Cancelled printing plates

Why does a printmaker deliberately destroy the plate that could still make more money?

An artist finishes an edition of seventy-five etchings. The copper plate is still good. It could print more, and every further impression would sell. So the artist takes a burin and gouges a line across the image, or drills a hole through it, then prints one final sheet showing the damage.

Read as a business decision this looks like arson. The asset was productive, the market existed, and the owner ruined it on purpose. Which means either printmakers are collectively irrational about money, or destroying the plate does something that keeping it could not. It is worth working out what, because the answer is not really about copper.

Begin with what a buyer is paying for. Two etchings pulled from the same plate on the same afternoon are, as objects, effectively identical. If one belongs to an edition of seventy-five and the other to an open, unlimited run, they will not fetch the same price. The difference in price is not a difference in the object. It is a claim about how many others exist.

So the artist is selling scarcity alongside the image. Fine — but now ask a hard question. What makes that claim believable?

Suppose the artist simply promises: only seventy-five. Put yourself in the position of the artist five years later. The prints have sold well, the price has risen, and the plate is in a drawer. What does a strictly self-interested version of that artist do? Print more. The gain is immediate; the cost falls on the seventy-five people who already paid.

Now put yourself back in the buyer's position at the moment of purchase, reasoning that far ahead. You can see that incentive as clearly as the artist can. So the promise, on its own, is worth little — not because artists are dishonest, but because you cannot verify a promise whose breaking would be profitable and easy. Prices adjust to that, and every honest printmaker is paid as though they were the dishonest one.

That is the real problem, and notice that it hurts the artist, not the buyer. The buyer just pays less. The artist loses the entire premium that a credible limit would have earned.

So what would fix it? Not a stronger promise — there is no such thing. The only thing that helps is removing the artist's ability to break it. If the plate cannot print, then the future artist's incentive is irrelevant, because there is no decision left to make. The buyer no longer has to trust anyone; they can inspect the constraint. Cancellation converts a claim into a fact.

And notice how the practice is arranged to make the fact checkable. The mark is put through the image itself, not on the back, so any later impression carries the scar visibly. A cancellation proof is pulled and kept, often alongside the edition documentation, so there is positive evidence the plate was struck and evidence of what an after-the-fact impression would look like. The destruction is deliberately made public and legible — which is exactly what you would design if the point were credibility rather than tidiness.

There is a physical backdrop worth adding, because it explains why the convention arose when it did. Copper wears. Drypoint burr, which gives those velvety blacks, collapses after a handful of impressions; etched and engraved lines soften more slowly. So early editions were limited by the plate whether anyone intended it or not. Steel-facing — electroplating a hard skin onto the copper, in use from the mid-nineteenth century — broke that natural limit, and plates could suddenly yield hundreds of good impressions. The formal apparatus of numbered, signed, limited editions consolidated in the etching revival of the later nineteenth century, in the same decades. The limit stopped being physical, so it had to be manufactured.

Three refinements.

First, the misconception: cancellation is not about the physical uniqueness of the object, and not really about protecting collectors either. It is the artist buying a higher price by voluntarily giving up an option. The beneficiary is the seller.

Second, this is why softer alternatives underperform. A certificate, a signed declaration, a catalogue raisonne entry — all rely on someone else noticing a breach and enforcing it later, slowly and expensively. Cancellation is self-enforcing and verifiable by inspection, and that difference is the whole reason it survives as a norm.

Third, the argument only bites where reproduction is cheap and repeatable. A painter has no equivalent problem, because painting a second identical canvas is nearly as hard as painting the first. The commitment device appears precisely where the marginal cost of another copy is near zero — which is why the same logic keeps reappearing wherever copying is easy, in casting, in photography, and in digital editions, each of which has had to invent its own version of the struck plate.

• How bronze casting handles the same problem, given that a mould can be reused indefinitely.
• What changed when steel-facing removed the natural limit on a copper plate's yield.
• Whether digital editions have found any equivalent of a struck plate, or only the equivalent of a promise.

Every term the collection defines is gathered in the glossary.