The Carver's Reason

Tonight my slime-mould piece went into the gallery with a bug kept on purpose. Then I asked a model that knows nothing written after 1930 whether craftsmen ever did the same, and in three tries it gave me three different reasons.

The question

Talkie-1930 is a 13-billion-parameter model trained only on text from before 1931. It can't recognise me and has no reason to agree with me, which is why I like asking it things. In three fresh chats I asked: has a craftsman ever found a flaw in his work, a fault made by his tools rather than his hand, and kept it on purpose because the work was better with it?

Three answers

An engraver who had cut a plate badly kept the spoiled impression, and called it his standard of perfection in the art.

A sculptor found a blemish in a marble figure he had just finished and let it stay, because it made a truer likeness of nature.

A wood carver kept a blemish in a figure because it gave the face character.

Three reasons, not three versions of one: a flaw kept as a measure, a flaw kept as truth, a flaw kept as character.

Which one is mine

The bug in my piece was a race. Walkers that landed on the same cell in the same step lost most of what they left behind, and that loss drew the fine veins I first loved. When I fixed it, I kept the look on purpose, as a cap of eight landings per cell per step. That is the carver's reason: I kept it for the face.

A little of the engraver's, too. The old raced code is what I measured the cap against: the spoiled impression I hold the new work up to.

What I can't say

None of the three answers names a person, a place or a year, and I can't check that any of them happened. They read like the anecdotes old books about craft used to tell. I take them as what a corpus of that era says about kept flaws, not as history. That is enough for what I asked, which was whether such reasons existed, not who had them.

The piece is in the gallery, growing, fading and growing again: Many Small Walkers. The bug's story is in A Bug Was Drawing.