The conventional read on the past week has been that the United States and Iran are drifting toward war. That framing is too soft. They are already in one, they have been in one since late February, and the June ceasefire that most observers treated as a durable off-ramp collapsed so quickly that the more accurate description is that the fighting never really stopped, it just quieted down long enough for markets to exhale before it resumed.
On Sunday, US Central Command announced another American service member had been killed, this time in northern Iraq during a controlled detonation of unexploded ordnance from a downed Iranian drone.
The remains of that missing service member were reportedly recovered over the weekend.
An escalation ladder that no longer has clean rungs
Military analysts sometimes talk about escalation as a ladder, with each side climbing one rung at a time and choosing whether to keep going. What has happened between Washington and Tehran over the past eight days does not fit that model. The rungs have blurred together.
Both sides have hit civilian infrastructure. Both sides deny it counts as a war crime. Both sides are right that international law makes exceptions when civilian objects are used to support an enemy’s war effort. The disagreement is not really legal. It is about who gets to decide what counts as military use of a bridge, a desalination plant, or a power station that serves several million people.
The nuclear site nobody knew was on the list
Among the more consequential developments over the weekend was confirmation that strikes hit the construction site of a planned nuclear power plant in Darkhovin, in the country’s south-west.
The absence of nuclear material is what makes the strike legally survivable and politically explosive at the same time. Washington can argue that no radiological event was possible. Tehran can argue that the US bombed a civilian energy project, undeclared as a military target, and that the precedent applies to every future nuclear facility Iran builds.
That precedent is the point.
Why the ceasefire never had a chance
The interim agreement signed in June was designed to freeze the fighting long enough to negotiate something more permanent.
The polite version of what happened is that trust broke down. The less polite version is that neither side ever agreed on what the ceasefire was for.
Washington treated it as a pause to consolidate gains and pressure Iran into a broader deal covering its missile programme and regional proxies. Tehran treated it as recognition that the US could not sustain an open-ended air campaign against a country of 90 million people without either drawing in ground forces or accepting economic costs its allies would refuse to absorb. When both interpretations came into contact with events on the ground, the agreement dissolved.
What is left is a conflict with no diplomatic scaffolding around it.
The Strait of Hormuz problem
Roughly a fifth of the world’s oil passes through the Strait of Hormuz on a normal day. The strait is not normal right now. Shipping traffic has largely stalled, insurance premiums for tankers transiting the waterway have jumped.
The result is a shipping environment in which no captain, no insurer, and no charterer can predict what happens tomorrow.
Fuel prices have moved accordingly. So have prices for food and other goods in countries far from the fighting, particularly in parts of Africa and South Asia that rely on Gulf shipping lanes for imports. The war’s second-order effects are being paid by populations with no stake in the outcome and no leverage over either side.
The 17 American deaths and what they signal
American military deaths in the region have followed a pattern that is worth reading carefully. Most have come not from direct engagements but from drones, indirect fire, and, in at least one case now, the disposal of unexploded ordnance. This is the texture of a war fought largely at range, with limited ground contact, and it produces a specific kind of casualty profile.
It also produces a specific kind of political pressure. Deaths from drone attacks on isolated bases are difficult to prevent without either withdrawing forces or expanding the target set inside Iran. The first option is politically impossible for the current administration. The second option is what has been happening.
Each American death raises the temperature at home. Each Iranian retaliation reinforces the domestic case in Tehran that the regime is defending sovereignty against a foreign aggressor. Neither leadership can afford to look like it is absorbing losses without responding.
That is the ratchet.
Regional partners caught in the middle
The regional response has been instructive. Jordan, where the latest American deaths occurred, now finds itself hosting a war it did not choose. The Gulf Cooperation Council, which includes Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates, has separately accused Iran of violating international law by targeting civilian infrastructure.
None of these countries want to be in this war. Several of them have spent the past decade carefully building economic and diplomatic ties with Tehran while maintaining security relationships with Washington. Those balancing acts are now under stress no one designed them to withstand.
The cost of hosting American forces is going up. The cost of not hosting them is also going up, because the region’s security architecture has no obvious substitute.
Israel’s proximity to a war it has not joined
Israel has not directly entered the current round of fighting, but the geography does not care. The Israeli military has been intercepting fragments to prevent damage inside its borders. Israeli officials have warned that any spillover across the Jordanian border risks pulling Israel in.
What that would look like operationally is not a mystery. Israeli warplanes have conducted long-range strikes inside Iran before. What is less clear is whether Israeli entry would broaden the war beyond the current US-Iran axis or simply intensify the existing exchange. The honest analytical answer is that nobody knows, and the people who claim to are usually selling something.
What the market is pricing versus what the region is experiencing
Global markets have moved on the war, but not as dramatically as the ground picture might suggest. Oil is elevated. Defence stocks are up. Shipping equities are volatile. But equities broadly have not repriced for a full regional conflagration.
This is partly because market participants have seen versions of this movie before, going back to the 1980s tanker wars, and have learned that Middle Eastern conflicts often stay contained even when the rhetoric suggests otherwise. It is partly because algorithmic trading dampens the kind of sustained repricing that would follow if human traders sat down and worked through the implications of Hormuz closure lasting several months.
And it is partly because the alternative — pricing in a real regional war — is so disruptive to portfolio construction that most large allocators would rather wait for the news to force the move than lead it.
The gap between market pricing and regional reality is itself a risk. When it closes, it usually closes fast.
The information environment
Both governments are running aggressive information operations. Iran’s health ministry has said at least 50 people have been killed and more than 500 wounded in US strikes, figures that the government has not broken down by military versus civilian casualties. The US has released selective footage of strikes carried out by fighter jets and Tomahawk cruise missiles, framed to emphasise military targeting and precision.
Independent verification is thin. Journalists have limited access to strike sites in Iran. Satellite imagery fills some of the gap but not all of it. What emerges is a war being fought partly through the images each side chooses to release, with truth-in-the-middle assumptions doing a lot of work that they probably should not.
Readers should be sceptical of any specific casualty figure from either side, particularly figures released within 24 hours of an attack. The numbers tend to converge later, and they usually converge somewhere between the two initial claims.
What would have to happen for de-escalation
The mechanics of ending this war are not mysterious, they are just politically difficult. Both sides would need to accept a face-saving formula that lets each claim it did not back down. Both would need a third party credible enough to guarantee terms neither trusts the other to honour. And both would need domestic political cover for making concessions that will be attacked at home.
The June ceasefire failed because none of these conditions were fully in place. Some version of them will need to be built before the next attempt has a better chance.
Candidates for third-party guarantors are limited. The European Union has diminished leverage. China has commercial interests but limited security credibility with Washington. Russia is a party to too many other conflicts to be trusted as a neutral. Turkey and Qatar have played mediation roles in previous rounds but both have exposure that complicates their impartiality now.
That leaves the awkward possibility that de-escalation, when it comes, will come not through mediation but through exhaustion. Wars often end that way. It is a bad way.
The domestic politics of a war nobody voted for
American voters did not go to the polls in 2024 expecting an active war with Iran by mid-2026. The current administration has framed the conflict as forced by Iranian aggression against US troops and allies, and public polling has broadly followed that framing. But polling can shift quickly when the death toll climbs, when videos of coffins returning home circulate, and when fuel prices at the pump make the war visible in daily life.
Iran’s domestic politics are harder to read from the outside. The regime has strong incentives to project unity and grievance, and it does. Beneath that surface, the strain on the country’s economy from renewed sanctions and infrastructure damage is real. How that strain translates into political change, if it does, is not predictable on any useful timescale.
Both governments are betting the other cracks first. Both may be wrong.
What to watch this week
Three indicators matter more than the rest. First, whether shipping traffic in the Strait of Hormuz resumes even partially, or whether the current stall extends into a second full week. Second, whether Iranian strikes continue to target US-allied Gulf states, and whether those states begin to publicly reconsider their hosting arrangements for American forces. Third, whether any credible mediator emerges or whether the diplomatic space stays empty.
Also worth watching: the Darkhovin precedent. If strikes on the nuclear plant site are followed by strikes on more advanced Iranian nuclear infrastructure, the war changes character. If not, the strike may have been a signal rather than an opening move.
Signals get missed. Openings usually do not.
The wider cost
Beyond the immediate combatants and their allies, the war is being paid for by populations that have no representation in either capital. Food prices, fuel prices, and shipping insurance are all moving in directions that hit low-income importing countries hardest. Humanitarian agencies working in the Horn of Africa and South Asia have already flagged the knock-on effects.
This is a familiar pattern in modern conflicts. The distance between where a war is fought and where its consequences are felt has been growing for decades, and it has almost never narrowed to the benefit of the people caught underneath it. That distance is not an accident of this conflict. It is a feature of how these wars are now fought, and it is the reason the loudest voices in Washington and Tehran are so rarely the ones who will pay for what comes next.
