A collector asked me once whether it bothered me that the code could make another one. It was a good question and I gave a worse answer than I wanted to, so I've thought about it since.

The honest version is that the code could make an effectively unlimited number, and that fact is the entire reason stopping at one means anything.

Scarcity that comes from restraint

There are two kinds of rare object. Something can be rare because making more is hard, or rare because somebody decided not to.

A hand-thrown pot is the first kind. The limit is real: hours, kiln space, a body that gets tired. Nobody has to promise you they'll stop, because the world stops them.

Generative work is the second kind. The system has no natural limit. I could run it tonight and have four hundred compositions by morning, any of which could be cut and assembled and sold. The scarcity is a decision, held in place by nothing but my continuing to hold it.

That's a weaker guarantee on its face, and I understand why it makes people hesitate. It's also the only kind of scarcity that says anything about the person making the work, because it's the only one that could have gone otherwise.

What the numbering on an edition does not tell you

An edition of 500 is a promise about how many prints exist. It says nothing about how many were made from the same file, which is one, or about how much of the artist's attention each of the 500 received, which is close to none after the first.

I made editioned work for years on fxhash before moving away from it. Multiple outputs from one algorithm, sold as digital pieces, and the interesting part of that period was the community rather than the model. What I noticed eventually was that the edition size was doing the work my judgment should have been doing. The algorithm decided what existed. I decided how many.

Reversing that is what I do now. I decide what exists. The number is always one.

The part that gets discarded

Every finished piece has an invisible mass of alternatives behind it, and I think the discarded work is where the value actually sits.

For each surviving composition there were dozens that didn't survive. Most were competent. They followed the rules and arrived nowhere in particular, and I looked at them and moved on. Then something held, for reasons I still can't articulate cleanly, and that one got made.

There are also the pieces that survived selection and then failed in translation to physical form. Looked right on screen, did nothing as a stack of layers, and got abandoned partway. That happens often enough to be a normal part of the work rather than a bad week.

So when I say one of one, the claim isn't that a rare thing exists. It's that a very large number of near neighbours were seen and declined, and the one on your wall is what was left.

What you can actually verify

Every piece carries a hash. That's a deterministic seed value: feed it back into the same generator with the same parameters and you get the identical composition, exactly, every time.

That property cuts in an uncomfortable direction and I'd rather name it than skate past it. The composition is reproducible, in principle, by anyone with the code and the hash. What isn't reproducible is the object. The board, the twelve or twenty cut layers, the hours of assembly, the registration held to a hair's width, the specific decisions about which elements sit on which plane. Someone with the hash has a picture. What's on your wall went through a laser and a workbench.

The certificate that comes with a piece is linked to that hash, so what you can check is specific rather than a signature on paper that means whatever you're willing to assume it means.

What it is not

It isn't an investment thesis. I don't know what any of this will be worth later and I'd be suspicious of anyone in my position who claimed to.

It isn't a guarantee that I'll never make something similar. The same generator produces a family, and family resemblance is the point of working in a series. What won't happen is a second instance of the piece you own.

And it isn't a promise about my restraint being permanent in some cosmic sense. It's a commitment I've made and kept, published where anyone can hold me to it, which is written down more fully in the manifesto.

The thing worth owning

Somebody standing in front of a piece is looking at the surviving output of a system with no natural edge, chosen by a person who ran it and waited, then cut and assembled by hand in a room that smells like burned paper.

The mathematics gave me endlessness. Deciding what deserves to exist is the work, and the object on the wall is the record of one of those decisions.

If you want the mechanical version of how the system works and why it produces what it does, that's here. If you want to know what the result does under changing light, that's here.

Shawn Kemp makes one-of-a-kind generative artworks fabricated as dimensional layered relief sculptures in Bellingham, Washington. The current work is here.