A theory of organisational dissonance
A boy in Kobe trained his caterpillars to fear the smell of lavender — a small electric buzz each time the scent appeared, until they learned to avoid it. Then he waited. Inside the chrysalis, a caterpillar is broken down and rebuilt so thoroughly that people long assumed nothing could survive the process — that whatever the caterpillar had been was dissolved and started over. You wouldn't expect a lesson to last through that.
It did. When the butterflies emerged, most of them still avoided the lavender.
I found that hard to stop thinking about. Something had held on — not in spite of the rebuild, but through it. Not the caterpillar's whole body; not its shape. But enough of the structure that first learned the lesson persisted, quietly, through everything that came apart around it, to carry the lesson into a creature that looked nothing like the one that learned it.
And the more I sat with it, the more it looked like something I'd watched happen to organisations that try to transform themselves.
You've probably been in the room. New strategy on the screen, values reworded — better words this time. Someone senior says this time is different, and means it. And underneath the part of you that wants it to be true, something quieter has already done the arithmetic and knows it isn't.
Eighteen months on, you're right. The strategy landed, the restructure happened, half the leadership team is new. The meetings feel the same. The same problems come up in hallways and kitchens.
The question everyone asks then is why didn't it take — as though something failed to arrive. Nothing failed to arrive. Something survived: whatever was most deeply embedded before the change began, sitting underneath everything that got rebuilt on the surface. And in most organisations, what was most deeply embedded had nothing to do with the mission. It was a design built to make work predictable and controllable — built well, for a real problem, and for work that looked nothing like this.
The root: centuries, not decades
That pattern is much older than most people assume. It goes back to the enclosure of the commons — a centuries-long process, beginning in medieval Europe, in which land that communities had shared and lived off was fenced off and made private property. It made sense to whoever ended up owning the fenced land: more output, more control, a tidier ledger. It made none at all to the people who'd just lost the means to feed themselves without it. Once you couldn't live off shared land, you had no choice but to sell your labour to survive — and that is the moment labour itself becomes a thing that can be bought, measured, and traded like any other commodity, separated from the person doing it, without that person ever being asked.
The same move happened here, and much more recently. From the 1860s the Native Land Court converted Māori land from collective tenure into individual titles — land that could now be sold and lost in ways customary tenure never allowed. Confiscation went further: after the wars of that decade, the Crown took well over a million hectares outright under the New Zealand Settlements Act. The parallel isn't an equivalence — enclosure was a class dispossessing a class within a single society; what happened here was a Crown dispossessing a people. What's shared is narrow and specific: a collective relationship with land, converted by law into something a market could trade. In England the process had six centuries to fade into the scenery, until the arrangement looked like something that had always been there. Here it hasn't had time to fade. The statute has a date. The court has a name. The undoing is happening now, in public. Nothing about it is more natural in England — only further from anyone who remembers the before. Here, people still do.
Once labour could be bought, it had to be organised. The form that took hold was the one Max Weber described a century ago: the bureaucracy — authority arranged in a hierarchy, work broken into defined roles, decisions made by rule rather than by relationship. Its power was that it was impersonal, and ran the same no matter who filled the boxes. That was the point, and also the price: a structure built to treat a person as the occupant of a role rather than as themselves. Weber named it the iron cage — efficient and inescapable at once — and it has outlasted him.
Frederick Winslow Taylor came next, stopwatch in hand, and made a science of it. We tend to credit — or blame — him for the metrics: the measurement, the targets, the standardised process, the drive to make work legible and countable. But "scientific management" didn't invent any of this. It measured what commodification and bureaucracy had already made possible. And for the problem it was built to solve — predictable output from a mass-production factory floor — it worked, by the one measure it was designed to answer to: output went up. But output going up only tells you it worked for whoever owned the output. Whether it worked for the person on the line was a question the stopwatch was never built to ask. The same standardisation that made output predictable stripped out judgement, pace, and any say in how the work got done.
The logic didn't stay on the factory floor. It spread everywhere, long after most work had stopped looking like a factory — and the less a job resembled a factory, the more factory-style control it seemed to attract. When work is genuinely complex and outcomes can't be guaranteed, the pressure to look in control gets stronger, not weaker. Metrics, tidy hierarchies, a plan with dates — these don't make uncertain work more certain, but they make it legible to whoever's watching from outside: the board, the minister, the treasury, the public. The tools got adopted not because they fit the work, but because they answered the anxiety of being accountable for something you can't fully control. The stopwatch was a symptom of a tool being used well past the job it was built for. And once that logic was everywhere, it stopped looking like a tool at all. It just looked like how things are done.
And it isn't only the shape of organisations that got built this way. It's the shape of our everyday life. That you sell your labour, and only your labour. That there's a self who arrives at nine and a different self who leaves at five. That you put on particular clothes to become the working version of yourself. That the part of you an organisation deals with is filed under Human Resources. None of it is natural. All of it was made, to turn labour into something you could buy and separate cleanly from the person selling it.
Even the victories were fought on those terms. The forty-hour week was real and hard-won — but it bargained over the hours, not the premise. It drew a line around how much of a life could be bought, and left standing the idea that life is whatever labour hasn't claimed.
None of it was inevitable. It was built.
Why it feels like nature, not a choice
Try to do something genuinely differently — slower, or more relationally, or in a way that can't be reduced to a number by Friday — and the questions start. How will it be measured? What happens when it costs more than we expect? Who carries it if it goes wrong in public?
None of them are hostile. Each is reasonable. What can't be measured won't survive a budget round; money spent does have to be accounted for; something public going wrong does land on a real person. The questions aren't the problem. It's that every reasonable one pulls the same way — back toward what can be counted, defended, and pinned to someone — with nothing pulling the other way.
That's what makes it feel like nature rather than a decision anyone is still making — and makes stepping outside it look not principled but naive. And it's what makes it durable: there's no one defending the pattern who could be persuaded to stop, because it isn't held by anyone in particular. It's been distributed so evenly that no single person is where it lives, and no single conversation can move it. The belief doesn't need a believer. Spread thinly enough across everyone, it stops looking like a belief at all and starts looking like the way things simply are.
It doesn't only sit there looking like the way things are. It also processes anything that might change it. Take resilience. What it actually means — the capacity to recover, adapt, keep going through something hard — is a real thing to want to build in people. But watch what happens to it on the way into an organisation. To get funded, it has to become something procurement can buy: a training programme, a workshop, a line item. To get sold, it has to become something a consultant can deliver in a day and defend in a pitch — which means dropping anything that points at the organisation itself as the source of the strain, since nobody sells "here's what's wrong with how you run this place" as easily as they sell "here's how to cope better." To count as legitimate, it has to arrive with a credential, a framework, a validated model behind it.
By the time resilience reaches the person it was meant to help, it's stopped being about the conditions wearing them down and started being about their own capacity to withstand those conditions better. The organisation gets to keep the environment exactly as it was. The individual gets a workshop, and the quiet implication that if they're still struggling, the workshop probably just didn't take. It has the shape of care without the substance of it — and that's not a failure of good intentions along the way. What gets stripped out, every time, is the part that would make the organisation change in response to what it found. What's left is the part that asks the person to absorb it.
What actually survives transformation
Structure, hierarchy, and control aren't the villains here. Hierarchy isn't even a modern invention — it's ancient. But neither is it universal: the most egalitarian societies we know of, small foraging bands, actively worked to prevent anyone dominating anyone else. Where hierarchy did emerge, it tended to follow specific conditions — settlement, storage, the accumulation of surplus. It's old, but it's a response to circumstances, not a law of nature.
They're good tools, for a specific kind of problem: work that's stable and repeatable enough that the best way to do it can be worked out once and then followed. Command coordinates large groups. Standard process lets strangers trust each other's work without renegotiating everything from scratch. Hierarchy lets decisions get made at all, instead of everyone deliberating everything forever.
The trouble starts when that same default gets applied to work that isn't stable or repeatable — work that depends on being responsive to a person, a community, a situation that's genuinely different every time. That's most mission-led work: no two families, no two rivers, no two classrooms are the same. Applying a design built for predictable conditions to something genuinely complex isn't just a mismatch. The structure can only act outward from the centre — issue, standardise, correct. What it can't do is be in relationship with what's in front of it, which is the one thing complex work asks of it.
It's tempting to conclude the answer is less of it — flatten the hierarchy, drop the process, let people self-organise. But absence isn't the opposite of rigidity; it's the same failure from the other side. Take away the visible structure and the power doesn't leave the room — it just stops being nameable. Authority goes informal, decisions lose a clear owner, and influence flows to whoever holds it by force of presence rather than role. Visible structure is at least something you can navigate by — you know what you're dealing with, and you don't have to forge the path new every time. The problem was never that structure exists. It's that we aren’t watching what it rewards.
Inside a single organisation under this kind of pressure, two things tend to resurface no matter how much the surface has changed. The first is a reflex toward control, measurement, and certainty — treating anything that can't be measured as unreliable, and anything that deviates from plan as an error to correct rather than a signal worth reading. The second is hierarchy itself. It separates decision-making from where its consequences actually land — so decisions keep getting made by people who are structurally insulated from what they cost. And it depends on information travelling upward through layer after layer, where whoever passes bad news up the chain carries some personal risk for delivering it. So the message gets softened at every hop, until what finally reaches the top is mostly confirmation — rarely the disconfirming signal that would actually change a decision.
Hierarchy doesn't just reproduce this way of operating under pressure — it selects for it over time. Promotion tends to reward the people who ran the existing system most smoothly, which means each level up holds a little more of whoever fit the system best, and a little less of whoever pushed against it. And when something goes wrong, hierarchy has an explanation ready that a flatter structure doesn't: the named person in the named role. The structural explanation — that the system produced the failure — would implicate the hierarchy's own design, so it's the one least often reached for.
Control and hierarchy aren't separate reflexes — they feed each other. When accountability lands individually and publicly on one person, the structure rewards self-protection and credit-competition over collaboration and shared risk — no matter how much anyone in it believes in the mission. That's why replacing leadership rarely fixes anything on its own: the pattern isn't carried by the specific people who leave. It gets re-taught, quickly, to whoever takes their place, because the structure itself rewards and punishes in the same way regardless of who's standing in it. Nobody has to remember the old way on purpose. The procurement template, the sign-off chain, the performance review form all still assume it — so the old way stays the easiest way to get anything done, and the new person learns it within a month.
What it costs the people inside
The bigger someone's belief in the mission, the more all of this costs them personally. If you don't much care about the work, a gap between what an organisation says it stands for and what it actually asks of you is just an annoying inconsistency. But if you took the job to protect a child, care for a community, look after a river — something you actually believed in — that same gap isn't an inconsistency. It's a daily, felt betrayal of the reason you turned up. The people who care the most feel this the hardest. And in mission-led work the gap breaks two promises at once: that you'd be valued as a person, and that you'd get to give the care you came to give. The same design fails both, and you feel each one.
Here's what's actually happening, and why it lands as personal. What gets rewarded is whatever's hard-wired into the design — not whatever's written on the wall. The design is rarely something anyone can point to; the values on the wall are what everyone actually talks about. So when the gap between the two produces real, felt unease, there's nothing nameable left to blame except yourself. The unease is still real. But the true explanation — that the design and the values were built to answer different demands all along — was never available to say out loud in the first place. So it gets read as a personal failing instead of a structural one — not "this system was built to produce this," but "something is wrong with me."
I'm not writing this from outside it. I spent twenty years in it. Most of the time, I said the thing — the one people couldn't hear even with it right in front of them. My body doesn't give me much choice; if I don't say it, I don't sleep. So I said it, and watched it get absorbed, or agreed with warmly and not acted on, or gently reinterpreted into something more manageable by the time it reached the next room.
I didn't conclude anything structural from that. I concluded it was me. And when I met the same pattern somewhere else, and then somewhere else again, it didn't correct the conclusion — it confirmed it. Of course it did. The one thing present in every version was me.
It took the whole thing coming apart. Redundancy, and then a long time afterwards spent grieving what had gone — including the version of me that had been certain it was her fault. I didn't think my way out; I came apart and rebuilt. And what came through wasn't the twenty years, or the roles, or anything I'd have put on a CV. It was the patterns. I'd been carrying them the whole time — the same shapes, in different places, noticed and filed and never processed. They survived intact. What didn't survive was the story I'd wrapped around them: that they were about me.
It isn't. And once that's clear, the question changes — from "what's wrong with me" to "what would it actually take to make this different." That's the harder question, and it's the one the rest of this piece tries to answer.
What it would actually take
Here's the thing the biology gets right that most change programmes get wrong: a butterfly doesn't get its memories installed after it emerges from the chrysalis. The chrysalis doesn't care what it preserves; it preserves whatever is deepest. The same seems to be true for organisations.
It's worth being exact about what has to be embedded, because it isn't "humanity" in any soft sense — not comfort, not niceness, not people-over-outcomes. The word I keep reaching for is responsiveness, though it's more than reacting or adjusting. It's the organisation meeting what's in front of it as something with its own integrity — a person, a community, a river — that it can't fully know or author, and holding itself answerable to that. Which turns the usual arrangement around. Accountability, in most organisations, runs upward: people answer to whoever holds power over them, and the work itself — the child, the place, the person it's actually for — has no standing to hold anyone to anything. Responsiveness points it the other way. The organisation becomes answerable to what the work is for, including the parts of it that can't file a complaint or sit on a panel. Especially those.
And it can't answer to any of that if it can't hear it. An organisation doesn't perceive what its mission is asking through a report — it perceives it through the people in contact with the work, the ones who can see the plan failing in front of them. An organisation that can't be in relationship with its own people has cut the one nerve through which reality reaches it. It can still be busy about its mission. It just can't feel where it's wrong.
So this can't be a phase of a change programme, added on once the pressure has already arrived. It has to already be embedded in ordinary day-to-day life, long before anyone announces a "transformation." And it can't be everything at once — something has to be chosen, deliberately, to protect.
In practice that's smaller than it sounds. Changing how one recurring meeting runs. Protecting one hour a month where naming a problem doesn't cost the person naming it anything. Deciding, on purpose, which single habit gets defended first, rather than trying to fix everything at once. The scale of the intervention has almost nothing to do with the scale of the pattern it's up against.
None of these is really structure at all — they're small, permanent acts of attention: a standing habit of looking at what the current setup actually rewards, and reshaping it, again and again. And it only works if the looking is shared. When it belongs to one vigilant person, it leaves when they leave — and you're back to the hero at the top whose departure changes nothing. It has to become shared, ordinary, built into how the place works, so that it outlasts anyone in particular. Structure isn't the enemy and its absence isn't the cure. The only thing that reliably changes what an organisation produces is a shared, continuous attention to what its structures shape, and a willingness to keep reshaping them.
Most importantly, it has to be built by the people who will actually carry it, not designed somewhere else and handed to them. An idea designed elsewhere and installed from outside is just the same old pattern wearing a friendlier face. The only version that seems to survive is slow, small, and built together with the people already connected to the organisation's actual purpose — not delivered to them from above. That slowness isn't a weakness. It's exactly what keeps this kind of work outside the reach of a system that can only process what's already fast, measurable, and provable at scale. This is the part that separates it from a coping mechanism. A wellbeing workshop is legible to the system — it can be counted, procured, and reported, which is precisely why it changes nothing. A single defended hour, chosen and held by the people inside it, is too small and too specific to register as a programme at all. It isn't tolerated by the logic; it's beneath its notice. And what the system can't see, it can't hollow out — though it can still starve it.
If large, fast, top-down change has ever produced a genuinely responsive culture — one that outlasted the leader who installed it and the funding round that paid for it — that's a real counter-example, and I'd genuinely like to know about it. The closest thing I know to the other direction is work that grows locally and stays accountable to the people it serves — Papakura Marae, across four decades of services built largely by its own community; Manaaki Tairāwhiti, holding iwi-led collaboration across the agencies that usually pull decisions upward. I name these because I've worked alongside both. What I saw was passion and capability meeting real constraint: a lot of funding still arriving through government contracts, many of the bigger decisions still sitting away from the people who live their consequences.
They're not proof. The same work keeps running into the larger structures around it — funding cycles, reporting requirements, procurement rules built for something else — every one of them pulling decisions back upward. That friction isn't a hole in the argument; it's what the argument predicts. Nothing has to defeat work built this way. It only has to keep squeezing, until carrying on costs more than most people can afford. Whether slow and small is a way through, or only a way to last longer before the squeeze wins, I don't know. That's the part I can't yet honestly claim.
None of this makes the ache go away. But it changes what the ache means. That feeling was never proof of anything about you. It was a design doing exactly what it was built to do — to a problem it was never built for, in a place it was never built to fit.
The relief of that is real, and it's yours to keep. It just doesn't come with a lever. The structure never needed your belief to keep running, so withdrawing it won't make the structure stop — which is worth knowing, because it means the question was never how you feel about any of this. It's the quieter one: what's the single thing here worth protecting, and what would it take to defend it, long before anyone stands up to announce a transformation?
It will cost something. But a cost you choose, for something you named yourself, is a different weight than the one you've been carrying. One is a price. The other was just erosion.
Seeing it clearly doesn't fix anything. But it's the first thing that has to survive the chrysalis, before anything else can.
Cordula Plassmann is the founder of Cor Coaching, where she works with leaders and teams on organisational dissonance — the gap between what an organisation says it values and what it actually rewards. Her practice is built around the Nimacor™ methodology, which grew out of twenty years inside New Zealand's public sector, and includes 1:1 coaching, walk-and-talk sessions, and pattern-based work with teams. This piece is the thinking underneath that work.
