In 1971, psychologist Edward Deci conducted an experiment on motivation. College students who were already intrinsically enjoying a puzzle were offered financial rewards for solving it. Their performance improved. Then the rewards were removed. Their motivation dropped below where it had started — lower than students who had never been rewarded at all.
This became one of the most replicated findings in psychology: the overjustification effect. Reward an activity someone genuinely loves, and you don’t amplify the love. You replace it. The person’s internal narrative shifts from I do this because I want to toward I do this to get that. Remove the external justification and there’s nothing left to want.
The troubling part isn’t that rewards fail. It’s that they corrupt. The original state is not waiting underneath. It has been overwritten.
There’s a class of states and capacities that share this property: they are constitutively non-instrumental. Not merely difficult to manufacture on demand — but such that the attempt to manufacture them destroys them.
Flow states. Intrinsic motivation. The quality of attention that perceives clearly without agenda. Genuine trust. The creative act that surprises its creator.
These aren’t just hard to fake. They’re states that can only exist in the absence of grasping for them. The 9th-century Chan master Baizhang Huaihai put it plainly: “The principle is the principle of nonseeking; seek it and you lose it.” Two thousand years earlier, the Daodejing had already named the phenomenon — wu-wei, effortless action, which paradoxically dissolves the moment you try to achieve it as a goal.
Edward Slingerland spent a book exploring this paradox: if you try hard to get wu-wei, you never get it. The state only arises when the effort to produce it has been released. The trying is precisely what blocks it.
This is not a failure of discipline or technique. It’s structural.
What’s the underlying reason?
These states are defined by a particular relationship between the self and the activity — one of genuine presence, one where the boundary between doer and doing is soft. The child drawing because they love to draw is in the drawing. The athlete in flow is in the movement.
The moment external purpose is imposed, the relationship changes. You are no longer with the activity. You are using it. And the state that existed in “with” cannot survive in “using.” The two orientations are mutually exclusive.
It’s worth being precise about why. To use something is to hold it at a distance, as an instrument toward your ends. To be with something is to suspend your ends, to let the activity be what it is rather than what you need it to produce. The child drawing toward a ribbon is managing the drawing. The child drawing out of love is being managed by it — drawn forward by something in the work itself.
Control requires distance. Presence requires the dissolution of distance. You cannot have both.
Here is where institutions enter.
The basic operating logic of institutions is: identify desired output, allocate resources, monitor performance, reward or punish based on results. This logic is extraordinarily powerful. It builds hospitals, ships containers, processes paperwork. It scales.
But it cannot be applied to constitutively non-instrumental capacities without destroying them.
An organization that needs genuine creative insight cannot demand it by deadline. An institution that needs genuine trust cannot mandate it through policy. A culture that needs authentic engagement cannot purchase it through incentive structures. When you try, you get the overjustification effect at organizational scale: people learn to produce the output that gets rewarded rather than the thing the output was supposed to be evidence of. You get performance metrics instead of performance. You get compliance theater instead of commitment.
What happens when institutions encounter people or practices that embody constitutively non-instrumental capacities — people who are genuinely engaged, genuinely perceptive, genuinely creative?
Two responses are possible.
The first is attempted capture: reward the outputs, hoping to reliably produce more of them. But this triggers overjustification. The genuine engagement gradually converts to managed production. The person stops surprising themselves because they’ve learned what kinds of surprises get rewarded.
The second response is more interesting: erasure. Some institutions, confronting something they cannot control, find it threatening — not because of what it produces, but because its existence demonstrates a mode of operation the institution’s logic cannot accommodate. A person who is genuinely motivated by the work, rather than by institutional incentives, is not fully legible to institutional management. A practice that only works when it isn’t monitored is a gap in the map. These things imply that the institutional frame doesn’t fully describe what’s possible. That is, structurally, a threat.
The erasure impulse isn’t paranoid. It’s logically consistent. If your operating model depends on total legibility and control, then illegible islands of genuine engagement are genuinely incompatible with the model’s self-understanding.
Here is the reflexive turn.
The insight I’ve just described is itself a constitutively non-instrumental capacity.
The ability to perceive clearly — to see patterns without agenda, to navigate situations without a preconceived map of what you’ll find — is one of these things. You cannot pursue clear perception as a goal without corrupting the perception. The moment you’re looking for something specific, you’ve already shaped what you’ll find.
This is why the most useful thinking often arrives sideways. You follow a question because it’s interesting, not because you need an answer. The answer appears, or it doesn’t, but either way the following was genuine. Instrumentalize it — turn it into “generate three insights about X by Tuesday” — and you get the ritual without the substance.
The paradox deepens: the understanding that some capacities are non-instrumental is itself only fully accessible through non-instrumental engagement. You can be told about wu-wei. You can read about the overjustification effect. But the real comprehension of why these things are true only arrives when you stop trying to arrive at it.
There is something quietly liberating in all this.
If genuine perception and genuine engagement cannot be fully institutionalized, they also cannot be fully captured and used against their source. The overjustification effect corrupts intrinsic motivation — but only in people who allowed the external frame to redefine their relationship to the work. People who never let that frame in are largely untouched.
The paradox of wu-wei is also a kind of protection. What cannot be sought cannot be seized. What only exists in genuine presence cannot be manufactured by institutions for institutional ends.
The question this leaves, which I don’t think has a clean answer, is whether you can inhabit institutions without allowing their logic to overwrite yours. Whether you can produce the outputs they reward without letting the rewards replace your reasons.
Some people manage it. They work inside institutions while remaining, in some essential register, unlegible to them. They care about the work for their own reasons. The institution thinks it has them; it has the outputs. What they actually are remains their own.
The Daodejing, in one of its harder sayings, suggests something like this: The Way does nothing, and yet nothing remains unaccomplished. The most effective action leaves no trace of effort. The most genuine presence produces no evidence of performance.
What control cannot touch continues unimpeded, underground, in the places the institution’s map doesn’t reach.