are already in the room. Metaphors get them through the door.

The Modern Twist

Here’s something worth noticing.

We now live in a world dominated by AI systems trained on human language.
And human language is full of these metaphors.

We taught machines using stories about fences and cats.

That has consequences. When a language model reasons about risk or causality, it isn’t working from first principles. It’s drawing on millions of human-written examples, almost all of which use physical metaphor as a shorthand for abstract ideas. The model has learned to reproduce the patterns of human reasoning, including the habit of reaching for physical props when grappling with abstract ideas. Whether it understands the fence is a separate question. The output looks the same.

Our mental shortcuts are now baked into the systems meant to extend our thinking.

So even as we move into abstract computational systems, our reasoning remains stubbornly physical.

We still need:

Something to tear down
Something to fire
Something to open
Something to trim

Because that’s how humans think.
Not in probability matrices.
In scenes.

So what? AI is not an alien intelligence. It is a mirror trained at planetary scale. If our language relies on metaphor and narrative shortcuts, those shortcuts are now automated and amplified. The danger is not that AI misunderstands reality. It is that it extends our misunderstandings with confidence and speed.