The Cities That Built the Thesis

Part one ended with a teenager and a health crisis. Part two is about geography — or really, about why geography turned out to be a red herring. I've lived in eleven cities across twelve countries, and every single move came with the same private promise: this time I become the disciplined one. It almost never held.

The clean slate that wasn't

For years I treated a new city like a factory reset. New kitchen, new commute, new strangers who didn't know my old habits. I'd unpack and think: here is where the version that finally has it together gets installed.

Then three weeks would pass. And I'd watch myself rebuild the exact rhythms I thought I'd left at the airport. Same late nights. Same skipped meals when work got loud. Same instinct to answer one more message before bed.

The surfaces were genuinely different. Lisbon doesn't smell like Istanbul. The light in one apartment was nothing like the light in the last. But the person standing in all that new light kept reaching for the same old defaults.

What made it useful instead of just frustrating was my work. Across those moves I sat with roughly 7,000 people — founders, students, parents, people who looked fine on paper. And I kept meeting the same person inside very different lives. The founder who couldn't stop working. The student who couldn't sleep. The woman smiling through a day that was clearly costing her something.

Eleven cities. Twelve countries. One pattern underneath all of it. That repetition is what eventually became the thesis.

The mechanism, and it isn't romantic

Here's why the clean slate fails, and it's not a character flaw.

Behavior is cued mostly by two things: your environment and your identity. James Clear makes this point well — willpower is a poor engine, and most of what you do runs on context and on who you believe you are. Move cities and you change one of those two things hard. The environment flips overnight. New fridge, new street, new faces with no expectations of you.

But identity doesn't go in a box. It doesn't get shipped separately. It walks through the new front door wearing your face.

So what actually happens after a move is a quiet negotiation. The new environment offers different cues. Your old identity reaches for familiar outputs. And within about three weeks, the identity wins, because it's older and stronger than any apartment. It reshapes the new setting until the setting starts producing the old behavior. The desk migrates to where you can work late. The phone ends up by the bed again.

Research on habit formation backs the timeline loosely — new contexts can interrupt a cue, but only briefly, and only if you also change the response, not just the room. The view changes fast. The wiring changes slowly, and only on purpose.

Why "fresh start" is the wrong tool

The fresh start is seductive because it's real for about a week. There's a genuine window after any big change — a move, a new job, a Monday — where old cues go quiet and you feel like a different person. People mistake that quiet for transformation.

It isn't. It's a gap. The old triggers are temporarily out of range, so the old behaviors don't fire. You feel free. You credit the city, the calendar, the diet, the new planner. Then your identity reasserts itself, rebuilds its cues in the new space, and the behavior comes back. Now you've also added a fresh layer of self-blame, because you had your chance and "wasted" it.

This is why people chase the next move, the next program, the next reinvention. Each one delivers that real-but-temporary window, and each one ends the same way. The relocation becomes the strategy. The actual operating system never gets touched.

I did this for years before I understood it. I wasn't lazy and I wasn't unusual. I was running a high-effort plan aimed at the wrong layer. Changing where you do something does almost nothing to change why you do it.

What to change instead, concretely

If the setting is a costume and the system runs the show, then you work the system. Three places, in order.

First, name the identity that's driving the loop, in plain words. Not "I want to sleep better" but "I'm the person who proves her worth by being available at midnight." The behavior makes perfect sense once you see the belief paying for it. You can't edit a story you won't say out loud.

Second, change the response to one existing cue — keep the cue, swap what it triggers. You don't need a new city for this; you need the same trigger to produce a different output. The bed is the cue. The new response is the phone charging in another room. Same environment, rewired loop.

Third, make the new identity cheap to act on once and let evidence accumulate. Every time you act like the calm version, you cast a small vote for being her. Clear's framing again: you're not trying to hit a goal, you're trying to become the type of person for whom the behavior is normal. That happens through repetition, not relocation.

None of this requires a plane ticket. That's the point. The change you keep trying to buy with geography was always available where you already are.

But sometimes the city really is the problem

I want to be honest about the limit of this, because it's easy to flatten into "environment never matters." It matters enormously.

Some environments are genuinely toxic — a job that runs on fear, a home that punishes rest, a social circle where every cue points at the behavior you're trying to leave. In those cases moving isn't avoidance. It's removing real triggers that no amount of inner work can fully outvote. If you're standing next to the bottle every night, willpower is a bad plan; distance is a good one.

The distinction is this. Change the environment to weaken specific cues you've already identified, and pair it with a changed response — that's surgery. Change the environment hoping the new scenery will rewrite you while the identity rides along untouched — that's the costume swap I kept making.

So the question isn't "should I move." It's "what exactly am I expecting the move to do, and have I changed the thing inside me that will otherwise rebuild the old life on arrival." Move with that answer in hand and a relocation can genuinely help. Move without it and you've just bought a more expensive version of the same week-one window.

The thesis the cities handed me

Eleven cities taught me something one city never could have: the human is portable, and so the human is the leverage point. Everything else is set dressing.

That's the spine of what I build now. We hand the repeatable work to systems and agents so the surface of your life gets quieter — fewer fires, fewer 11pm tasks. But quieter surroundings don't reprogram anyone. I watched my own apartments prove that for a decade. The calm has to be installed in the person, or the person will quietly rebuild the chaos in the calm.

This is also why I stopped believing in the dramatic before-and-after. The real work isn't a clean slate. It's the unglamorous, in-place rewrite of the loops you carry everywhere — the same loops that survived twelve countries with you. You don't have to go anywhere to start. You're already standing in the only environment that was ever going to matter.

The setting is a costume. The system runs the show. That sentence cost me eleven moves to learn. It's free now. The behavioral science behind changing the system, not the scenery, is most of what we write about at marsa.ai.