Directory · R
Not sure this is the right entry? Run the check · New here? Start with the premise
Regret
2 min read · 441 words
Regret is you, now, reviewing a decision made by someone with less information — and finding it wanting. That review has exactly one useful output, and an unlimited appetite for your attention.
Is this you?
Check whatever lands. Nothing is sent anywhere.
Regret is the system reporting that a past decision, on review, is not the one the person would make now.
The report is honest as far as it goes. The current self, with current information and current values, evaluates an old decision and finds it does not match what it would choose today. That is data: standards have evolved, or the consequences have surfaced information that wasn’t available at the time, or the situation turned out to be other than it looked.
It rarely arrives as data, though. It arrives as the scene replayed at 2am — the fork in the road, the path not taken, the life that runs alongside this one in which the better choice was made. The mind builds that parallel life in detail, then grieves it, as though it had ever been real.
TWO COMPLICATIONS
The person who made the decision is not the person reviewing it. The choice was made with the information, capacity, and values available then. Judging it by what only became available later is judging someone for not seeing in the dark. The evaluation isn’t wrong, exactly — it just doesn’t translate cleanly into a verdict about who that earlier person was.
And the review can run forever without producing anything. This is rumination in regret’s clothing: the endless replay of if I had only and I should have, generating the full felt cost of the mistake again and again while changing nothing that happened and improving nothing ahead. Some people run this loop for decades, paying continuous interest on a debt that cannot be settled by paying it.
FROM THE CHAIR
Regret has one useful function and an unlimited appetite for bandwidth. The useful function: extract whatever the past decision teaches about what to do now and next. The wasted one: relitigating closed material to no operational benefit.
The discipline that converts regret into output: ask once, with deliberate attention, what the decision teaches. Take the lesson if there is one. Then close the file. Processing past the lesson adds only cost, never information.
When the regret involves harm to someone that can still be addressed, the response is not more internal regret. It is repair, where repair is possible. Continuing to feel bad about old harm — instead of doing the repair — converts potentially useful guilt into self-focused suffering. The other person is not helped by how badly you feel. They are helped by what you do now.
The past decision cannot be changed. Your relationship to it can.
Extract the lesson, repair what is repairable, release the rest.
Two things that make it unfair, and one that makes it endless
You're judging someone who couldn't see. The person who made that decision had the information, capacity and values available at the time. Reviewing it with what only became available later isn't a fair test, and whatever verdict you reach about them is answering a question they were never asked. The evaluation isn't exactly wrong — standards do evolve, consequences do surface information — it just doesn't convert into a judgement about who that earlier person was.
The parallel life was never real. Your mind builds the road not taken in considerable detail — the version where you took the job, stayed, left, said it — and then grieves it as though it existed. It didn't. It's a construction, and it's built to the specifications of the thing you currently want, with none of the costs the actual path would have carried.
And the review can run forever. This is the part that does the damage: rumination wearing regret's clothes, generating the full felt cost of the mistake again and again while changing nothing behind you and improving nothing ahead. People run this loop for decades, paying continuous interest on a debt that cannot be settled by paying it.
Which gives you the whole discipline. Ask once, with real attention, what the decision teaches. Take the lesson if there is one. Then close the file — processing past the lesson adds cost and never adds information.
Extract once, then close the file
Twenty minutes, on paper, on the regret you've carried longest. The point is to do deliberately and once what the loop has been doing badly and repeatedly.
Real questions
How do I stop regretting a decision I can't change?
Do the review once, properly, instead of repeatedly and badly. Write down what you actually knew at the time, what you'd choose now, and which specific new information accounts for the difference. Extract the lesson in one sentence — or note that there isn't one, which is often the honest answer. Then close it deliberately. The loop isn't searching for information any more; it has all of it. It's generating the felt cost again, and that's the part you're allowed to decline.
Why do I keep thinking about the life I didn't choose?
Because your mind constructs it in detail and then treats the construction as something lost. The parallel version is built to order — it has the thing you currently want in it and none of the costs the real path would have had, because those were never incurred and so never imagined. It isn't a memory of an alternative. It's a projection assembled from present dissatisfaction, which is why it tends to intensify when things are hard now and recede when they're not.
What's the difference between regret and guilt?
Guilt is about having violated your own code — it points at a wrong done. Regret is broader and often has no wrongdoing in it at all: you can regret a career, a city, a missed opportunity, with nobody harmed. They overlap when the decision you regret also hurt someone, and that overlap is worth noticing, because the responses differ. Regret asks what the decision teaches. Guilt asks what repair is still possible — and that one has an action attached.
Is regret ever useful?
Once, briefly, and then not at all. The useful function is extracting what a past decision teaches about now and next — standards you've developed, patterns you can see, information you'd gather differently. That's genuine, and people who refuse to feel any of it tend to repeat themselves. The waste is relitigating closed material. The whole skill is in noticing when the first has finished, because the second will happily continue for the rest of your life.
How do I deal with regret about hurting someone?
With repair, if repair is still available — and that's the key redirection, because continuing to feel terrible is often doing duty as a substitute for doing something. The other person is not helped at all by how badly you feel. They may be helped by an acknowledgement, even a late one. If repair genuinely isn't possible — they've gone, or contact would cost them more than it gives — then the remaining useful move is changing the behaviour that produced it, which is the only thing left with any value in it.
Why does regret get worse at night?
Because the conditions favour it. The distractions that competed for your attention all day have gone, you're tired enough that the part of you that would normally say this isn't productive is offline, and nothing is available to check the parallel life against. So the construction runs unopposed and feels more conclusive than it is. Verdicts reached at 2am are famously unreliable, and the same material in daylight is usually noticeably smaller.