AI channels John Holt

=


I asked ChatGPT this, May 16, 2026:
What might John Holt have said about kids and computer use if he had lived long enough to see it?
ChatGPT's response:

I think John Holt would probably have been fascinated by computers and wary of adult control over them at the same time.

Not “computers are automatically good,” and not “screens are ruining childhood,” but more along the lines of:
“What are children actually doing with this tool, and what real needs or curiosities is it serving?”
That was very much his pattern.

By the late 1970s and early 1980s, before home internet and before modern gaming, Holt was already defending children’s right to follow interests that adults didn’t necessarily value. He trusted self-directed engagement far more than imposed curriculum. So I doubt he would have lined up comfortably with modern anti-screen panic.

He probably would have noticed several things quickly: All of those fit his existing philosophy very well.

I suspect he would have loved some early hacker/computer-club culture—the way kids learned from one another, traded discoveries, stayed up absorbed for hours, and developed competence because they wanted to.

He also distrusted coercion deeply. So I think he would have objected both to: He often argued that adults confuse their own anxieties with children’s needs. So if a parent said:
“My child spends all day on the computer!”
I can imagine Holt asking:
“Doing what?”
Because “computer use” is not one thing: those are socially and intellectually different activities sharing one machine.

And he was very good at separating the medium from the living experience.

At the same time, I don’t think he would have been naïve about commercial manipulation. Holt cared a lot about institutions shaping children into compliant consumers and performers. I think modern algorithmic attention-capture systems would have bothered him—not because they were electronic, but because they attempt to override autonomy and curiosity rather than support them.

So he might have distinguished between: One thing that feels especially Holt-ish to me is that he probably would have resisted simplistic categories like: He understood that deep learning often looks aimless from outside.

A child memorizing Pokémon statistics, building Minecraft contraptions, or participating in an online fandom might have made perfect sense to him if the child was alive with attention, connection, competence, and meaning.

And he would almost certainly have noticed what many unschoolers later noticed: that computers collapsed the old gatekeeping system.

A kid no longer needed: to pursue serious interests.

That would have seemed revolutionary to him.

I suspect he would also have appreciated your recurring point that curiosity “pulls learning in.” Computers are extraordinarily good at that when people are free to use them expansively rather than under constant evaluation.

He might have said something close to:
“Trust the learner more than the system.”

Because it was directed to me and ChatGPT "knows" my beliefs about learning, I was addressed with "I suspect he would also have appreciated your recurring point that curiosity 'pulls learning in.'"

A Second Opinion:

Karen James asked Claude the same question, the next day, and got this:

John Holt died in 1985, just as personal computers were beginning to appear in homes and schools, so he never wrote directly about the internet, tablets, or modern software. But his body of work — How Children Learn, How Children Fail, Instead of Education, Teach Your Own — gives us a pretty clear window into how he would have thought about it.

He would have been enthusiastic, but on the child's terms
Holt was fascinated by tools that let children pursue their own curiosity without needing an adult intermediary. He loved that kids could use typewriters, calculators, and musical instruments on their own schedules, at their own pace, driven by genuine interest. Computers — especially the open-ended, exploratory kind — would have fit that perfectly. He'd have loved watching a kid fall into a rabbit hole about volcanoes or medieval armor at 11pm because they wanted to, not because it was assigned.
He would have been deeply suspicious of "educational" software
Holt was critical of anything that dressed up coercion in a friendlier costume. Gamified learning apps, adaptive curriculum platforms, anything designed to get kids to learn things adults decided they should know — he'd have seen through the candy coating immediately. He wrote extensively about how the structure of school damages curiosity even when individual teachers are kind. He'd likely say that a lot of ed-tech just moves the cage online.
He would have trusted children's judgment about screens more than most adults do
Holt was consistently skeptical of adult alarm about what children chose to do with their time. He thought kids who seemed to be "wasting time" were usually doing something important that adults couldn't see or value. He'd probably have said that a kid playing video games for hours is doing something — learning systems, building persistence, socializing, exercising agency — and that the anxiety adults feel about it says more about adult discomfort with child autonomy than about any real harm.
He'd have been interested in programming specifically
Holt believed deeply in children as makers and experimenters. The idea that a child could make the computer do things — writing code, building games, automating tasks — would have excited him enormously. This aligns directly with his admiration for Seymour Papert, whose Logo programming language and constructionist philosophy Holt was aware of in his final years. He and Papert were traveling in very similar circles intellectually.
He would have worried about one thing: passivity versus agency
The distinction Holt cared about most was whether a child was doing or being done to. A child choosing to watch YouTube for four hours is exercising some agency; a child in a classroom being walked through an "interactive lesson" they didn't ask for is not, even if the screen looks the same. He'd have asked: Is the child in charge here? Are they following their own thread? That question, not screen time in hours, would have been his metric.

On social media and the attention economy, he probably would have been more conflicted than his followers sometimes assume. He trusted children, but he also wrote carefully about environments designed to manipulate — and he understood that not all constraints are visible. He might have distinguished between tools that serve the child's purposes and platforms engineered to capture attention against the child's interests.
The through-line in everything Holt wrote was: children are competent, curious people who learn best when trusted and worst when managed. He would have brought exactly that lens to computers — celebrating what they make possible while being sharp-eyed about who, exactly, is being served.

(I'm grateful to Karen James for thinking to get this!)


John Holt quotes and notes



Screentime



Video Games