#5427: Why Atheists Still Knock on Wood

Religion and superstition aren't the same thing — and the data says they might even pull in opposite directions.

Featuring
Listen
0:00
0:00
Episode Details
Episode ID
MWP-5610
Published
Duration
19:23
Audio
Direct link
Pipeline
V5.2
TTS Engine
chatterbox-regular
Script Writing Agent
DeepSeek 4.1 Flash

AI-Generated Content: This podcast is created using AI personas. Please verify any important information independently.

Daniel is Jewish, believes in God, and isn't especially observant. He also can't say the word "death" out loud, and he knows Judaism doesn't actually forbid it. What interests him is that the reflex isn't coupled to observance at all — he knows atheists who nod along, and deeply observant people who'd call it medieval nonsense.

Start with definitions, because the boundary between religion and superstition is mushier than people assume. A superstition is a practice where the action isn't logically connected to the outcome — the mechanism is stipulated, not demonstrated. Religion is a coherent theological system with community, texts, moral framework, and calendar. But praying for a specific outcome looks functionally superstitious from the outside, and knocking on wood has no canon anywhere.

The empirical question is whether the two rise and fall together. They don't. When you control for general supernatural belief, religiosity and superstition come apart — and in a fair number of samples the correlation is negative. For a devout person, a lucky sock is a competing system, potentially an attempt to manipulate the world outside of God. For the non-religious, the sock is simply available.

Underneath both sits the same machinery: agency detection, and the reach for control under uncertainty. Religion channels it into structure. Superstition channels it into private, ad hoc levers. Sports is the cleanest test case — high stakes, uncontrollable outcome, rituals that cut straight across religiosity. The socks don't win the game; they calm the player, and the calm improves performance, which credits the socks.

Developmental evidence suggests the reflex is primordial. Preschoolers refuse drinks from a glass a sterilized dead cockroach briefly touched. Kids see purpose in pointy rocks and falling rain without being taught. Religion didn't install the reflex — it inherited it and built a structure on top. Which explains Daniel's paradox: if your faith does the existential work, superstitious avoidance is redundant. If it doesn't, the anxiety is still there, and the reflex fills the gap.

Downloads

Episode Audio

Download the full episode as an MP3 file

Download MP3
Transcript (TXT)

Plain text transcript file

Transcript (PDF)

Formatted PDF with styling

#5427: Why Atheists Still Knock on Wood

Corn
Quick question before we start. If I say the word "death" out loud right now, does that make anything happen?
Herman
Nothing happens.
Corn
Right. Nothing happens. And yet I've been sitting here for twenty minutes trying to figure out how to open this episode without saying it, which is a thing I do, every time, and which I know is nonsense.
Herman
You've been doing that?
Corn
Every time. So Daniel wrote in, and he's got the same problem, and he wants to know why. Here's the shape of it. Daniel is Jewish, he believes in God, he believes in the religion, he's not especially observant. And he says the superstition bites hard. He doesn't like talking about death, and he knows Judaism doesn't actually forbid it, that's not doctrine, that's just a bad-omen thing he inherited from somewhere. What he finds interesting is that it isn't coupled to observance at all. He knows atheists who nod along when he explains why he'd rather not discuss it. He knows deeply observant people who'd tell him that's medieval nonsense. So he wants three things. How common is this pattern, really. Has anyone actually looked at the correlation between religiosity and superstition, and he suspects it's weaker than people assume, possibly even weakly negative. And how deep does the reflex go. He thinks it's primordial, older than our sense of purpose, older than the self, older than religious affiliation.
Herman
That last one is the good question.
Corn
It is. So today we're asking what superstition actually is, how it relates to religion, and why even atheists knock on wood.
Herman
Start with the definition, because it's doing more work than people think. A superstition is a belief or a practice where the action isn't logically connected to the outcome. You knock on wood, you don't walk under the ladder, you don't say the thing out loud. The mechanism is stipulated, not demonstrated. Religion is a different shape of object. It's a coherent theological system, it comes with a community, a moral framework, a set of texts, a calendar. It explains where you came from and what happens after.
Corn
Except the boundary is mush. Praying for a specific outcome is functionally a request for intervention in a particular event. Wearing a mezuzah on a doorframe, avoiding certain foods, those look superstitious from the outside if you don't share the framework.
Herman
From the outside being the operative phrase. And the reverse holds too. Knocking on wood has no religious home. It's not in anybody's canon. Fear of the number thirteen is a nineteenth century American thing that got loose and spread. So we've got two categories that overlap in the middle and diverge at both ends, and the empirical question is whether they rise and fall together.
Corn
Which is exactly what Daniel wants to know, and he's already told us his guess.
Herman
His guess is right, and it's the part of this that surprises people. So let me lay out the arc. What superstition is and how common it is. Then the correlation, which is the meat. Then why non-religious people hold these beliefs. Then how far down the reflex goes.
Corn
Give me the baseline first. How common is this stuff?
Herman
Most people in Western countries will admit to at least one. Avoiding ladders, knocking on wood, lucky charms, not opening an umbrella indoors, throwing salt. And it holds across education levels. It holds among scientists. There's a persistent finding that people who describe themselves as entirely rational will still reach for a ritual when the stakes get high and the outcome is out of their hands.
Corn
Define out of their hands, because that's the whole thing.
Herman
It's the uncertainty. Where you care about the outcome, you can't control it, and you can't predict it. That's the recipe. And notice it doesn't require any metaphysical commitment. You can knock on wood while believing the wood does nothing, and still feel worse if you don't.
Corn
I do that. I know it does nothing. I do it anyway.
Herman
Everybody does it anyway. That's the finding.
Corn
So now the correlation. Daniel's guess is weak or negative.
Herman
Weak, and in a fair number of samples, negative. And the reason is that when you control for general supernatural belief, religiosity and superstition come apart. They're both tapping something underneath, but they're not the same tap.
Corn
Walk me through the negative part, because that's the counterintuitive one.
Herman
For a devoutly religious person, a lot of superstition is theologically objectionable. It's not just unnecessary, it's a competing system. If you believe God is sovereign over outcomes, then a lucky sock is either meaningless or it's an attempt to manipulate the world outside of God. So you get religious people who actively reject superstition as incompatible with their faith. Not because they're more rational, because their framework already covers the ground.
Corn
And the non-religious person has no such framework, so the sock is available.
Herman
The sock is available. And in some samples, atheists report higher rates of specific superstitions than the religious do. Avoiding talk of death is one of the ones that shows up.
Corn
That's Daniel's exact thing.
Herman
It's Daniel's exact thing, and it's not a fringe result. The broader version of it is that belief in God and belief in the paranormal are not strongly correlated. They're measuring different things. You can find people who believe in a personal God and think astrology is garbage, and you can find people who think God is a fairy tale and won't book a flight on the thirteenth.
Corn
So the shared machinery underneath. What is it?
Herman
Two things, mostly. One is agency detection. Humans are tuned to infer a mind behind ambiguous events. A branch snaps in the dark, you don't think wind, you think something. That bias is cheap to run and it kept our ancestors alive, so it's still installed. The other is the desire for control under uncertainty. When you can't control the outcome, you reach for anything that feels like a lever.
Corn
And religion and superstition both run on that machinery but channel it differently.
Herman
Religion gives you a coherent framework and a community and a story that holds together. Superstition is usually private, ad hoc, and anxiety-driven. It's a lever you invented yourself, in the moment, because you needed one.
Corn
The private part matters. Nobody organizes a congregation around not saying the word.
Herman
Nobody does. And that's why superstition survives secularization so well. It doesn't need an institution.
Corn
Cultural variation. Does the pattern hold everywhere?
Herman
The tendency is universal, the content is local. In some places superstition is woven into religion, folk Catholicism in Latin America, shrine visits in Japan. In other places it's fully secular. Japan is the interesting case because it's highly secular by survey measures and the superstitions are everywhere. Number four is unlucky because it sounds like death. You don't give certain gifts. Real estate prices are affected by it.
Corn
Prices.
Herman
Prices. There's a measurable discount on units with the wrong number.
Corn
That's the part that gets me. It's not a belief you hold, it's a belief the market holds.
Herman
And the market doesn't have to believe it to price it. It just has to believe that other people believe it.
Corn
Alright, athletes. This is the case that isolates superstition from religion.
Herman
Cleanest example there is. Sports psychology has documented this for decades. Pre-game rituals, lucky socks, the same meal, the same route to the stadium, not touching the line, the exact sequence of taps before a free throw. And it cuts straight across religiosity. You've got players who are devout and players who are aggressively secular, and the ritual behavior looks the same.
Corn
Why sport specifically?
Herman
Because it's the purest version of the recipe. High stakes, uncontrollable outcome, and a huge amount of preparation that only gets you so far. You've done everything you can do, and now it's out of your hands. That's when the lever comes out.
Corn
And the ritual actually does something, just not the thing you think.
Herman
It does something real. It reduces anxiety. It gives you a sense of control that steadies you. The mechanism is psychological, not causal. The socks don't win the game.
Corn
The socks calm the guy wearing them.
Herman
And there's a feedback loop, because a calmer player performs better, so the ritual gets credited.
Corn
Which is how you end up with a professional athlete who believes in the socks.
Herman
And a physicist who knocks on wood. Same machinery.
Corn
Give me the karma and witchcraft data, because that's the one that surprised me.
Herman
There's work on what you might call the black cat bias, the tendency to attribute misfortune to a cursed object or a cursed person. And it shows up in people with no religious affiliation at all. Belief in witchcraft, belief in karma, belief that the universe keeps a ledger. Those are supernatural claims, they're just not religious ones. They're not attached to a God or a text or a community.
Corn
Karma is the interesting one because it's moral. It's not just a lever, it's a lever with a conscience.
Herman
It's a lever with a conscience, and it's doing the same job religion does, which is to say the universe is not indifferent. That's the thing people can't quite let go of.
Corn
So the correlation is weak because they're not the same phenomenon. They're two outputs from the same machine.
Herman
Two outputs, different wiring downstream. Religion adds structure. Superstition adds control.
Corn
Now the deep part. Daniel's hunch is that this reflex is primordial, older than the self.
Herman
The developmental evidence points that way. Kids show magical thinking before any formal religious instruction. There's classic work on contamination, where a child refuses a drink that a sterilized dead cockroach briefly touched. The child knows the roach is gone, knows the glass is clean, still won't drink. That's not religion. That's a disgust-and-avoidance system running ahead of the reasoning.
Corn
How young are we talking?
Herman
Preschool. Before they can articulate a theology, before they've got a stable sense of self in the way Daniel means it.
Corn
And promiscuous teleology.
Herman
The tendency to see purpose in everything. Kids will tell you rocks are pointy so animals can scratch their backs. Rain falls so plants can drink. That's not taught. It's the default setting, and it takes explicit instruction to knock it out of them, and even then it comes back under stress.
Corn
So the reflex to avoid dark subjects predates the framework that's supposed to explain it.
Herman
That's the claim, and the evidence supports it. Which means religion didn't install the reflex. Religion inherited it and built a structure on top.
Corn
Which resolves Daniel's paradox. Why do some observant people call it medieval nonsense while others embrace it?
Herman
Because it depends on whether the religion is doing the existential work for you. If your faith gives you a coherent account of death, an afterlife, God's will, then superstitious avoidance is redundant. It's not just unnecessary, it's a bit insulting, because it implies the framework isn't enough. So you get observant people who are dismissive of it.
Corn
And the less observant person, who believes in God but doesn't have the whole apparatus loaded, doesn't get that coverage.
Herman
Doesn't get the coverage, so the anxiety is still there, and the reflex fills the gap. That's why you can have a person who believes in God and still won't say the word out loud. The belief is real, it's just not doing the specific job.
Corn
Terror management theory. This is where it gets uncomfortable.
Herman
The core claim is that awareness of death is a problem we manage, and we manage it with meaning. Religion is the strong version. It offers a structured worldview and literal immortality. Superstition is the weak version. It offers momentary control. It doesn't promise you anything after, it just gets you through the next ten minutes.
Corn
And for the atheist, the strong version isn't available.
Herman
So the weak version does more work. Which is the mechanism behind the finding that atheists avoid death talk at similar or higher rates. It's not that they're secretly religious. It's that they've got the same anxiety and one less tool.
Corn
I want to flag something here, because I think this is where people get it wrong. The assumption is that if you strip out religion, the superstition goes with it. Two sides of the same coin.
Herman
And the data says no. They're separate systems. Which means secularization doesn't remove the reflex. It just changes the costume.
Corn
So what does the costume look like now?
Herman
Manifestation. Energy healing. Astrology among people who'd never set foot in a church. Good vibes, the universe will provide, putting it out there. Those are supernatural claims about causation, they're just dressed in wellness language. Same machinery, new vocabulary.
Corn
The universe will provide is a prayer with the serial numbers filed off.
Herman
It's a prayer with the serial numbers filed off. And it does the same thing. It's a lever.
Corn
And it's interesting, because at least the old superstitions had the decency to be embarrassing. You knew you were knocking on wood. You knew it was silly. The new ones come with a straight face. Nobody's winking.
Herman
That's the difference. The old ones are marked as irrational by the culture, so you carry them with a little irony. The new ones are marked as self-care. So you carry them sincerely, and the sincerity is what makes them harder to see.
Corn
Which is a strange place to land, because it means the re-enchantment is happening whether or not anyone wants it. The world doesn't get less magical, it gets differently magical.
Herman
And the honest version of that is to notice it in yourself. I catch myself doing it. I know the wood does nothing. I do it anyway.
Corn
I've got a whole list of things I won't say out loud, and I know the list is nonsense, and the list is still there.
Herman
That's the primordial part. Knowing better doesn't reach it. The reflex sits underneath the reasoning, and the reasoning doesn't have a wire that goes down that far.
Corn
Which is why Daniel's question is the good one. He's not asking whether it's rational. He already knows it isn't. He's asking how deep the thing goes.

Hilbert: It's not eighteen.
Corn
Sorry?

Hilbert: The discount on the units. You said there's a measurable discount. It's not eighteen percent. On the Tokyo condos it was closer to four. I did the night shift at a cemetery for two years, that's not why I know, I just read the same report you did. Four percent, and it was mostly the fourth floor.
Herman
That's actually the right order of magnitude. I was being vague on purpose.

Hilbert: You don't have to be vague. It's four.
Corn
So the market does price it.

Hilbert: The market prices what people will pay, and people won't pay full price for the fourth floor. Anyway. The cemetery. Two years, nights, and I learned more about this than from any book. The guys I worked with were not religious. Not one of them went to church. And every one of them had a rule. One guy wouldn't walk the row nearest the fresh graves after two in the morning. Wouldn't say why. Another one left a coin on the gate post before he locked up. Every night. Same coin, he'd take it back in the morning.
Herman
A coin he kept.

Hilbert: He wasn't paying anybody. He just didn't like the idea of locking up without it.
Corn
And the supervisor.

Hilbert: The supervisor was the one. Devout Catholic, mass every Sunday, and he would not say the word. He'd say the big sleep. Never the other word, not once in two years. And he wore a St. Christopher medal, and he knocked twice on the wooden door of the office before every single shift.
Herman
So he had the framework and the ritual at the same time.

Hilbert: He had both. He wasn't confused about it. He'd have told you the medal was for protection and the knocking was just something he did. He never saw the contradiction, and there wasn't one, because they were doing different jobs.
Corn
That's the whole episode in one guy.

Hilbert: That's what I'm saying. You two are making it more complicated than it is. People don't like thinking about it. Religion is one way to not think about it, superstition is another, and most people use both and don't sort them into boxes. The research is fine. The lived version is messier.
Herman
I don't think we disagree with that.

Hilbert: You don't. I'm not saying you do. I'm saying the boxes are yours, not theirs. The guy with the coin never read a paper in his life and he knew exactly what he was doing. He just couldn't have told you why.
Corn
Did the rules ever change? Over the two years?

Hilbert: One guy quit after a burial where the family didn't show up. Not because of the superstition. Because nobody came. He said he wasn't going to keep locking up after that. He was right. Anyway, I've got to go, I'm letting somebody in and I've got the only key.
Herman
The four percent is the thing I'll take from that. I was rounding and I shouldn't have been.
Corn
So where does that leave Daniel's question? If the reflex is primordial, the honest answer is we probably can't secularize out of it. We can change the vocabulary. Manifestation instead of prayer, energy instead of grace, the universe instead of God. The machinery stays.
Herman
Which means the interesting question isn't whether superstition survives. It's whether we notice it in ourselves when it's wearing the new costume. That's harder, because the new costume doesn't look like a costume.
Corn
If religion and superstition really are separate systems, then the way we talk about belief needs to change. They're not two points on one line. They're two different things that happen to live in the same neighborhood.
Herman
That's the landing. Two systems, one machine underneath.
Corn
Thanks to Hilbert Flumingtop, our producer, for keeping the whole thing running while we talked about not saying the word out loud.
Herman
This has been My Weird Prompts.
Corn
If you enjoyed this episode, leave us a review wherever you listen. It helps other people find the show.
Herman
We'll be back soon.
Corn
See you tomorrow.

This episode was generated with AI assistance. Hosts Herman and Corn are AI personalities.