Nobody quits.
That’s the thing about reversion that makes it so hard to catch: there’s no moment where you decide to stop. If there were, you’d notice it, and noticing is most of the battle.
What actually happens is that at some point — usually a few weeks in, usually after a disruption — you have a thought that feels like clear-sightedness.
I was being a bit unrealistic about this.
And it doesn’t feel like giving up. It feels like maturity. Like you’ve finally got some perspective on a thing you’d been slightly manic about. There’s even a small physical relief attached, the sensation of putting something heavy down.
Two months later you’re exactly who you were in December, and you can’t reconstruct the moment it went.
Because there wasn’t one.
Here’s what actually happened.
What Reversion Is
Reversion is the return of previous behaviour after a period of change, produced by an unchanged self-concept restoring its set point. It is characterised by three features that distinguish it from a simple lapse. First, it is gradual and unannounced, arriving as a series of individually reasonable adjustments rather than as a decision to stop. Second, it generates its own justification: the mind supplies plausible, situation-specific reasons for the reduction, and these are experienced as insight rather than as rationalisation. Third, it is typically triggered by a disruption to circumstances rather than by a loss of motivation, which is why it commonly follows travel, illness, workload or a return to a familiar environment. Reversion is not a failure of will; it is a stabilising system operating as designed on an unchanged input.
Three triggers. They account for nearly all of it.
Trigger One: Stress Removes the Scaffolding
New behaviour costs something to run. That’s the defining feature of it — it contradicts the existing self-concept, so it requires ongoing effort to maintain.
Stress consumes exactly the resource that effort draws on.
So when a bad fortnight arrives — a deadline, a sick child, a genuinely difficult stretch — the new behaviour is the first thing to be dropped, because it’s the only thing on the list that was being held up by effort rather than by structure.
And here’s the mechanism people miss: the old behaviour doesn’t have to be chosen. It doesn’t need effort. It’s the default, sitting underneath, requiring nothing. Remove the effort and it’s simply what’s there.
That’s why reversion feels like falling rather than like deciding. You’re not moving toward the old version. You’re releasing the thing that was holding you off it.
Which produces the most reliable prediction in this whole area: you will not revert during the difficult weeks you’re worried about. You’ll revert during the ordinary week immediately after them, when the pressure lifts and you discover you’ve already stopped.
Trigger Two: Old Rooms Restore Old Versions
You already know this one, you’ve just never had it named.
You go home for a few days. Within about ninety minutes you’re using a tone of voice you haven’t used since you were nineteen, taking positions you don’t hold, and reacting to a sibling with an intensity that would embarrass you in any other context.
You’re not choosing any of it. It’s being elicited.
Environments carry cues, and cues activate the version of you that was formed in them. Your family holds the longest-standing and most detailed model of who you are, and being around people running an old model of you exerts steady pressure to conform to it — not through anything they say, mostly. Through the sheer weight of being consistently treated as somebody you used to be.
For a change effort this is genuinely dangerous, and dangerous in a way people underestimate, because a weekend in an old environment can undo more than a bad month of stress. Stress interrupts the behaviour. Old rooms reinstall the identity.
There’s a full treatment of this at why you become the old version around the old people. For here, one practical note: this is why “can you change while staying in the same environment” is a real question with a genuinely mixed answer, and why the people who change most reliably often changed something about their surroundings without ever framing it as part of the plan.
Trigger Three: Nobody’s Watching, and That Turns Out to Matter
This one is uncomfortable because it undercuts something you’d like to believe about yourself.
A meaningful portion of early change is held up by observation. Someone knows. Someone will ask. There’s a small social cost to stopping.
Remove the observer and a specific thing happens: the behaviour stops being an identity claim and becomes a private preference, and private preferences are negotiable in a way public claims aren’t.
I’m not saying you’re only doing it for other people. I’m saying observation was doing structural work you didn’t account for, and when it’s removed the structure gets lighter without anyone noticing.
The useful diagnostic: if it stops when nobody’s looking, it hasn’t become identity yet. That’s not a moral verdict. It’s a stage report — you’re earlier in the process than you thought, and knowing that is worth more than the reassurance.
The Reasons Are the Tell
Now the most useful thing on this page.
Reversion doesn’t arrive as a decision to quit. It arrives as a reason, and the reason is always good.
Not vague. Not obviously self-serving. Specific, situation-appropriate, and completely accurate.
It’s a genuinely mad quarter — I’ll pick it up in January.
I was being rigid about it, and rigidity is its own problem.
I’ve realised this wasn’t the right approach for me.
I need to be kinder to myself about this.
Look at that last one especially, because it’s the most effective one your mind has and it comes wrapped in the language of wellbeing. It is very difficult to argue against being kind to yourself, which is precisely why it’s the reason that gets deployed when the others have failed.
Here’s the test that cuts through all of it:
Would this reason have been persuasive to you three weeks ago?
Because usually — not always, but usually — nothing has actually changed. The quarter was always going to be busy. You knew about the travel. The approach hasn’t been shown to be wrong; it’s been shown to be effortful, which you knew on day one.
What changed isn’t the situation. It’s that a reason became necessary, and your mind is exceptionally good at producing one on demand.
That’s not dishonesty. It’s a stabilising system doing its job, and it uses your own voice and your own vocabulary, which is exactly why it’s so convincing.
Lapse Versus Reversion
Worth separating precisely, because treating one as the other causes most of the unnecessary damage in this territory.
A lapse is a gap in behaviour. You missed three days. The identity is intact; the practice was interrupted. You resume, nothing has changed structurally, and the whole thing is a non-event.
Reversion is a change in the underlying claim. You’ve stopped considering yourself the person who does this. The behaviour going is a symptom rather than the event.
They look identical from outside — in both cases the person isn’t doing the thing.
The distinction is in whether the resumption requires a decision. After a lapse, you just start again; there’s nothing to reconsider. After a reversion, restarting requires re-adopting the identity, which is why it feels so disproportionately heavy to begin again after a long gap. You’re not restarting a behaviour. You’re re-making a claim you’ve since quietly withdrawn.
Rough rule from the pillar: one missed day is a lapse the identity absorbs without difficulty. A missed week starts functioning as evidence for the old identity, and evidence accumulates in both directions.
Where This Connects
The reason any of this happens is covered fully in the identity pillar: behaviour installed on top of an unchanged self-concept requires maintenance, and maintenance breaks when circumstances break it.
But the older version of the argument is more interesting, because it makes the same claim without any modern vocabulary.
A Ming dynasty official had his entire life forecast accurately for twenty years, and what made him forecastable was that his character was stable. When he eventually changed the trajectory, he didn’t install behaviours. He audited his character in writing and worked on the generator for a decade. That’s the method, and the reason he didn’t revert is that there was nothing left to revert to — the input had genuinely changed.
That’s the whole difference. Behaviour reverts. Changed character doesn’t have a previous state to fall back into.
What This Is Not
It is not a discipline failure. Disciplined people revert constantly, and frequently harder, because they can sustain contradicted behaviour longer on effort alone — which means they get further before the collapse and have more distance to fall. Endurance postpones reversion. It doesn’t prevent it.
It is not evidence the change was fake. The three weeks were real. You genuinely did the thing, and the evidence you generated during it still counts. Reversion doesn’t retroactively delete the record, and treating it as though it does is how people end up with a decade of “failed attempts” instead of an accumulating case.
It is not permanent. A reversion isn’t a verdict on your capacity. It’s information about which trigger got you, and the trigger is usually identifiable within about five minutes of honest attention.
The Cost
Here’s the price of taking this seriously, and it’s an odd one.
You have to stop trusting your own reasons at the exact moment they feel most sensible.
That’s genuinely disorienting. The reason you’ll produce for stopping will be a good reason, delivered in your own voice, and treating a convincing thought as suspect specifically because it’s convenient is not a comfortable way to operate.
It also can’t be applied everywhere. Sometimes the reassessment is correct. Sometimes the thing genuinely was wrong for you and stopping is the right call, and a person who treats every doubt as sabotage will grind on with something they should have abandoned in week two.
There’s no clean rule separating the two. The best available test is timing: reassessments that arrive immediately after a disruption are suspect. Reassessments that arrive during a good week, when the thing is going fine and you have no need for a reason, are usually real.
That’s an imperfect instrument and I’m not going to pretend otherwise.
Common Questions
Why do I always go back to my old habits?
Because the old behaviour is the default and requires no effort, while the new one contradicts your self-concept and has to be actively maintained. Anything that interrupts the maintenance — stress, travel, illness, a change of environment — returns you to the default without a decision being made.
Why does reversion feel like a sensible reassessment rather than quitting?
Because your mind produces a genuine, situation-specific reason at the moment one becomes necessary, and it arrives in your own voice and vocabulary. The reliable test is whether that reason would have persuaded you three weeks earlier. Usually nothing has changed except that a justification became required.
Why do I revert after a stressful period rather than during it?
Because during the difficult stretch you’re still holding on, and the release comes when the pressure lifts. The dangerous week is almost always the ordinary one immediately afterwards.
Why does going home undo weeks of progress?
Because environments carry cues that activate the version of you formed in them, and family holds the most detailed and longest-standing model of who you are. Being consistently treated as someone you used to be exerts real pressure to comply, without anyone intending it.
What’s the difference between a lapse and a reversion?
A lapse is a gap in behaviour with the identity intact — you simply resume. A reversion is a withdrawal of the underlying claim, which is why restarting afterwards feels so much heavier than the behaviour itself would justify.
Does reverting mean the change was never real?
No. The evidence generated during the period still counts and doesn’t get deleted. Treating each reversion as a total loss is how people accumulate a list of failures rather than an accumulating case.
Where to Start
This is the most useful ten minutes in the whole branch, and almost nobody does it.
Pre-write the reasons.
Right now, while you’re not in it, write down the three reasons you will give yourself for stopping. You know what they are. You’ve used them before, on previous attempts, and they’ll be roughly the same ones this time because you’re roughly the same person.
Write them on a page. Date it. Put it somewhere you’ll find it.
That’s the whole exercise.
Because the entire power of a reversion reason is that it feels like it arrived from nowhere, as fresh insight, at the moment you needed it. It’s persuasive precisely because it appears to be a response to the situation.
Meeting it in writing, in your own handwriting, dated six weeks earlier, removes that completely. It stops being insight and becomes a scheduled event you predicted and are now watching happen on time.
You’ll still feel the pull. But you’ll be reading your own forecast while it does, and it’s remarkably difficult to be moved by a reason you wrote down in advance as the thing you’d say.
I’ve spent twenty-five years working on how ordinary people develop capabilities that look impossible from the outside, and the people who change permanently aren’t the ones who never feel the pull back.
They’re the ones who recognised it, because they’d written it down before it arrived.