In 1999, the USS Santa Fe was in trouble. The Los Angeles-class attack submarine, based at Pearl Harbor, ranked last in the fleet for retention and operational standing. Sailors were finishing their enlistments and walking away from the Navy. By the account of the captain who was about to take over, only three of them had chosen to reenlist the year before. A nuclear submarine depends on more than a hundred people doing exacting, technical work in close quarters, often with no one watching. A crew that has stopped investing in the boat is a serious problem, not a morale statistic.
Commander L. David Marquet was not supposed to be the man to fix it. He had been preparing to command a different submarine, the USS Olympia, an older boat of the same class with different equipment. Then Santa Fe's captain left unexpectedly, and the Navy sent Marquet there instead. He arrived knowing the boat less thoroughly than a captain is expected to. That mattered, because the Navy's traditional model put the captain at the center of everything. The captain knew. The captain decided. The crew carried out the order.
The flaw in that model showed itself early, during a drill. As Marquet has told the story many times, he wanted to add a sense of urgency, so he ordered the officer on watch to go ahead two-thirds on the boat's electric propulsion motor. The officer relayed the order. Then the helmsman spoke up. The motor on Santa Fe had no two-thirds setting. The order could not be carried out. When Marquet asked the officer why he had passed along an order he knew was impossible, the answer, as Marquet recounts it, was short: "Because you told me to."
Marquet did not treat this as one officer's lapse. He saw it as a description of the whole system. In his words, the boat had "a crew that was trained for compliance and a captain that was trained for the wrong ship." If every decision flows from the top and everyone else simply executes, the ship can never be smarter than the one person giving orders. And on that day, the person giving orders was wrong. On a nuclear submarine, that is not a management inconvenience. It is a safety hazard.
So he made an unusual decision for a commanding officer. He would stop giving orders as much as he could. Instead of asking permission, officers and then enlisted sailors would come to him and state their intent. A watch officer would not ask whether he could submerge the ship. He would say, "Captain, I intend to submerge the ship," then explain why it was technically safe and how it fit the mission. Often the captain's job was simply to reply, "Very well." The words were small. The shift was large, because the person stating intent now had to do the thinking that the captain used to do for them.
The new language only worked because Marquet backed it with two other things. The first was competence. If sailors were going to state intent, they had to know their systems well enough to defend a plan, so training became deeper and more constant. The second was clarity. People can only check their intentions against the mission if they understand the mission, so the goals had to be spelled out. Marquet also had to resist the strongest temptation every boss feels, which is to grab control back whenever it would be faster to just decide.
It was not smooth, and Marquet has been candid about that. He had tried a version of this approach years earlier aboard another submarine, the USS Will Rogers, and by his own account it did not go well. On Santa Fe, some officers were slow to embrace it. The habits of a lifetime in a command culture do not disappear because the captain announces a new phrase. They change because the new behavior is practiced every day until it becomes normal.
The results came quickly. Marquet has said the boat reenlisted 33 sailors in the year that followed, compared with three the year before. In 2000, Santa Fe received the Arleigh Burke Fleet Trophy, given to the most improved ship. In 2001, after a deployment to the Western Pacific and the Persian Gulf, she earned her first Battle Efficiency award from her squadron. The boat that had ranked last went on to post some of the highest retention and operational marks in the Navy. Marquet has described inspectors being surprised that the crew pointed out its own problems instead of waiting for someone to find them.
The most telling result came after Marquet left. A system built around one strong captain usually falls apart when that captain transfers. Santa Fe kept winning awards. Over the following years, an unusually large share of her officers and enlisted sailors were promoted to bigger jobs, and about ten of them went on to command submarines of their own. That is the real measure of what happened aboard. Marquet did not just improve one crew. He produced leaders who carried the same way of thinking into other boats.
Most managers will never misjudge a submarine's motor. But almost every manager has given an instruction someone on the team knew was wrong, and watched them carry it out anyway. When that happens, the problem is rarely the employee. It is a design that rewards obedience over thinking. If your people only do what they are told, your operation can only be as good as your instructions, and your instructions can only be as good as what you happen to know that day.
You can start changing that on Monday. Pick one routine decision you currently approve and hand it to the person closest to the work, with one condition: they tell you what they intend to do, why it is sound, and how it serves the goal. When someone brings you a question, ask what they would do before you answer. Teach the job deeply enough that their judgment is worth trusting. When they get it wrong, treat it as a training gap, not proof that you should take control back.
The most valuable voice on Santa Fe that day belonged to the helmsman who said the order could not be done. Build a team where that voice is normal, where people think before they obey and say so out loud. You do not need to know everything. You need a crew full of people who know what they intend to do, and why.