The War With Iran Keeps Ending. The Bombing Never Does.

On paper, the war with Iran has already ended twice. A ceasefire took hold on April 8 and was later extended indefinitely; the White House still maintains that hostilities "terminated with the ceasefire on April 7th."¹⁶ Then came the 14-point memorandum of understanding signed on June 17, which formalized the ceasefire, lifted the US naval blockade, and reopened the Strait of Hormuz⁶, the narrow channel at the mouth of the Persian Gulf that carries roughly a fifth of the world's oil. President Trump declared that agreement "over"⁶ on July 8 after Iranian attacks on tankers. On Sunday, US Central Command completed its ninth consecutive night of strikes¹ on Iranian command centers, air defenses, coastal radar, and missile and drone launch sites, promising to "continue degrading Iranian military capabilities" used against shipping in the strait.

The latest round was triggered by American deaths. Two service members were killed in Jordan on July 17² when an Iranian ballistic missile got through air defenses, with one more initially listed as missing; a third died in northern Iraq³ during the controlled detonation of ordnance from a downed Iranian drone. That brings US fatalities in this five-month war to 17, with more than 430 wounded. Brent crude pushed back above $90 a barrel⁴ on the news, and gasoline is crossing the politically radioactive $4-a-gallon line⁵.

So the question worth asking on night nine is not whether the strikes are hitting their targets. They plainly are. The question is what would have to be true for the strikes to stop, and after five months of reporting on this war I cannot find an answer that survives contact with the record. This campaign is bounded in its means and unbounded in its ends. Every night's target list is precise; the definition of victory is not, and the second problem swallows the first.

The administration's defenders have a real case, and I want to state it fairly because it is the strongest argument against everything that follows. From the beginning, senior officials have described four consistent military objectives⁷: prevent an Iranian nuclear weapon, destroy Iran's navy, destroy its ballistic missile arsenal, and destroy its capacity to build more. The nightly strike logs match that list. There is no ground invasion, no occupation authority, no provisional government in a Green Zone. Measured against 2003 Iraq, whose objectives explicitly included overthrowing a regime and building its successor, this looks like a different animal: a capabilities campaign, punitive and airborne, structurally resistant to the state-building quagmire.

But a target list is not an endstate, and the record of what the president himself has said victory means is a moving staircase. When the war began on February 28 with strikes that killed Supreme Leader Ali Khamenei, Trump called it the Iranian people's best chance to take back their country⁷, an explicit invitation to regime change that his own defense secretary contradicted within days, insisting this was "not a so-called regime change war."⁹ By March 23, with Khamenei's son Mojtaba installed as successor and the Islamic Republic very much intact, Trump declared "There's automatically a regime change"⁷ because the old leaders were dead. By June 16 he had arrived at "I never cared about regime change."⁸ Along the way, unconditional surrender was demanded and then quietly shelved, and the president decided Iran could keep some ballistic missiles after all⁸, since Saudi Arabia and Qatar have them. Each of these is defensible in isolation. Together they mean the war's purpose has been retrofitted to whatever the previous month's bombing achieved.

The deeper problem is mechanical, and you can see it in the oil data. The Energy Information Administration's July outlook¹⁰ records the whiplash: Hormuz effectively closed from the war's first day, Middle East production shut-ins peaking at 11.2 million barrels a day in May, Brent averaging $103 in the second quarter, then falling below $70 on July 1 as the MOU reopened the strait and tanker traffic surged. Three weeks later the strait is closing again and Brent is back above $90. What that sawtooth pattern tells you is that five months of degradation strikes have not removed Iran's ability to shut the world's most important oil chokepoint whenever it chooses. Mines, small boats, anti-ship missiles, and drones are cheap, dispersed, and regenerable. Iran is producing roughly 10,000 Shahed drones a month at about $30,000 each¹¹, while the US fires $4 million interceptors to stop them; within the first seven weeks the Pentagon had burned through half its THAAD interceptors and nearly half its Patriots¹⁴. You cannot bomb your way to a finish line when the enemy's escalation currency costs three orders of magnitude less than your defense of it.

Which is why the operative test of victory has quietly shifted from Iranian capability to Iranian behavior. The capability boxes have largely been checked; the president says Iran has "lost everything almost militarily." Yet the strikes continue, because the actual exit condition is that Iran stops attacking ships and bases, and Iran, having survived decapitation, has every incentive to keep demonstrating that it can impose costs. A behavior test has no finish line an air campaign can reach. It has only rounds.

The costs of each round are compounding. The Pentagon told Congress the first six days cost $11.3 billion¹¹; the Penn Wharton Budget Model put the first month at $27 to $28 billion¹²; by late June, independent estimates ranged from $38.6 billion to $45 billion in direct costs¹³, with Harvard's Linda Bilmes projecting the full bill, including a quarter-trillion dollars in base repairs across seven countries, will exceed $1 trillion¹⁴. The average household has already paid about $240 in higher gasoline prices¹³, a figure that trebles if the war runs a year.

And Congress has formally objected to paying it. In June, the House voted 215-208 and the Senate 50-48¹⁵ to direct the president to remove US forces from hostilities against Iran absent a declaration of war or a new authorization for the use of military force, the statutory grant that presidents have relied on since 2001 to wage war without fresh votes. The administration's response was to note the resolution is nonbinding and to argue there are no hostilities to end, a position it maintains while announcing nightly strike tallies. Asked whether the war had taught him anything about the limits of executive power, Trump answered: "There are no limits."¹⁷ That is the constitutional posture of a war with no defined end: the branch that could impose a definition has been told its opinion is decorative.

The administration would say, with some justice, that recurrence is not the same as failure, and that a standing policy of punishing attacks while diplomacy continues is itself a strategy. Secretary of State Marco Rubio insists the US remains open to a diplomatic solution¹. But a punitive equilibrium that must be re-enforced every few weeks, at a billion dollars a day, against an adversary that regenerates its leverage faster than we deplete it, is not a war heading toward victory. It is a subscription, and the American public has already declined to renew: a June Reuters/Ipsos poll found just 24 percent believe the war has been worth the cost¹⁸.

That number is the real exit condition, and it belongs to the voters, not to CENTCOM. Wars without defined endstates do not end when objectives are met; they end when the domestic coalition sustaining them breaks, and this one is breaking with midterm elections fifteen weeks away. The administration has five months of evidence that Iran can keep closing Hormuz and one summer of $4 gasoline to explain away. Any Republican defending a Senate seat in November now owns night nine, and every night after it.