Cold-process soap is chemically finished in forty-eight hours. The reaction between oil and lye has run its course; the bar holds its shape, takes a cut, behaves like soap. And yet it is not done. What follows is another four weeks in which nothing visibly happens. Water leaves the bar slowly. The structure firms. A soap used too early lathers thin and wears fast, technically soap, but not finished soap. The difference is not in the chemistry. It is in the waiting.
Knowing when a thing is finished is among the harder judgments a maker carries, precisely because it cannot be reduced to a checklist. There is a clock for curing, a calendar that says four weeks, and the calendar is right often enough to trust. But the calendar is a proxy. What it stands in for is a state of completion that has to be recognised rather than measured, felt in the hand, the weight, the way the surface takes light. Rules approximate the judgment. They are not the judgment itself.
Two ways to get it wrong
There are precisely two failures, and they pull in opposite directions.
The first is stopping too early. The bar pulled from the rack before its time. The sentence sent before its weakness has been found. The painting signed while a corner still asks for attention. Premature completion has a particular flavour: a thinness, a sense that the thing was let go before it had become fully itself. It is the easier mistake to forgive, because it usually comes from impatience or modesty rather than vanity.
The second failure is the more insidious one, because it disguises itself as diligence. The painter who returns to the canvas once more. The writer who revises the life out of a paragraph until it lies flat and lacquered. The maker who cannot leave the thing alone. Overwork does not announce itself as damage. It feels like care. And then one pass too many arrives, and the looseness that gave the work its life is gone, sanded into a deadness that no further effort can recover. The over-worked thing is harder to repair than the unfinished one, because there is nothing left to push against.
Both failures share a root: a refusal to trust the judgment of done. One refuses it out of haste. The other refuses it out of an anxiety that nothing is ever quite enough.
What Leonardo understood
Leonardo da Vinci is supposed to have said that art is never finished, only abandoned. The line endures because it admits something most makers feel and few state plainly: that completion is rarely a clean arrival. There is no point at which the work declares itself done and lays down its claims on you. There is, instead, a moment when continuing would cost more than it returns, when the next adjustment is as likely to take away as to add.
Abandonment, in this reading, is not failure. It is the recognition that the work has reached the limit of what further attention can improve. To call it abandonment rather than completion is honest about the asymmetry: the maker leaves, and the work stays. Da Vinci, who carried the Mona Lisa with him for years and returned to it again and again, knew the temptation to keep going better than most. The line is not cynicism. It is the confession of someone who found stopping genuinely difficult.
This is why completion is a skill and not a procedure. It is learned the way any judgment is learned, through the accumulation of cases, the soaps cut too soon and the ones left too long, the sentences killed by revision and the ones saved by it. Over time a sense develops that cannot be written down. The maker comes to recognise the moment the way one recognises a face: not by inventory of features but by an immediate, whole apprehension.
The four-week wait, then, is not really about time. It is about deferring the judgment until the thing is ready to be judged, and then trusting yourself to make the call. To stop. To not pick the bar up again, turn it over, decide it needs one more day, and one more after that.
A thing is finished when leaving it alone improves it. The discipline is learning to feel that moment, and then to honour it.