A geography of leverage
2 March 2026. Midnight.
No tankers left on the automatic identification system. Not because a fleet had been sunk. Not because mines had been laid or a naval battle had taken place. Because drones are cheap enough — and because insurance companies in London calculate faster than admirals plan.
That was how the strait closed.
Anyone who found this surprising had been reading the wrong text.
What the system always knew
There is a persistent belief in geopolitics that power reveals itself in action. Visible, measurable, military. It is a useful belief. It is also, most of the time, wrong.
Power sits more often in what does not happen. In the background, unexercised but structurally available, shaping decisions not through visible moves but through the anticipation of moves that never have to be made.
The Strait of Hormuz was exactly that — for decades. Not because it was secured. Because nobody had to test it.
Before 28 February, roughly twenty million barrels of oil moved daily through thirty-eight kilometres of water between Iran and Oman. The movement was routine, almost uneventful. The system appeared to confirm its own resilience.
That appearance was doing more work than it seemed.
What presented itself as stability was a system that continued to function as long as nobody tested how little disruption it could absorb. Stability as restraint. Not as condition.
Restraint as strategy
Iran did not need to close the strait to shape outcomes.
Closing it would have collapsed ambiguity into confrontation. It would have invited a level of response difficult to control and costly to absorb — including for Iran itself. The leverage would have been spent in a single move. The advantage of operating below the threshold of open conflict: gone.
That was the strategy of the preceding decades. Apply pressure without exhausting it. Send signals without formalising them. Disruption not as an end, but as a possibility — permanently available, never deployed.
Then came Operation Epic Fury. And with it, Khamenei’s death.
A state that calculates behaves differently from a state that is trying to survive.
The grammar of selection
Since the war began, oil flows through the Strait have collapsed by more than eighty percent. Lloyd’s List Intelligence counted sixteen oil tankers in the first two weeks of March — against more than a hundred vessel passages daily before the conflict. But the selection matters more than the number.
On 5 March, the IRGC announced the strait would remain closed — but only to ships from the United States, Israel, and their Western allies. Turkish vessels were permitted passage. Indian gas carriers. A Saudi tanker carrying a million barrels bound for India.
And today, 25 March: Iran tells the UN that non-hostile ships may transit the strait — in coordination with the competent Iranian authorities.
In coordination with the competent Iranian authorities.
That is not the language of a closed war port.
It is the language of controlled access.
Who passes is no longer determined by geography alone, but by designation. Friendly ships move. Others wait — alongside more than a thousand vessels anchored outside the Gulf of Oman.
The strait is not closed.
It is being filtered.
Control as revenue, not closure
What emerges from this is not simply disruption, but a different model of control.
The ability to deny access is one form of leverage. The ability to grant access selectively is another — and often a more durable one. It allows pressure to be applied without collapsing the system entirely. It preserves the flow, while changing the terms under which that flow takes place.
Control, in this sense, does not require permanence. It requires recognition.
Once passage becomes conditional, it can be priced — formally or informally. Not necessarily as a declared fee, but as a function of alignment, coordination, or acceptance of authority. The mechanism does not need to be institutionalised to be effective. It only needs to be acknowledged by those who depend on it.
A weakened, post-Khamenei Iran is unlikely to stabilise such a system immediately. That is not a model of governance. It is a condition of strain.
But conditions change.
Capabilities, once demonstrated, do not disappear. The infrastructure of selection — the ability to distinguish, permit, and deny — remains available to whoever inherits it.
Exposure without ownership
Iran has taken a page from the Houthi playbook: selective access for friendly ships, restriction for others. That may appear improvised. It is not.
China imports a substantial share of its energy through this passage. Militarily and politically, its role in securing that flow has been limited. Chinese state carriers initially suspended new bookings for Middle East routes. Then individual Chinese-linked ships transited — and one was hit by shrapnel despite broadcasting its affiliation.
Exposure, in this case, does not translate into control.
It rarely does.
A configuration in which one actor guarantees access, another retains the capacity to threaten it, and a third depends on its continuity without assuming a corresponding share of responsibility can persist — for a time.
It tends to become visible only when it is tested.
What the silence was always saying
The strait did not need to close to shape the system.
For decades its significance lay in remaining open under conditions that were neither fully secure nor entirely predictable. What appeared as continuity rested on a shared understanding — unspoken, but reflected in insurance premiums, military deployments, and diplomatic positioning.
“One of the core conundrums of this conflict is the Iranians have real leverage with this, and there’s not an obvious fix for it” — a US intelligence official, this week.
No obvious fix.
That is an assessment from the most capable military apparatus in the system. About thirty-eight kilometres of water.
The strait was never silent.
It was simply not necessary to listen.
Christopher Angel — currently posted, officially silent, and biting anyway.