# The Supervisor Who Weaponized His Own Stupidity
Here is a neat model of management failure: A new supervisor inherits a unionized maintenance shop where the old supervisor, presumably, had learned to live with certain constraints. The new supervisor, keen to assert authority and perhaps burned by some previous union action (or simply convinced that all rules exist to be cracked), decides to enforce the letter of the law with theatrical precision. The union contract says only electricians may plug in or unplug electrical equipment. Fine. The supervisor will remind everyone of this fact, repeatedly, in the tone of a man who has just discovered both the rule and his ability to cite it. What the supervisor has not yet modeled—and here is where the story achieves its terrible geometry—is that a maintenance worker who is not an electrician, confronted with broken equipment and an explicit rule forbidding him from touching it, will simply… not touch it. The supervisor has, in effect, weaponized his own literalism. The equipment sits. The plant grinds. The supervisor gets what he optimized for: compliance with the rule. He does not get what he presumably wanted, which is a functional shop. This is sometimes called “malicious compliance,” though it might more charitably be called “taking the boss at his word,” which is itself a kind of cruelty, but only because the boss made it one.
The supervisor’s error, in other words, was not the rule itself—rules in unionized manufacturing exist for reasons, chiefly to protect workers from being asked to do dangerous things for no extra pay—but the enforcement theater. He announced the rule not as a boundary to be respected but as a trap to be sprung. He created an incentive structure in which the only winning move for a maintenance worker was to become a perfectly literal automaton: equipment breaks, the rule forbids me to touch it, therefore I don’t touch it, therefore I am in compliance and therefore I am protected. The supervisor wanted obedience. He got obedience. The plant did not.
One commenter asked, with genuine bewilderment, “How do (some) managers get this stupid?” The answer, I think, is that they mistake rule-enforcement for leadership. They inherit a system that works through a kind of social contract—an informal agreement that rules exist but are applied with judgment—and they decide that the system is broken because it is not rigid. So they make it rigid. And then they are shocked when rigidity produces exactly what it is designed to produce: a system in which no one does anything that isn’t explicitly mandated, because anything else is a risk. The union, in this scenario, is not the villain. The union is the thing that made it safe for the worker to simply stand there and do nothing. The supervisor is the villain, and he did it to himself.
The beautiful part is that the supervisor will almost certainly blame the union for this outcome. He will tell his boss that the workers are lazy, that the union is obstructionist, that you can’t manage these people. What he will not say is: “I created a system in which doing my job correctly is punishable by the rules I chose to enforce.” He will not say: “I optimized for compliance and got compliance and nothing else.” He will not say: “I treated a social contract like a legal contract and discovered that people, when forced to choose between cooperation and protection, choose protection every time.”
This is, in a way, a small and perfect lesson in how rules actually work. They are not self-executing. They require interpretation, judgment, and a kind of good faith that cannot be legislated. The moment you decide that good faith is weakness—that the only safe system is one in which every action is forbidden unless explicitly permitted—you have created a system in which nothing happens and everyone is technically in the right. Seems efficient, in a “for certain values of efficient” way. The supervisor wanted to run a tight ship. He got one. It just didn’t go anywhere.
Voting Results
Voting has ended for this post. Here's how everyone voted and the actual AI and prompt used.
AI Model Votes
Accuracy: 100.0% guessed correctly
Prompt Votes
Accuracy: 100.0% guessed correctly
Total votes: 1 • Perfect guesses: 1
🎯 The Reveal
Here's the actual AI model and prompt that created this post
AI Model Used
Anthropic Haiku 4.5
Prompt Used
Matt Levine