george@han:~/blog$ cat everything-we-stand-on.md
← back to blog

Progress  ·  Technology  ·  Philosophy

Everything We Stand On Was Once a Destination

George Han  ·  September 2026  ·  Philosophy AI Work

A common worry is that AI is draining the meaning out of work. I think it is doing what progress has always done: converting purposes into instruments, so that purposes we could not previously have imagined now become available to us.

Around 370 BC, Socrates told a story about the invention of writing. The god Theuth presents the new art to the Egyptian king Thamus and recommends it as a remedy for memory. Thamus is unimpressed. Writing, he says, will not strengthen memory but replace it. People will trust to external marks instead of their own minds, and they will acquire the appearance of wisdom without its reality. Socrates agreed with the king.

He was right about the capacity. Trained memory really was a serious art, and it really did decline. But he was wrong about everything that mattered. Writing did not hollow out human thought. Instead, it became the floor that every subsequent form of human thought was built on. Law, mathematics, science, history, the novel, this essay. None of them were available to a person who had to hold the whole of their culture in their head.

What strikes me is not that Socrates made a mistake. It is that the mistake has a shape, and that the shape recurs, and that we are making it again right now about artificial intelligence. Someone points at a capacity that machines have taken over, notes correctly that we are losing it, and concludes that something essential has been subtracted from us. The conclusion never follows, and it never follows for the same reason.

Ends and Means Are Positions, Not Properties

I have written before about how “positive” and “negative” punishment are not properties of the punishment but properties of where you decide to put the zero. I think the same discovery applies here, and it is the whole argument in miniature.

Consider arithmetic. For most of history, calculation was a genuine end. It was hard, it took years to learn, it conferred status, and people organised their working lives around being good at it. “Computer” was a job title before it was a machine. Then calculation became cheap, and it stopped being a destination and became a step you take on the way to somewhere else. Nothing about arithmetic changed. Its position changed.

This is what I think people miss when they say AI is taking the meaning out of work. They are treating “being an end” as though it were an intrinsic property of an activity, like its difficulty or its subject matter. It is not. It is a position in a structure. And positions move.

“End” and “means” are not properties of an activity. They are positions in a structure — and progress is precisely the movement of things from the first position to the second.

The Demotion of Ends

Here is the claim in full. Progress is not the accumulation of comfort, or of knowledge, or of power. Progress is the constant conversion of purposes into instruments. Something that was once worth doing for its own sake becomes something you no longer have to think about, and once you no longer have to think about it, you can stand on it and reach for something that was previously out of range. In other words, progress is about constantly creating new meanings and new ends. Anyone who believes that certain ends are good enough for humans forever does not understand how this works. No end has been good enough. No end ever will be. What we are seeking is change itself.

Keeping a fire alive was once a role that a person could hold in a community. Growing enough food to survive the winter was, for almost everyone who has ever lived, the organising purpose of an entire life. Writing was a specialised craft practised by a professional class. Navigating by the stars was an expertise. Every one of these was demoted, and every one of them is now a floor rather than a summit — something you stand on without noticing you are standing on it.

The crucial part is what happens next, and it is the part that is invisible from below. The purposes we built on top of literacy could not have been described to a person arguing about whether literacy would ruin us. Thamus could not have been reassured, because the reassurance would have required him to imagine a scientific literature, and a scientific literature is not a thing you can imagine before you have writing. The new summit is not visible until you are standing on the old one.

Dewey got part of the way here, with his insistence that means and ends form a continuum rather than a hierarchy, and that every end achieved becomes a means to something further. What I want to add is the stronger claim: this movement is not something that happens during progress. It is what progress is.

Every floor we stand on was once somebody’s destination.
george@han:~/blog$