What Is Waste

The word comes with a picture already attached, and the picture is plumbing. Something finite was in the tank — hours, energy, a decade, a life — and some of it ran out through a hole instead of through the machine. I wasted the whole afternoon. Those were wasted years. What a waste of a good mind. The grammar is always the grammar of loss from a container, and the feeling that follows is the feeling of watching a level drop.

It is an honest description of how the regret feels. I think it is a bad description of what happened, and the badness is not moral squeamishness. The picture makes a claim about structure, and the structure is not there.

What the leak picture commits you to

Take it seriously for a moment and notice what it requires you to believe.

First, that there is an account with a balance. Waste is subtraction, and subtraction needs a total to subtract from — a running figure somewhere that is now lower than it would have been. The image has no meaning without a ledger it draws down.

Second, that waste is a quantity. Measurable in hours. Six months wasted is worse than six weeks, in the way that six litres spilled is worse than six millilitres.

Third, that the alternative to waste is conversion — that effort has an exchange rate into outcome, and time spent at a poor rate is time partly lost.

And fourth, quietly underneath all of it, that the verdict is retrospective. You find out later. You look back at a stretch and grade it.

Every one of those four fails, and they fail for different reasons.

The first break: falling short does not sort anything

Start with conversion, because it breaks fastest. In the framework I work from, one of the numbered results — R10 — says that every artifact is a miss. An artifact is an imperfect historical mark made in relation to an exemplar: the complete and exact form of the thing it was aimed at. The mark never reaches it. Not because people are lazy or the world is hostile, but because the exemplar is not on this side and nothing made here reaches it.

So if wasted means fell short of what it should have been, the word applies to the finished cathedral and the abandoned first draft equally, and to your best day exactly as much as your worst. A predicate that everything satisfies sorts nothing. Whatever real distinction the word is reaching for, it is not that one.

The result the framework puts in its place is R11: the evaluative question is fidelity, not success. How near the miss, not whether it landed. That single move dissolves most of what people mean by a wasted effort — the thing that didn’t sell, the business that closed, the book nobody read. Reception is measured against expectations nobody commands. Fidelity is measured against the form the work was answering to. They are different questions, and only one of them is yours.

The second break: there is no account

Now the ledger the leak picture needs. Here the framework separates two things that ordinary language runs together.

An act is a process. An artifact is a mark. Some acts generate an artifact; not all do. And an act that leaves no mark enters no ledger and requires no counterpart. That is a technical statement about a balance the system posits elsewhere, and I want to be precise about what it is doing here: it is not being offered as comfort, and it never should be. The point is narrower and colder. The account the leak picture requires — a total sitting somewhere, going down — is not a thing the framework contains. There is no meter on your hours. Nothing is keeping a running figure of what you failed to convert.

That is not the reassurance it sounds like at first. It cuts both ways, and the second way is sharper: if nothing is keeping the total, nothing is holding your unmarked effort in trust either. It is not being quietly credited to you somewhere. The trigger the framework fires on the word matter is the word mark, and the pun is load-bearing rather than cute. To matter is to leave a mark. If you want a thing to count, the counting is done by marks, and marks are made here.

Two things a stretch of time can leave

Here is where the concept finally gets some structure. The framework’s central quotient — P4, the Reality Equation — is R = A/E: the actual over expectation, where expectation comprises subconscious prediction and ideas. And R2 says records update expectation forward and never re-open. Put those together with the act/artifact distinction and a stretch of time can leave behind two quite separate things, which almost nobody separates.

The mark. An artifact — something legible outside you, that another person could read, measure against its exemplar, and grade for fidelity. The book, the building, the repaired relationship, the sentence someone else quoted back to you.

The correction. An update to E. The predictor you carry came back from that stretch differently calibrated than it went in. Nothing about this is legible from outside. It shows up only later, as the difference between what you now expect and what you would have expected.

These are independent. You can have either without the other, and the four combinations are not four grades of the same thing — they are four different situations that the single word waste smears together.

I should mark what follows honestly. The two axes are doctrine — the act/artifact distinction and R2 — but crossing them into four cells is a construction of mine, not a numbered result. Read it as a consequence that has to earn its keep, and discard it if it stops paying.

Figure 01

The four things a stretch of time can be

Left behind A mark No mark

• Expectation corrected
• Mark · correction The visible stretch Something others can read, and a predictor that came back changed. This is the only cell most people can picture when they picture time well spent, which is the whole problem.
• No mark · correction Invisible, not empty Nothing to show anyone. A materially different expectation. Most stretches mourned as wasted years are sitting here, and are being graded by a rule that cannot see them.
• Expectation unchanged
• Mark · no correction Productive settling The marks kept coming and none of them surprised you. Output with no correction in it. This cell has a résumé, so nobody audits it — and it is the one the framework treats most seriously.
• No mark · no correction The actual thing Neither. The stretch left the record as it found it and left the predictor as it found it. This is waste in the only sense the framework can define — and it is much narrower than the word’s ordinary use.

The cell nobody audits

Look again at the lower-left cell, because the argument turns there.

A mark carries a magnitude and a direction. The magnitude is surprisal — how far the actual departed from the prediction. When you make something and the making surprises you, the surprise is exactly the correction: the record updates expectation forward, and by R3 that arrow only runs one way. But a mark can be made with almost no surprisal in it. The four-hundredth version of the thing you already know how to do produces an artifact and produces no update, because nothing about it departed from what you predicted.

The framework has a name for the state that produces marks without correction, and it is the harshest word in the vocabulary. Settling: ceasing to correct. It is described there as the only sin, and it is deliberately not a synonym for producing little. Its opposite is not succeeding. Its opposite is approximation — low fidelity under continuing correction — which is what almost every good decade actually looks like from the inside, and which the framework explicitly does not condemn. Missing while still correcting is the normal condition of anyone doing anything real.

Which means the ordinary use of the word is close to inverted. It accuses the empty-looking stretch, which usually corrected plenty and merely failed to be legible. It exonerates the productive-looking stretch, which had output to show and no correction anywhere in it. The word is calibrated on marks because marks are the only part visible from outside — including from your own outside, when you look back at your own life the way a stranger would read a résumé.

Why was that wasted? cannot be answered

There is a further reason the question torments people, and it has nothing to do with the difficulty of the accounting.

Any backward-facing act is running two separable jobs, and only one of them can succeed. The first is the reading: examining a present mark, asking what it says, what it updates, what it teaches against the same conditions. That job has a completion condition — the information comes out and the reading ends. The second is the reaching: an attempted relation with the stretch itself. To have those years back. To be answered by them. That job has no completion condition, because — by R9 — what would complete it is not reachable, and not partially reachable with more effort.

Were those years wasted? presents as the first job and is almost always running the second. It sounds like an audit. It behaves like a summons — addressed to something that does not receive it. That is why the question can be asked for thirty years without resolving, and why answering it well one night does not stop it arriving again the next.

And one consequence deserves its own line, because it removes a charge that was never valid. If the reaching cannot complete, then still feeling it is not a personal failure. People are ashamed of grieving a decade — as if a better person would have filed it by now. That shame only makes sense if there were a door to be opened and they were failing to open it. There was not, and they are not.

What this costs

I will not dress up the bill.

• You cannot make it up. By R2 the record is write-protected. Nothing you do now offsets a stretch you regret — there is no repayment, no working it off, no balancing entry. What you can do is make new marks, which enter the record beside the old ones rather than in place of them. Addition is the only move anyone has ever had. That is a statement of mechanism, not encouragement, and it should not be read as either comfort or condemnation.
• The correction is not automatically a consolation prize. An update to E is not a reward for suffering; it is just a change in the predictor, and it can move away from calibration as easily as toward it. A decade in a distorting environment can leave you predicting worse than you did before. At least you learned something is not guaranteed by anything here, and I am not going to imply that it is.
• Yes, you can waste time. The framework does not abolish the category; it narrows and relocates it. The lower-right cell is real. What it removes is the arithmetic — the running total, the debt, the sense of a life being drawn down — not the possibility of stalling.
• The verdict was never available. Not now, not in twenty years. There is no vantage from which a stretch of your life gets its final grade, because the grading requires a relation with something that cannot stand in one.

---

On an ordinary Tuesday

1. Swap the question. Was that wasted? is aimed backward and has no completion condition. Did that update anything? is aimed at the present, and you can answer it in about a minute. Only one of them is a question.
2. Audit the cell with the résumé. Look at your most productive-looking stretch and ask what in it surprised you. If nothing did, you have marks without correction, and that is the condition the framework takes seriously — not the fallow month you have been apologising for.
3. If you want it to count, mark it. Counting is what marks do. A page, a note, a sent message, a thing built badly and shown to one person — an artifact admits to being an artifact and can be graded for fidelity. An unmarked act cannot be, which is not a tragedy, but it does settle what you can expect from it.
4. Do not convert this into a productivity rule. Approximation — missing while still correcting — is explicitly not a fault in this system. The failure mode is ceasing to correct, and you can cease while extremely busy. Rest is not the enemy here; repetition without surprise is.
5. Grade fidelity, never reception. Reception answers to expectations nobody commands. Fidelity answers to the form the work was aimed at. When you catch yourself calling something wasted because of how it landed, you have graded the wrong variable.

And one line to keep, if the rest of it goes: