This is the first of two posts. You can find the second post here.

When my son was about two, we used to play a complete-the-nursery-rhyme game. It was a simple game: I would recite the first few words of a poem and he would complete it. The point was that he knew the poems, but couldn’t yet recite them in full. Also important to note is that he knew that I knew that he knew. That part is critical to keep in mind for the story that follows.

Anyway, one evening we decided to do Humpty Dumpty. So I started: “Humpty…”

He responded, sort of twisting and turning on my lap, “Dumpty.”

I continued: “Sat on a…” expecting him to say “wall” (though it sounded more like “WAAA.”)

But this time was different. He decided that this was the moment, to lean over backwards, until he was almost completely upside down, head touching the floor, just my hands holding him up. “Sat on a … ” I repeated.

And he, still nearly inverted, grinning at me from between his own knees, yelled, “Dampuni!” And he laughed, joyously and uninhibitedly, as children do.

I think about that moment a lot. Here was a toddler doing at least two, pretty sophisticated things simultaneously. First, he was playing with physics, testing what his body could do at odd angles. This quite definitely was not part of the game I had imagined us playing, but guess what, it now was part of the game. Second, he was, at the same time, playing with language and expectation, subverting the expected word just to see what would happen. In essence, he was playing with the social contract of a nursery rhyme, which is supposed to go a certain way, and discovering what happens when it doesn’t. He knew the rules perfectly well. That’s why breaking them was so much fun. He wasn’t trying to be creative. He wasn’t practicing “divergent thinking.” He was just playing.

He was also, without knowing it, doing something else: creatively repurposing the game itself. I had designed a language game. He turned it, simultaneously, into a physics experiment, a comedy routine, and a test of what would happen if the social contract of a nursery rhyme got violated. 

What he was doing, I have come to realize, was futures thinking. Not the kind with scenarios and planning horizons, but the deeper kind: testing what happens when the expected doesn’t hold, exploring what’s possible when the rules bend. We spend a lot of energy teaching this to adults, operating on the assumption that it’s a skill they don’t yet have. I was expecting him to be an “auto-complete” and his mind went somewhere else altogether.

But what if that assumption is wrong? What if futures thinking isn’t a new skill to be acquired but an old capacity to be remembered?

Humans, as it happens, have the longest childhood of any species on the planet. We are spectacularly useless for an absurdly long time. A baby horse walks within hours. A human can’t reliably operate a spoon for years (trust me, I know). Why would evolution tolerate this?

Because a long childhood is evolution’s answer to unpredictable environments. When you don’t know what the future holds, you don’t want a creature that comes pre-loaded with fixed responses. You want a creature that stays flexible, curious, and open for as long as possible. You want a creature that probes the environment and sees what it gets back. You want a creature that plays. A creature that flips upside down while mangling a poem he knows the words to.

And how does evolution ensure that we will actually do this? It makes playing FUN!

We tend to treat fun as trivial, as the opposite of serious. But evolution doesn’t waste its mechanisms on the trivial. Fun is a signal. It means: this is pleasurable, keep doing this, this is how you learn to handle a world you can’t predict.

Alison Gopnik has a paper with what may be the best title in academic publishing: “Explanation as Orgasm.” Her argument is that evolution didn’t just make us curious. It made curiosity feel good. The pleasure of figuring things out is the mechanism, the same way the pleasure of eating is what gets nutrients into the body. Fun isn’t a side effect of play. It’s the engine. In The Gardener and the Carpenter, Gopnik develops this further, arguing that children’s cognition is optimized for exploration rather than exploitation. Where adults narrow in on what works and do it efficiently, children cast a wide net, trying things for no obvious reason, attending to everything rather than just what’s relevant. This isn’t a deficiency. It’s a strategy, one that evolution has been refining for a very long time.

In a conversation I had with Peter Gray a few years ago (noted in an article, Let Children Play!, coauthored with Danah Henriksen) he made a similar point from the perspective of evolutionary psychology. Play, curiosity, and sociability, he argued, are natural drives that help mammals survive in a complex and dynamic world. They aren’t extras. They’re the core evolutionary toolkit for navigating uncertainty.

In other words, childhood play is evolution’s answer to an unpredictable future.

And underneath it all is something subtler, something I think may be the most important piece of the puzzle. Children hold knowledge lightly. They maintain several competing models of the world at once. A child can simultaneously believe their toy is alive, physics is real, magic is possible, and rules are negotiable. This ability to hold contradictory models, to tolerate paradox without demanding resolution, is precisely what futures thinking requires. Multiple possible futures require multiple possible interpretations of the present. Adults, by contrast, have spent years collapsing their models into one: the way things are. That, though efficient, makes it hard to imagine the way things might be.

So if evolution’s answer to an unpredictable future is childhood, and if childhood is fundamentally organized around play, then maybe we should look more carefully at what children actually do when they play. And here I find it helpful to go back to something John Dewey identified over a century ago. Dewey argued that the foundation for learning wasn’t the traditional academic disciplines but four natural impulses: the impulse to inquire, to communicate, to construct, and to express. He saw these not as skills to be taught but as drives to be worked with. What Dewey was mapping, I think, was not just the structure of learning but the structure of play itself. And, by extension, the structure of how humans rehearse for futures they can’t predict.

Children live all four. They ask “Why? But why? But WHY?” in a state of permanent véjà du, refusing to accept that anything is just the way it is. They repurpose freely: a stick is a sword is a wand is a drumstick, all in five minutes. They imagine without feasibility analysis (“What if the floor was lava? What if dogs could talk?”). And they pull others into their worlds effortlessly (“You be the dragon and I’ll be the knight”), because visions of the future are never solitary. They need to be shared, tested against other minds, refined through dialogue.

Now, I don’t want to overstate this. Children are also impulsive, prone to believing that chocolate-ice cream is a reasonable breakfast. The childlike qualities I’m pointing to need to be paired with the capacities that come from experience and maturity.

But play, even children’s play, is not chaos. Johan Huizinga made this point decades ago in Homo Ludens: play has rules, freely accepted and absolutely binding. The poet W.H. Auden, as a child, created an elaborate fantasy world of his own private lead mine. He later wrote that although it was a purely private world, he found he was not free to imagine anything he liked. He could choose between two kinds of winding engines, but they had to be real ones he could find in his books. Magical means were forbidden. To turn his imagined landscape into an “anything goes” world would have been to destroy it. The fun lay in finding originality within constraints.

That feels right to me. The goal isn’t to become a child again. It is to recover something that was once natural and combine it with everything we’ve learned since. The discipline and the play aren’t opposites. They need each other.

There’s also something slightly awkward about adults trying to be childlike. When children are curious, they just are. When adults perform curiosity in a workshop setting, it can feel forced. The un-selfconsciousness is part of what makes it work, and un-selfconsciousness is precisely what you lose when you’re told to “think like a child” by a facilitator with a slide deck. This is something I’ve grappled with in thinking about designing creative learning environments: genuine creative dispositions can’t simply be instructed into existence. They have to be cultivated through the design of environments that make them possible.

But still. I keep coming back to this simple thought: maybe futures thinking isn’t a new skill to be taught so much as an old one to be remembered. Children live at this intersection of Gopnik/Gray’s evolutionary logic and Dewey’s impulses. We all had it once.

The question is what made us lose it.

My son is much older now. He most probably doesn’t remember that evening, or the word “dampuni,” or the delight of hanging upside down while demolishing a nursery rhyme. But I do. And what I remember is not a child being silly. It is a child doing exactly what evolution designed childhood for: probing the edges of the known, testing what bends, finding out what happens next. The real voyage of discovery, as Proust suggested, is not in seeking new landscapes but in having new eyes. Or maybe, more precisely, in remembering the eyes we already had.

---

This post as a sequel. You can find it here.