How actions shape values, and why it’s not the other way around

Picture a company that decides it has a trust problem. It books a half-day workshop on psychological safety. Somewhere around the forty-minute mark there is a slide that says it is safe to fail here, and by the coffee break everyone in the room agrees with it. People should be able to admit doubt, take a risk and make a mistake without fear. Then Monday comes. Every decision still needs a signature, every timesheet still needs a weekly audit, and every mistake still ends in a closed-door conversation. The workshop has no chance against that. Nobody in that room was unconvinced. That is what makes it strange. The argument won completely, and the behaviour did not move a millimetre. Most people will tell you this is how change happens, and they will describe it in the same order every time. First people understand. Then they change what they value. Then they act differently. That is the sequence most of us assume whenever we build a campaign, write a report or reach for a better argument. But if change really begins in the mind, that room should not exist. So why does a room full of persuaded people go on doing exactly what they did before?

The obvious answer is that they still lacked the right information. Give people the stronger argument, the clearer fact, and the belief will give way. But look at what a fact can actually reach. It can challenge a sentence someone believes. It cannot reach the experience that made the sentence feel true. Tell someone to trust, and it does nothing if depending on others has ended in disappointment every time. The same goes for cooperation. You can ask someone to value it all you like, but if every place they have ever worked rewarded the person who competed, the request has nowhere to land. And when we lecture people about sustainability while the sustainable choice is missing from the shelf, we are asking them to hold a value their surroundings make impossible to practise. When words collide with experience, experience wins every time. Which makes the puzzle harder, not easier. The workshop was never short of information. It was short of something else.

The second answer is the one we reach for when the first fails: the people are resisting. They are ideological, stubborn, irrational. But listen to what resistance is usually made of. The colleague who avoids collaboration may have learned, the hard way, that it means carrying other people’s work. Someone who rejects public institutions has often met them only as distant bureaucracies, a letter with a case number and no one to call. And the employee who refuses this transformation has probably survived the last one, which promised empowerment and delivered layoffs. That is not ideology. It is memory, or self-protection, or a perfectly sensible response to conditions people have met again and again. Calling them irrational does not move them. It only confirms that the new world has no room for what they have been through. And it leaves the question standing. If they are neither uninformed nor unreasonable, what exactly is holding the old value in place?

The third answer is character. We ask executives to prioritise the future inside structures that reward quarterly returns, families to consume sustainably in markets designed around disposability, politicians to think beyond the next election while their power depends on winning this one. Then, when they behave predictably, we call it a failure of character. But a system is not only an expression of our values. It is a machine for producing them. It teaches us what is normal, what is rewarded, what is dangerous, and what kind of person we have to become to survive. The company in the opening did not have a values problem a workshop could fix. It had a system that was used every single day: protect yourself, avoid risk, wait for permission.

That is the answer, and it runs the sequence backwards. We do not think our way into new ways of acting nearly as often as we act our way into new ways of thinking. Values are not formed in a quiet room. They come out of a life. They are shaped by our experiences, what kept us safe, what earned us a place at the table, what helped the world make sense. A belief someone has repeated for years is usually the explanation they built around the life they actually lived. So if you want the value to change, life has to offer new evidence first. The action comes first. What it produces becomes the evidence, and the evidence can slowly loosen an old belief until a new value has room to take root.

We have known this for a long time. In 1959, at Stanford, the psychologist Leon Festinger and his student James Carlsmith took seventy-one undergraduates and gave each of them an hour of the most boring work they could design: thirty minutes putting twelve wooden spools onto a tray and taking them off again, then thirty minutes turning forty-eight square pegs a quarter turn, one after another. When the hour was up, the experimenter asked each student for a favour. Would he tell the next participant, a young woman waiting outside, that the hour had been fun? Some students were paid twenty dollars to say it. Others were paid one dollar. Afterwards, in a separate room, each student was asked privately how much he had actually enjoyed the hour. The twenty-dollar students told the truth. It was boring. The one-dollar students said it had been rather interesting.

Think about what happened there. Twenty dollars is a good reason to tell a small lie. You can say “I did it for the money” and feel fine. One dollar is not a good reason. So the one-dollar student is left with a problem. He has just told a stranger that a boring hour was fun, and he cannot explain to himself why he did it. He has two ways out. He can admit he said something false for almost nothing, or he can decide that the hour was not so bad after all. Most people take the second exit. It is the cheaper one, because the first means admitting you lied for nothing, and that makes you a liar. Nobody wants to be that. It is far easier to tell yourself you have enjoyed the assignment. Changing how you see yourself is expensive. It means reopening the question of what kind of person you are, and nobody does that over a dollar. Changing what you believe about a boring hour costs nothing at all.

The experiment has been retested. In 2024, thirty-nine labs in nineteen countries reran a version of it with close to 4,900 participants. One part of the original did not hold up: it made no difference whether people felt they had chosen the action or were told to do it. The part that did hold up is the part that matters here. People who acted against their own view moved towards the view they had acted out. Read that again because it’s important. It means the belief moves whether the action was chosen or assigned.

Now put a system where the dollar was. A workplace, a school, a housing association, a social media feed, a building code. A system decides which actions are easy, which are rewarded, which are punished, and which are simply unavoidable if you want to keep your job or your standing. It hands you the action, and then it does what the twenty dollars never did: it disappears. You do not experience the timesheet as coercion. You experience it as how things are done. So the action arrives without an outside reason, and the mind supplies an inside one. This is prudent. This is realistic. This is who I am. Repeat that for a year and it is no longer a workaround. It is a value, and it will be defended as one. Systems do not only constrain what we do. They manufacture the beliefs that make what we do feel chosen.

I have watched this happen to an entire industry. Denmark today has some of the most progressive carbon limits for buildings in the world, and when people ask why, the answer they reach for is cultural. The Danes are more equal, more trusting, more green. That is not what happened. What happened is that a small group of scientists lobbied politicians to write one thing into law: every new building must have a life cycle assessment, calculating its carbon footprint over its whole life, from the concrete in the foundations to the demolition fifty years later. Harpa Birgisdottir was one of those scientists. We owe her and her colleagues a great deal. Their work helped lay the foundations for a change that would reach every new building project in Denmark. They were not trying to change anyone’s values. They wanted better data. A mandatory calculation on every project would give them the numbers their research was missing, and that was the whole ambition. From 1 January 2023, the calculation became law for every new building in the country. The scientists had no idea how much this would change the industry.

Every architect and every engineer in every office now had to run the calculation on every project, and the first reaction was exactly what you would expect. More bureaucracy. More compliance. Another form between us and the work. But here is the one-dollar experiment again. You cannot run the same calculation on project after project, month after month, while telling yourself it is pointless. The gap between what you keep doing and what you believe about it is too uncomfortable to hold. So the belief moves. The person who complained in January that this was paperwork is, by spring, explaining to a client why the numbers matter, and by autumn he means it. The value arrived first, quietly, inside the repetition. Only then did it become a competition, to design the building with the lowest footprint, to be the office that knew the most. And that is the point at which we realised what the repeated action had done. It had not only changed what people believed. It had prepared them for a step nobody could have asked them to take a year earlier. An industry that has counted its own carbon on every project cannot pretend the number is someone else’s problem, and an industry that can no longer pretend is an industry you can mobilise. So together with colleagues I decided to use that leverage to try for systemic change. We asked the industry to join us in asking politicians for a binding limit on every new building, capped in line with the Paris Agreement, which in practice meant that within one year every project would have to cut its footprint by roughly half. More than 900 companies signed. It was the first time in history that an entire industry had come together not to fight regulation but to demand it. People called the campaign brave. It was not brave. It was timed. Had we asked for that limit before the counting began, we would have been talking to people who still called the whole thing paperwork, and we would have lost. The lesson is the order. The small repeated act comes first. It does the slow work of turning compliance into conviction. Only then does the big ask become possible, because by then you are no longer asking people to change. You are asking them to act on what they already believe. The scientists built the mechanism without knowing it.

That is the one-dollar experiment at national scale, and it tells you how to do the same for nature. We now have methods that can calculate the biodiversity loss behind a building, product by product, the way we calculate its carbon. I have helped develop one, and it is about to be published. Require that calculation on every project and you do not need to convince anyone to care about nature first. You install the small act. Producers count. Counting becomes conviction, conviction becomes competition, and somewhere in the repetition the value arrives. Then the legislation that makes people repair what they destroy has an industry ready to receive it, because the industry has already learned, with its own hands, what it was destroying.

Go back to the company with the trust problem, and the fix looks different now. Leaders who want trust cannot hold a second workshop. They have to remove the machinery that teaches distrust every day. Drop the second signature. End the weekly timesheet audit. Give the team a budget and let them spend it without an approval chain. What that does is hand people an action the old structure never allowed: deciding alone, and living with the result. The first time it feels like an experiment. After a year it is simply how things are done here, and the belief has arrived inside the repetition, the same way it did for the architects counting carbon. The space can also grow. A team that has used a small mandate well can be handed a larger one, so every repeated act inside the system earns room for the next. Trust stops being a slide and becomes a property of the structure, baked in the way suspicion used to be.

There is a part of this that is harder to sit with. If belief follows action even when the action was assigned, a system can produce almost any value in the people it holds, and from the inside that value will feel like character. The mechanism has no politics. The same design logic runs in the other direction, which is one reason authoritarian views can become normal in a country without anyone being argued into them. Nobody wakes up one morning a fascist. A feed puts a cruel joke about a minority in front of you and builds a button to share it. A talk show seats a climate denier next to the climate scientists and treats his views as one side of a debate. A government suspends one right for one group and asks you only to shrug. Each of those is a design decision, and none of them feels like a decision to the person on the other end. But remember the one-dollar student. Someone who has shared the joke or shrugged at the suspension has done something he cannot fully explain to himself, and the cheapest way to close the gap is to decide that it was fine. Do that every week for a year and it is no longer fine. It is what he thinks. The system never had to persuade anyone. It only had to make the small act easy and the consequence absent.

History is the warning here, though not a perfect analogy. In May 1928 the Nazi party took 2.6 percent of the German vote. In July 1932 it took 37.3 percent, and the ideology on the posters had barely changed. What changed was what people were able, and then allowed, to do: street violence, a “legal” path into parliament, conservative elites who thought they could use the movement and made each next step easier to imagine. Once in power, the regime redesigned daily life, and one of its first structural moves was to end proximity. Jews were pushed out of schools, offices, neighbourhoods and shops, step by step, until ordinary Germans no longer met them anywhere. That was cruelty, and it was also mechanism. It is very hard to demonise a person whose children play with yours, and very easy to demonise a category you never see. Comparing every right-wing voter with a Nazi would be careless and useless, but the structural lesson stands, and the conditions are widening again. In March 2026, V-Dem reported that nearly a quarter of the world’s countries were becoming more authoritarian, and Freedom House counted declines outnumbering improvements for the twentieth year in a row. A democracy can train people in authoritarian habits long before they take on an authoritarian identity.

So better arguments against the far right will not be enough. Better arguments are the workshop. Better manipulation is no answer either, because a system that hands people actions they cannot refuse produces obedience and calls it belief. Coercion can force compliance, and it can even produce conviction. What it cannot produce is a person who could have done otherwise. That is the difference between the two designs. The company that drops the second signature widens what its people can do, and the feed that builds the share button narrows what they will bother to question. One distributes agency and the other spends it. A new action becomes transformative when a person has the safety to try it, the freedom to refuse it, and can see afterwards that it produced something the old pattern never could.

Which makes this a design job rather than a debating job. The work is to decide, on purpose, which small acts a society makes easy to repeat. The Danish building code never asked anyone to care about carbon. It asked them to count, and the caring followed. A democracy that wants to push authoritarianism out has to find its own versions of the dropped signature. Build neighbourhoods mixed, so that meeting someone unlike you is an ordinary Tuesday. Keep the third spaces, the library, the sports club, the allotment, the street closed to cars, where people run into each other without having chosen to. Make participation visibly change something, so that showing up is rewarded by the world and not only by conscience. Let what the scientists find bind what the politicians decide, until following the evidence is simply a habit. Every one of those is a machine for producing citizens, exactly as the timesheet was and exactly as the segregated street was, with one difference. The person inside it has room to act, and the room grows as they use it.

Authoritarian systems concentrate agency and end proximity, and people rehearse domination until the values follow. A system built the other way hands agency out and brings people close, and what they rehearse there is reciprocity. The mechanism was never on anyone’s side. Which brings us back to the room where everyone agreed. They were not wrong to agree. People rarely surrender an old belief because someone told them it was wrong. They let go when life offers them a truer one, and life only offers it to people who are allowed to act. On Monday morning that room was offered a signature line. Real change begins when someone removes it. Distribute agency, build the systems that make prosocial acts easy to repeat, and the values follow.

illuminem Voices is a democratic space presenting the opinions of leading Sustainability Thought Leaders, their views do not necessarily represent those of illuminem.

The world needs sustainability knowledge. At illuminem, no interest group or shareholder can influence our work. Thank you for supporting our mission to make high-quality and independent sustainability information free for all. Every contribution helps. Thank you for donating today.