Daniel's been chewing on something that's been bugging him since we talked about intoxicants. He points out that fermentation looks, on paper, almost impossible to discover by accident. Something sweet rots, someone drinks the liquid, fine. But then someone has to figure out there's a replicable process in there and systematize it. And then he zooms out. Coffee. A non-nutritious seed, roasted, ground, brewed. In a world where everyone's supposedly optimizing for calories, the cost-benefit ratio on that experiment is absurd. So his question is: how on earth did humans discover these wildly unlikely ways to prepare raw ingredients, centuries before anyone knew any chemistry? Were early humans just unusually dogged experimenters, or is something else going on?
The coffee one really gets me. Because you have to sit with how many steps delivered zero reward before the first cup.
Right. Roasting a green coffee bean doesn't smell good.
It smells like burning plant matter. Which, if you're a hunter-gatherer, is the smell of a mistake. And grinding it is just labor. And then you pour hot water on the result and... you have to decide to drink that. The first cup of coffee was an act of faith performed by someone who had no reason to have any.
So today we're going to follow the trail of unlikely discoveries, from the first beer to the first cup of coffee, and ask what they tell us about human nature.
The core paradox here is that a lot of foundational food technologies require multiple steps that individually provide no benefit. Fermentation, coffee, chocolate, cheese. Each one involves invisible biological or chemical transformations that early humans couldn't have understood. And yet they harnessed them.
Two ways to explain this. One is the accidental genius model. A series of lucky accidents that observant humans noticed and systematized. The other is what you might call the deep drive model. Humans have an innate, non-caloric exploratory urge that pushed them to experiment with inedible things, and culture preserved the wins.
And this isn't just a food history question. It's about whether human curiosity is a side effect of calorie-seeking or a primary drive in its own right. Whether we explore because we're hungry, or whether we explore because we're us.
So let's start with the oldest of these unlikely discoveries. Fermentation. And look at what the archaeological record actually tells us.
There's a researcher named Patrick McGovern at the University of Pennsylvania Museum. He's sometimes called the beer archaeologist, which is a title I deeply respect. McGovern has analyzed pottery residues from a site called Jiahu in China, dating back nine thousand years. And what he found is a mixed fermented beverage. Rice, honey, and fruit.
Nine thousand years.
Which predates systematic bread-making by millennia. This is the beer before bread hypothesis. The argument is that alcohol production wasn't a byproduct of agriculture. It may have been one of the reasons agriculture got going in the first place.
So the sequence isn't we grew grain, then we figured out what to do with it. It's we wanted the drink, so we figured out how to grow the grain.
That's McGovern's argument, and the residue evidence supports the timeline. But the question Daniel's really asking is about the discovery mechanism. How do you get from zero to that Jiahu beverage?
Walk me through the actual steps.
Wild yeasts are everywhere. They settle on sweet fruits naturally. If you leave a container of something sweet out in the right climate, you get a mildly alcoholic liquid. That's the accident part, and it's plausible. Someone leaves some fruit mash in a gourd, comes back a few days later, it smells different, they taste it.
And they feel something.
That's the key. McGovern argues that the psychoactive effect was the initial reward. Not calories. The buzz. Because the caloric gain from early fermentation was minimal. You're not doing this for nutrition. You're doing it because it does something to your brain.
But here's where the accident model runs into trouble. Drinking spoiled liquid once is one thing. Connecting cause to effect is another. Sweet plus time plus container equals buzz. That's a three-variable equation that nobody wrote down.
And you have to hold those variables in memory across days or weeks. The sweet thing went in. Time passed. Now there's this. The cognitive leap from this rotten mush tastes interesting to I can replicate this by mixing grain, water, and waiting requires observation, memory, and intentionality.
It implies early humans were already thinking in terms of processes, not just one-off events.
And McGovern's point is that the psychoactive reward was strong enough to motivate that systematization. You don't go through all those steps for a slightly tangy porridge. You do it because the result makes you feel different.
The buzz is the bridge between accident and technology.
And once you've got the basic loop, you iterate. Different grains, different fruits, different containers, different waiting times. The people who got good at it passed the knowledge down. And the ones who got it wrong...
Got the wrong kind of famous.
There's a broader pattern here. The same logic applies to cheese. Someone stores milk in a container made from an animal stomach. The residual rennet curdles the milk. They open it expecting liquid and find solids. Most people, I think, would throw that out.
Most people now would throw that out.
But someone tasted it. And then someone figured out that the stomach container was the variable. And then someone started deliberately adding rennet. Each step requires overriding the normal response to unexpected food, which is disgust.
Chocolate is even worse. The cacao pod has a sweet pulp around the seeds, which is nice. But the seeds themselves are bitter and astringent. To get to chocolate, you ferment the seeds, dry them, roast them, grind them. The fermentation step alone takes days and the beans look and smell like something that's gone wrong.
And bread. Grinding hard seeds into powder, mixing with water to make paste, then leaving it out until it bubbles, then applying heat. The bubbling step is fermentation again. If you don't know what yeast is, dough rising looks like it's been colonized by something.
Which it has.
Which it has. But every one of these processes involves an invisible transformation. Early humans couldn't see yeast or bacteria. They couldn't see the Maillard reaction browning the bread crust or the coffee bean. They were working blind.
So we've got a pattern. Psychoactive reward or strong sensory payoff driving systematization. Fermentation gives you the buzz. Cheese gives you a way to preserve milk. Chocolate gives you... well, chocolate.
But coffee is harder. Because the initial encounter with a raw coffee bean gives you nothing.
Let's get into that. Fermentation gives us one pattern. Coffee gives us another, and it might be even stranger.
The canonical story about coffee's discovery is Kaldi the goat herder. Ninth century Ethiopia. He notices his goats acting energetic after eating red coffee cherries. He tells a monk, the monk makes a drink from the beans, and coffee is born.
That's the story.
It's almost certainly a myth. There's no historical evidence for it. The earliest credible evidence of coffee consumption comes from Sufi monasteries in Yemen in the fifteenth century, where it was used to stay awake for night prayers.
Fifteenth century. That's already a processed beverage. Someone had to figure out the whole chain before that.
And the chain is genuinely strange. Raw coffee cherries are mildly sweet but not especially nutritious. The seed inside is hard, bitter, and indigestible. Roasting transforms the chemistry. The Maillard reaction creates hundreds of aromatic compounds that don't exist in the raw bean. But the first person to roast a coffee bean had no reason to think the result would be edible.
Let alone desirable.
And then you grind it. And then you brew it. And at no point before the first sip does anything about this process signal this is going to work.
So how do you get there?
The answer, I think, is that the roasting and brewing came late. They're optimizations of something people already knew. Before anyone brewed coffee, Ethiopians chewed the leaves and cherries for stimulation. The coffee plant contains caffeine in its fruit and leaves, not just the seed. You can get the psychoactive effect by just... eating the plant.
So the initial discovery wasn't roast, grind, brew. It was chew this leaf, feel awake.
That's the stimulant-first model. Across cultures, humans discovered psychoactive plants by chewing raw leaves or fruits. Coffee, tea, coca, khat, betel nut. The alertness effect is noticeable quickly. You don't need processing. You just need to put the thing in your mouth.
And then what?
Then you want to preserve it, or concentrate it, or make it portable. The coffee cherry is seasonal. If you dry the seeds, you can store them. If you roast them, they last even longer. If you grind and brew them, you get a faster, stronger dose. The elaborate preparation method isn't the starting point. It's an optimization of an already-known effect.
That reframes the whole question. The unlikely preparation method isn't what was discovered. The effect was discovered. The method is just engineering.
And engineering is cumulative. Nobody sat down and invented coffee from scratch. Someone chewed the leaves. Someone else tried the cherries. Someone figured out the seeds could be dried. Someone else tried roasting them. Each step is small, and each step has a reward at the end.
But some of those steps still seem irrational in the moment. Drying the seeds, fine, preservation. Roasting them? You're applying fire to something that's already hard and inedible. What's the thought process there?
I don't know that we can reconstruct the thought process. But I can imagine a scenario. You've got dried coffee seeds stored somewhere. A fire gets too close. The seeds get roasted. They smell different. Not good, necessarily, but different. And you're a person who already knows these seeds contain something valuable. So you investigate.
The known effect lowers the bar for experimentation.
Right. If you know the seed has caffeine, you're willing to try things with it that you wouldn't try with a random bitter seed you found on the ground. The cost-benefit ratio Daniel's talking about only looks impossible if you assume the experimenter had no prior knowledge.
But the person who first chewed the leaf had no prior knowledge.
That's true. And that's where the broader pattern comes in. Foraging humans taste everything. They have to. You can't know which plants are edible without trying them. And the ones that produce a noticeable physiological effect get remembered fast.
Bitter seed makes your heart race. Sweet leaf makes you alert. Mushy fruit makes you dizzy and happy.
Those are salient experiences. They stand out against the background of normal eating. And once an effect is known, the plant becomes valuable. People trade it, cultivate it, experiment with preparation methods.
So the pattern is three stages. Initial exposure through broad foraging. Taste everything. Identification of a non-caloric effect. Stimulation, intoxication, pain relief. And then cultural elaboration to improve delivery.
Which is less dogged pursuit of cooking processes and more opportunistic optimization of psychoactive rewards.
The thing I keep coming back to is that none of this requires anyone to understand what's happening chemically. You don't need to know what yeast is to notice that waiting makes the drink better. You don't need to know what caffeine is to notice that chewing this leaf makes you less tired.
And that's the part that's easy to forget from a modern perspective. We think of these discoveries as requiring theory. They don't. They require attention. Someone paying close enough attention to say, when I leave the grain paste out for two days instead of one, the result is different, and I prefer it.
Pattern recognition without mechanism.
And humans are extraordinarily good at that. We're pattern-matching engines. We don't need to know why something works to know that it works.
But we do need a reason to keep trying when the first ten attempts fail.
And that's where the psychoactive reward comes back in. If the thing you're trying to produce makes you feel good, you'll tolerate a lot of failed batches. If it just tastes slightly better than the raw version, you probably won't bother.
So the answer to Daniel's question, were early humans just unusually dogged experimenters, is... yes, but with a specific motivation. They weren't dogged about cooking processes in general. They were dogged about things that altered their mental state.
And the things that didn't, they probably tried once and abandoned. We don't hear about those because they didn't become technologies.
Survivorship bias in the archaeological record.
For every coffee, there were probably a hundred plants that someone tried roasting and brewing and the result was just bitter brown water with no payoff. Those experiments died with the experimenter.
So the cost-benefit ratio Daniel's worried about is real, but it's being applied to the wrong set of experiments. The ones that succeeded are the ones where the benefit was strong enough to justify the cost. The ones where the benefit was weak, we never found out about.
And the benefit doesn't have to be caloric. That's the key insight. Caloric optimization is a good model for understanding what animals eat. But humans have been doing things for non-caloric reasons for a very long time.
The buzz is the calories of the mind.
I'm going to pretend you didn't say that.
You're going to use it in the next episode.
But the point stands. If you model early humans as pure calorie-maximizers, coffee makes no sense. If you model them as experience-maximizers, it makes perfect sense.
There's a question hanging over all of this that we haven't touched. What does this mean for how we think about innovation now?
Well, we're trying to engineer novel foods right now. Lab-grown meat, precision fermentation. And the question is whether the same non-caloric drives will shape adoption.
Or whether modern efficiency thinking kills the experimentation that led to coffee and beer.
If you optimize purely for nutrition and cost, you don't get coffee. You get nutrient paste. And nobody gets excited about nutrient paste.
The psychoactive angle is interesting here too. A lot of the early novel food adoption is being driven by environmental and ethical concerns. Which are... not caloric.
They're values. They're experiences of a different kind. The feeling of doing the right thing. Which is a kind of psychoactive reward, if you want to stretch the definition.
I think we're stretching it.
Probably. But the broader point is that food has never been just about calories. It's been about what it does to you, how it makes you feel, what it says about who you are. And that's been true since Jiahu.
Before we go further, Hilbert has been making a face behind the glass for the last ten minutes, and I think he wants to say something.
Go ahead.
Hilbert: A five-gallon glass carboy. Northern Brewer brand. Forty-seven dollars in 1998.
...Alright.
Hilbert: I worked quality control at a craft brewery in Portland for six months. Late nineties. My job was to stare at fermentation tanks and figure out why batch seven smelled like nail polish remover. I was bad at it and I hated it. But I learned one thing. You cannot control wild yeast. We had temperature gauges and pH meters and sanitized everything with chemicals that would peel the skin off your hands, and we still lost a third of our batches. A third. With modern equipment and a written protocol and a guy whose entire job was to prevent exactly that.
What was the failure mode?
Hilbert: Acetobacter. Turns alcohol into vinegar. Gets in through the air. You open a carboy for thirty seconds and it's in there. And you don't know for two weeks. You just open it up expecting a nice sour ale and you get salad dressing. So when you two talk about someone nine thousand years ago looking at a pot of fermented mush and thinking, I should do that again on purpose...
You're not buying it.
Hilbert: I'm saying the model's wrong. You're describing it like they were proto-chemists running experiments. They weren't. They were bored.
Bored.
Hilbert: My brother-in-law, Dale. He lives in a cabin in eastern Oregon. No internet. He makes things. He made a chair out of skis once. Not because he needed a chair. He had chairs. He was bored. And when you're bored and you have a full stomach and a long evening, you try things. You chew a leaf. You roast a seed. You leave the fruit mash out an extra day. Most of it fails. You throw it out. But the things that make you feel good, you do again.
So the cost-benefit ratio only looks bad if you assume early humans were optimizing for efficiency.
Hilbert: If they were optimizing for novelty because survival was already handled, then roasting a bitter seed makes perfect sense. It's something to do. Dale once tried to make wine out of dandelions. Took him three weeks. Tasted like grass clippings. He didn't care. He wasn't doing it for the wine.
He was doing it for the doing.
Hilbert: Modern people have this idea that ancient humans were all desperately scrabbling for calories every waking moment. But hunter-gatherers in decent environments had downtime. A lot of it. You can only eat so much. You can only sleep so much. The rest of the time, you're sitting around. And sitting around is where fermentation was discovered. By someone who had nothing else to do and a container of fruit that got away from them.
The boredom hypothesis.
Hilbert: Dale is not a genius. Dale is a man with time and a willingness to drink things that might be poison. I think that's the whole story. You don't need a deep drive or an accidental genius. You need leisure and a high tolerance for failure.
The thing about Dale that interests me is the chair made of skis. Because that's not food. That's not even useful. It's just... a thing he made.
Hilbert: He's got a whole shed full of them. He calls them his projects. His wife calls them reasons to buy insurance.
So the same impulse that gave us coffee also gives us furniture made out of sporting equipment.
Hilbert: I'm not saying it's dignified. I'm saying it's human.
There's something here about the relationship between leisure and innovation that I think gets underplayed. We talk about necessity being the mother of invention. But a lot of the transformative discoveries weren't necessary. They were optional. You can live a full life without coffee.
Hilbert: You can. I wouldn't recommend it.
Hilbert's point about downtime is interesting because it flips the framing. Daniel asked whether early humans were unusually dogged experimenters. Hilbert's saying they weren't dogged. They were just... unoccupied.
And the two things look similar from a distance. Someone who tries the same process fifty times could be persistent or could just have a lot of free time.
The output's the same either way.
It is. But the motivation matters for how we understand ourselves. If these discoveries came from a deep exploratory drive, that says something about human nature. If they came from boredom, that says something else.
What does it say?
I think it says we're easily entertained by our own experiments. Which is not a bad thing. A lot of what we do on this show is basically the same impulse. We take a question and we poke at it for an hour because it's more interesting than not poking at it.
We're Dale with microphones.
I was going to say we're the Jiahu brewers with a budget, but sure.
Hilbert: Dale did start a podcast once. Lasted four episodes. He recorded them in the shed. The audio was terrible. You could hear skis falling over in the background.
What was it about?
Hilbert: Chairs.
Of course it was.
Hilbert: He had opinions. I don't share them. He's not to be trusted on chairs.
The thing I keep coming back to is the role of the psychoactive reward. Even if boredom is the spark, the buzz is what keeps the fire going. Dale's dandelion wine failed because dandelion wine tastes like grass. If it had gotten him pleasantly drunk, he'd still be making it.
Hilbert: Probably. He's not a man who learns from failure in the traditional sense.
Maybe the synthesis is: boredom provides the experiments, psychoactive rewards filter the winners, and culture preserves the recipes. You don't need anyone to be a genius. You just need enough people with enough time trying enough things that something eventually works.
Then the thing that works spreads because it makes people feel good.
Coffee spread from Ethiopia to Yemen to the Ottoman Empire to Europe in a few centuries. That's not because anyone was calorie-deficient. It's because caffeine is a powerful drug and people wanted it.
The first coffeehouses in Europe were called penny universities because for the price of a cup you could sit and listen to intellectuals argue. The drink was the admission fee for a cognitive experience.
Which is another non-caloric reward. Stimulation plus social connection. The coffee itself was almost secondary.
We've got three layers now. Boredom starts the experiments. Psychoactive effects select the winners. And then social and cultural rewards drive adoption and elaboration.
The original cost-benefit question Daniel raised only looks at the first layer in isolation. If you ignore the psychoactive and social rewards, the behavior looks irrational. If you include them, it looks obvious.
The question that leaves me with is whether we're still doing this. Whether modern innovation follows the same pattern.
I think it does, but we're embarrassed to admit it. We tell stories about necessity and efficiency and rational problem-solving. And then someone invents a social network and a billion people join because it makes them feel something.
The buzz is still the product.
The buzz is always the product. We just got better at dressing it up.
One open question before we wrap up. If boredom and psychoactive rewards drove these discoveries, what does that mean for how we think about human innovation? Are our most transformative technologies accidental byproducts of a drive to feel different, not better?
I think a lot of them are. And we're entering a period where we're engineering novel foods from scratch. Lab-grown meat, precision fermentation. The question is whether the same non-caloric drives will shape adoption, or whether modern efficiency thinking kills the experimentation that led to coffee and beer.
If you optimize purely for nutrition and cost, you get something nobody wants to eat.
If you optimize purely for experience, you get something nobody can afford. The trick is finding the space where the old impulses still have room to operate.
We'll see if we manage it.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop, who has apparently been holding onto that brewery story for twenty-eight years.
If you have a weird prompt you want us to tackle, email the show at show at my weird prompts dot com. We read every one, and yours might be the one that makes Hilbert tell another story about his misspent youth.
We'll be back soon.
