The United States has spent fifteen years preparing for a war it believed it did not want.

Every doctrine review, every CENTCOM force posture assessment, every State Department diplomatic cable—for a decade and a half, American strategic planners built a shadow self: the nation that could fight a sustained military campaign against Iran over the Strait of Hormuz. They wrote the rules of engagement, drilled the naval integration, hardened the bases. They calculated the cost in billions, the casualty estimates in thousands. They warned every administration that the Strait was the red line, the tripwire, the one contingency that could not be worked around. And then, when the contingency arrived, they ran the plan as written—because that is what you do when the ghost walking the halls for fifteen years finally steps into the room.

President Trump said this week that “every time Iran kills an American Soldier” going forward, “they will pay for that killing many times over.” The sentence reads as resolve to a nation that wants to hear resolve. But what it describes is a machine. A machine whose logic runs like this: we bomb, they bomb back, we bomb harder, they bomb harder still, and at no point does any human being stop to ask whether the premise—that the Strait can be fought to a resolution—was ever true.

The Associated Press reports that the daily strikes have collapsed an interim ceasefire, that gasoline prices are climbing, that the political calendar is squeezing every decision the administration makes.

That political calendar is not an abstraction—it is measured in dimes per gallon. Every upward tick at the pump eats into the margin Republicans need to hold the House, and the administration knows it. The ghost plan did not budget for domestic fuel prices because fuel prices were always someone else’s problem; now they are the problem, and the escalation logic does not have a subroutine for a constraint it cannot bomb.

Mona Yacoubian of CSIS put it plainly: the escalation reflects “a fundamental misreading of the Iranian psyche and a lack of learning” from the prior rounds of fighting. She is right, and she is stating a truth that cannot survive the logic it is describing, because if the administration had learned anything from the prior rounds, the current rounds would not be happening. The learning stops where the plan begins.

I have been watching this ghost since the summer of 2007, when I was a combat medic in Iraq.

We knew then what the next war would look like if we never found a way to stop wanting it. The Navy drilled for Hormuz every year. The think tanks published the same scenarios with updated numbers. The hawks and the institutional realists agreed on the stakes—and disagreement on anything is the only thing that can check a plan. When every important constituent in the national security apparatus agrees on the premise, the premise does not get tested. It gets executed.

And now it is being executed: daily bombing campaigns, shrinking diplomatic space, a ceasefire that collapsed because the logic of “one more round” was stronger than the logic of “enough.” The administration has spent the last three weeks ratcheting exactly this way—each escalation presented as the one that will break the impasse, each round failing to break anything, each failure producing the next round. The news reporting on it calls it a standoff. The operational facts call it a loop.

The Catholic tradition on war is sometimes accused of being unrealistically pacifist.

It is not. It has a just-war tradition that gives leaders a framework to look at what they are doing while they are doing it. The criteria exist for exactly this moment. Legitimate authority: check—Congress has not declared war, but the Authorization for Use of Military Force is a document that has been stretched so thin you can see through it. Just cause: the free passage of oil through an international waterway is a genuinely important interest—but just cause requires proportionality, and proportionality requires that the thing you are doing not be worse than the thing you are trying to prevent. Right intention: is the intention to restore the free passage of oil, or to make Iran “pay many times over”? Those are different intentions, and they point toward different kinds of escalation. Restoration of passage is bounded—you control the corridor, then you stop. “Pay many times over” is unbounded; the only stopping rule is submission. The administration has not chosen between them; it has let the punitive language drive the next bombing run while pretending the mission is still the corridor.

Last resort: the prior ceasefire was the resort, and it collapsed. Reasonable hope of success: here is the question the ghost plan does not allow. What does “success” look like in a daily bombing campaign against a country with deep defensive positions, a dispersed missile capability, and no interest in surrendering the one piece of geography that gives it leverage? If you cannot describe the end state in concrete terms—not “breaking the impasse,” not “getting them to relent,” but the actual conditions on the ground on the day the bombing stops—then you do not have reasonable hope of success. You have a plan for escalation that has become its own reason for continuing.

The ghost is real now, and it is resistant to argument because it is not held in place by arguments.

It is held in place by institutions that wrote the exercises, officers who were trained in the plan, contractors whose revenue projections depend on the conflict continuing, and the political calculus that says stopping is weak while continuing is strong. The ghost does not have to persuade anyone anymore. It is already in the room.

The tradition calls this structural sin—pecado estructural—a condition in which the harm is not the product of any single bad decision but of the accumulated weight of decisions nobody stopped to unmake. The Romero who told the soldiers to stop the repression did so because the repression had become its own reason for continuing. The structure of the violence had taken over the purpose of the violence. And the only thing that could break it was someone standing at the microphone and calling it by its name.

The Strait of Hormuz is not a ghost anymore. It is real. It is killing American soldiers and Iranian sailors and filling the water with debris that will take decades to clear. The only thing that is still a ghost is the possibility that someone in the chain of command will stop and say: this is not working. The thing the plan promised—that the other side would relent—has not happened after fifteen years of preparation and weeks of daily bombing. It is not going to happen. What are we doing?

The answer to that question is not more bombing. It is the work of building a diplomatic off-ramp before the escalation curve becomes a straight vertical line. It is the admission that the ghost was never a war plan at all—it was a substitute for one, a way of having prepared without having to decide whether to fight. That time is over. The decision has been made. The only question now is whether the decision can be unmade before the casualties reach a number that the plan’s authors never had to look at in the exercise.