The Signing
On June 17, 2026, from a room at Versailles, Donald Trump signed a memorandum of understanding to end a war the United States had started in February by killing Iran’s supreme leader. Pezeshkian signed for Iran from his own capital. They never shared a room. The framework that made the signing possible came together in Islamabad, where Pakistan’s prime minister announced it, while Oman and Qatar worked the channels that kept Tehran and Washington in contact through the spring. The document carries the Pakistani capital’s name, the Islamabad Memorandum, a fourteen-point framework, and its terms run toward Iran: at least $300 billion in reconstruction, an end to American sanctions, immediate waivers for Iranian oil, the release of frozen Iranian funds, and the withdrawal of American forces from Iran’s approaches. Its opening clause comes before any of that. It commits both sides to end military operations on all fronts, Lebanon named in the text. Israel, whose forces are the ones striking in Lebanon, did not sign, and says the memorandum does not bind it.
The war began very differently. The United States picked the day it started. It killed Ali Khamenei in the opening hours, forced open and then closed Iran’s skies, and ran a naval blockade the administration said cost Iran half a billion dollars a day. No regional force matched it. Yet when the war had to end, Washington could not set the terms it wanted, and the talks that produced it happened in another country’s capital. The guarantee that makes a settlement hold was not Washington’s to give. The country that began the war when it chose needed other countries to finish it.
The strikes opened at 1:15 a.m. Eastern, and Trump announced them in an eight-minute Truth Social video shortly after 2 a.m., his face half hidden under a white USA cap, recorded at Mar-a-Lago, where he ran the opening hours from a makeshift situation room behind the ballroom after telling a black-tie gala to have a good time. He notified the Gang of Eight and gave no public address. The war began during active negotiations and without authorization from the UN Security Council, and the Council met only to react, in an emergency session the day the strikes landed. Eleven days later it condemned Iran’s retaliation against the Gulf states and left out the strikes that opened the war, with China and Russia abstaining over the omission. A war this size once ran through the Security Council. This one ran through a club ballroom and a 2 a.m. video.
That distance matters more than whether the United States won or lost. The capacity to start a war and to end one on your terms now belongs to different countries, and the United States holds only the first.
What It Takes to End a War
Winning the exchanges of fire does not hand you the close. Ending a war requires a separate set of capabilities: a neutral setting that the other side will enter, a broker both sides trust, a guarantor whose commitment the loser finds credible, and terms that the stronger party can impose. The United States went into this war with the firepower and little else, and the memorandum records the gap.
The firepower was never in doubt. The United States and Israel had rehearsed it in a twelve-day war in 2025, and the strikes across both campaigns set Iran’s enrichment and metal work back by years. Operation Epic Fury opened on a date fixed weeks in advance, while the Geneva talks were still reporting progress, which tells you Washington chose the war and did not stumble into it. On the first morning, it killed Ali Larijani, a central figure in Iran’s negotiations. In the same hours, a double-tap strike with Tomahawk missiles destroyed a girls’ elementary school in Minab, killing at least 165 people, most of them girls between seven and twelve. Independent investigations, and the Navy’s confirmation that it fired Tomahawks into southern Iran that day, point to an American strike that neither Washington nor Israel has claimed. The United Nations, Amnesty International, and more than a hundred members of Congress have demanded an accounting. When Iran answered, closing the Strait of Hormuz and firing missiles and drones at the Gulf states that hosted American forces, the United States forced the strait back open with an air campaign and sealed Iran’s ports with a blockade. The strikes did what overwhelming force does, and on the first morning that included a collapsed schoolhouse in Hormozgan.
The rest of the list lay beyond Washington’s reach, and the memorandum catalogs it. Washington could not dictate the terms, so the text hands Iran the money, the sanctions relief, the funds, and the withdrawal, and asks in return a renewed pledge not to build a bomb and a strait kept open for sixty days before the question reopens. The talks that broke the deadlock were held in Islamabad, not on American soil. The security guarantee had turned reactive. Washington refused Saudi Arabia a binding commitment, and Riyadh pulled its defense closer to Pakistan; then weeks later it handed Qatar an explicit pledge of defense after Israel bombed Doha. The promise that once anchored a regional order now arrives case by case, after the strike, to whoever was hit last.
A skeptic will say Washington did not lose the power to dictate terms so much as decline to pay for them, and the objection has force. About a fifth of the world’s oil and gas passed through the Strait of Hormuz, so closing it and blockading Iran’s ports drove fuel prices up and strained the economy Trump needed steady before the November midterms. The United States could have held out for more, and chose not to. But the reason holding out grew so costly is the reason the terms slipped away. With the broker, the guarantor, and the exit all in other capitals, each added week of war bought Washington pressure and no bargaining leverage. A preponderant power that must pay other states to end its wars bargains against a clock those states control. Calling it a choice does not make it less of a loss.
The battlefield did not produce this gap. The United States won there. The shortage behind it predates this war. Thomas Schelling described it sixty years ago in Arms and Influence, separating brute force, which seizes what it can reach, from coercion, which makes an adversary accept your terms. The United States brought the brute force. Coercion runs on the adversary’s consent, and that consent stayed outside what the strikes could reach. Fred Iklé, in Every War Must End, gave the problem of war termination its modern form: that governments are built to start wars and poorly equipped to finish them. This war added what neither anticipated. The trusted broker and the credible guarantee had left American hands for other capitals before the first plane took off.
That older work of ending wars has a name. Centuries before Grotius, the Hanafi jurist Muhammad al-Shaybani set down a systematic Islamic law of nations, the siyar, organized around treaty-making, reciprocity, and the conduct of truces with adversaries. The renewable sixty-day truce at the heart of the memorandum is a form that tradition worked out more than a thousand years ago. Reading the settlement only through Geneva and Grotius overlooks the legal world in which the parties themselves were operating.
The Clock and the Threat
Who chose the moment is contested. Hawks treat the strike as a move in the contest with China, a way to take an Iranian asset off Beijing’s board. The administration’s own account points elsewhere. Secretary of State Marco Rubio told congressional leaders on March 2 that the United States struck because it knew Israel was going to act against Iran regardless, and that American forces would draw Iranian fire once it did. The justification shifted the next day to an imminent Iranian attack, one of several rationales that rotated through the early days. Underneath the shifting was an alliance dynamic the theorist Glenn Snyder named decades ago: entrapment, the pull that drags a strong state into its ally’s war out of fear the ally will fight alone. The United States did not set this war’s clock. Israel set it, and the alliance’s pull brought Washington along.
The clock sat inside a threat picture both governments had spent years assembling. Washington and Israel had come to treat Iran’s allied forces, Hezbollah in Lebanon, the Houthis in Yemen, the militias in Iraq, Hamas in Gaza, as one enemy, the Axis of Resistance, and they made taking it apart an explicit aim of the war. The American precondition for talks was that Iran end all support for the network, and the plan was to dismantle the whole structure and move the fight to Iran itself. Israel had signaled the approach for years, in Netanyahu’s doctrine of the preemptive strike and in a campaign of assassinations that killed the network’s leaders one after another. The framing did the work that threat framings do. It collapsed a set of distinct actors, each with its own base and its own quarrels, into a single existential enemy, and promised that breaking the enemy would buy security.
The war combined two different threats. Iran’s nuclear program was a real concern, and the strikes set it back. The Axis of Resistance was not the same kind of threat, and the war tested the promise that breaking it would buy security and found it thin. Years of killing the network’s leaders had not dissolved it, and through this war parts of it fought while others held back, since defending Iran was never the first priority for all of them. The name itself runs two ways. To Tehran it is a banner of deterrence, the forward defense it spent two decades building. Washington came to use the same words for a monolith it meant to destroy. Security studies call the move securitization, the building of an adversary into an existential threat to license what would otherwise be unthinkable, and the Iraq war is the nearest reminder of where it leads. The Axis framing drove the United States into a war it could open at will, fought against a network its strikes never erased. The threat picture promised force would settle the matter, and the firepower, the strike on Minab among it, secured no terms.
This war was the newest front in a longer campaign. Since October 2023, Israel has used the freedom of action its firepower buys, with American weapons and American cover, across Gaza, Lebanon, Syria, and the Gulf. On September 9, 2025, Israeli jets fired ten missiles into a residential compound in Doha, trying to kill Hamas negotiators on the soil of Qatar, a country that hosts a major American base and was at that moment mediating a Gaza ceasefire. The strike failed and killed a Qatari security officer among others. The same freedom of action runs through Gaza, where the record is now formal. In September 2025, the United Nations Independent International Commission of Inquiry concluded that Israel is committing genocide against Palestinians in Gaza, finding four of the five acts defined in the 1948 Genocide Convention. The International Association of Genocide Scholars reached the same conclusion in a resolution that month. The case South Africa brought before the International Court of Justice in December 2023 remains open, and the International Criminal Court issued arrest warrants for Netanyahu and Gallant in November 2024. Israel rejects the findings, calling them ridiculous and a blatant falsehood and asserting its right to self-defense. The Iran war belongs to that campaign, where the same force met the limit of what it can settle, and where the United States, having armed it, could not end the war it produced.
Washington and Tel Aviv were not the only hands on the outcome. Mohammed Ayoob’s subaltern realism holds that postcolonial states act first on the insecurity closest to home, the survival of the regime, and only then on the roles the great powers assign them. Iran’s choices, the Gulf’s hedging, and Pakistan’s brokerage followed that logic, and the settlement emerged from those moves.
The Tempo Without a Table
The settlement ran on a faster and choppier clock than war termination used to, and the reason sits in the missing table. Watch the talking instead of the fighting and the pattern comes clear. In three and a half months the two sides moved through a Pakistan-brokered two-week ceasefire in early April, a 21-hour direct round in Islamabad that collapsed without an agreement, an indefinite extension declared hours before the first ceasefire lapsed, a cancelled negotiators’ trip, a near-miss in May, and the 60-day ceasefire that carried the June memorandum, with separate Lebanon ceasefires running on their own clock alongside. CNN counted at least 38 times between March 23 and June 9 that Trump called a deal imminent, close to one every two days. The victory claims kept the same tempo. Trump declared Iran militarily defeated on April 6 after saying so on March 17, March 24, and March 26, each time contradicted within hours by the next strike or shipping disruption. The agreement was announced before it was settled, signed from separate capitals, and its central term, the open strait, broke within days of the signing.
Set that against the benchmark for ending a war at the table. The Korean armistice took 158 meetings over two years and 17 days, the longest negotiated armistice on record, close to 200 sessions and 400 hours of argument in a fixed building at Panmunjom, between professional military delegations, with the agenda alone consuming the first seventeen days. A single negotiation in a single room, settled over two years. This war ran the other way, dozens of public declarations and reversals packed into weeks, with no agenda, no fixed room, and no delegation seated across from anyone.
The table did more than house the talks. It set their pace. A fixed agenda and a standing delegation are friction, and friction slows a process and orders its steps. Take them out, and the tempo passes to whoever holds the deadline, reset here every few days on social media and carried by middle-power shuttle. What stands in for the forum runs faster and breaks more often, and the breaks carry a price, paid in the oil spikes and fuel shortages that followed each reversal in the strait.
The choppiness is real, and the pull is to read it as the end of an order. The smaller reading carries more weight. The deadlines that set the tempo were the President’s, improvised in real time and confirmed by his own officials as not part of any national security plan, timed to an economy and an election he needed steady, and broadcast to a home audience, with stocks and oil moving on the announcement. The forum that once stood between that kind of impulse and the use of force has thinned. Reading the result only at the level of the international system, as a leaderless world or a dying order, skips the step where the volatility is generated, which is closer to the ground, in domestic political time. What the war shows is narrower than a collapsing order and more specific. The process that used to absorb and slow these shocks has gone informal, and they pass through now with less to stop them.
The Verdicts That Overreach
The confident verdicts on this war each read a single number, up or down, and the record refuses them.
The China hawks, Zineb Riboua and others at Hudson, see a win over Beijing, an Iranian partner pulled from China’s column. That needs the war to have been about China, and the administration’s own timeline puts Israel at the controls. It needs the consequences for China to point one way, and they cut both ways: by one count China lost a partner and watched its mediation sidelined, by another it gained from an America that spent itself for little.
Ian Bremmer and Firas Maksad argue the reverse, an American defeat in a leaderless world tilting toward China. The defeat half mistakes the axis, since the United States won the fighting it sought. As for China rising, the states that brokered the truce were middle powers acting in their own interests. And the pact Bremmer leans on hardest as proof of a war-driven realignment, the Saudi-Pakistan defense agreement, predates the war by five months. What he calls the war’s consequence stood in place before it. The other half, the speed he reads as a world coming apart, has a nearer source in a formal process gone informal.
The wash verdict, from Barack Obama and, more carefully, Hassan Mneimneh at Carnegie, holds that the war cost much and changed little, exposing the architecture of American power for marginal gain. This reading comes closest to the truth, and it differs only in method. Mneimneh renders his verdict and closes the book. The count stays open here, because what the war exposed is still in motion.
The restraint camp, Trita Parsi and John Mearsheimer, calls it proof of the limits of American power and puts the limit in the wrong place. The United States could destroy whatever it chose. It could not settle. The counterproliferation hawks at FDD call it a necessary success, unproven so far, since Iran’s enriched stockpile survived and the IAEA has been blind inside the country since the first strike.
The single number fails for a structural reason. The capacities it averages move in opposite directions, and the belligerents cannot agree on the sign. Each reads the same memorandum as its own victory, Washington as strength, Tehran as the prudent flexibility its doctrine sanctions, Israel as a job done. No verdict that picks one reading survives the other three.
In Their Own Words
Across the table, the loudest claims of strength rested atop the largest concessions. War Secretary Pete Hegseth said the United States reached the agreement from a position of strength and warned it could clamp an ironclad blockade back on Iran if Tehran broke faith. Vice President J.D. Vance defended the deal by insisting the United States was not giving Iran a cent, even as the text released Iran’s frozen funds and promised $300 billion in reconstruction. Trump measured the outcome by the economy. He cast the deal as what spared the country an economic catastrophe and held up a rising stock market and cheaper fuel as the proof. The victory he claimed rested on American prosperity, and it left the terms he had signed out of the telling.
Tehran’s words ran the other way, defiant about a deal it had decided to take. The new supreme leader, Mojtaba Khamenei, said he had held a different view but approved the agreement, and that direct talks with Washington would not amount to surrender. Taken at face value, the statement looks like a leader defying a deal he just authorized. Inside Iran’s own tradition, it is coherent and even familiar. Ali Khamenei gave the move its name in 2013, heroic flexibility, when he licensed the nuclear talks that produced the 2015 accord, drawing on the Shia memory of Imam Hasan, who made peace with a stronger rival while insisting he had surrendered nothing. Behind it sits the doctrine of maslahat, the expediency that lets the survival of the Islamic system justify a bitter compromise. The precedent runs back to Ayatollah Khomeini accepting the 1988 ceasefire with Iraq, which he compared to drinking from a cup of poison. Mojtaba Khamenei approving a deal he disliked while denying any surrender is not improvisation. It is Iran ending a war the way its doctrine has taught, by taking the prudent truce and calling it by its right name. Tehran signed electronically and kept its leaders out of any photograph that might appear to be an acceptance.
At home, Tehran lets the concessions make the case. It presents the lifted sanctions and the unfrozen funds as what steadfastness extracted from a stronger enemy, and its lead negotiator called the memorandum a record of American failure.
Netanyahu told Israelis the war had crushed Iran’s nuclear facilities and achieved its aims, while admitting he had not seen the memorandum’s full terms and had been kept outside the talks that produced it. The leader whose timing started the war had no hand in how it ended and learned the terms secondhand. In response to those claims, Pakistan’s prime minister Shehbaz Sharif said only that the peace deal had been reached. His claim matched what he had done.
The Realignment Came First
The realignment offered as a consequence of the war was on the table before the war began. The clearest marker is the Saudi-Pakistan defense pact, signed in Riyadh on September 17, 2025, which treats an attack on either country as an attack on both and tucks Saudi Arabia under Pakistan’s nuclear umbrella. It was signed eight days after Israel’s strike on Doha and five months before the United States struck Iran, and analysts took it as a message from Riyadh to Washington that the American security guarantee no longer looked credible. The very pact offered as proof that the war reordered the region was signed before it.
The deeper marker runs back further, to March 2023, when China brokered the restoration of relations between Saudi Arabia and Iran and began a regional habit of routing disputes through capitals other than Washington. By the time this war needed an off-ramp, the brokers were in place and practiced. Pakistan, Qatar, and Oman did not improvise the settlement. They stepped into roles the region had been building for years.
The cohesion questions hanging over this settlement have an older name. The Arab historian Ibn Khaldun called it asabiyya, the group solidarity that lets a power rise and whose decay lets it fall. Can Iran’s revolutionary state hold together under a harder successor installed by a decapitation strike? The same doubt hangs over the battered Axis of Resistance, and over whether the Saudi-Pakistan alignment is bound by anything sturdier than a shared bad month.
The skeptics are strongest on that last question. The Middle East Institute notes that Pakistan’s army is built to deter India and is unlikely to fight Iran or Israel on Riyadh’s behalf, so the pact may signal more than it binds. The claim here stays narrower than a finished bloc replacing the American order. What the region had assembled by then was the habit of brokering its own disputes, and that habit was already standing.
What to Watch
Declining to crown a winner is only worth something if a measurement takes the place of the verdict. If the capacity to end a war on your terms is what is moving, it can be tracked by watching who ends them, war by war, in signals anyone can pull from the public record. The measurement takes the shape of an index, six indicators, each a plain question with an answer you can check:
Convening. Who provides the neutral ground and the informal channel that gets the parties to a table at all? In this war, the talks ran through Muscat, Geneva, and Islamabad, settings opened by states both sides would meet on, and Pakistan and Qatar proposed the Swiss resort for the close. Washington hosted no round.
Brokerage. Who carries the messages and is trusted by both sides to mediate? The named mediators were Pakistan, Qatar, and Oman.
Guarantee. Whose security promise do the parties seek to make the terms hold? Saudi Arabia asked Washington, Washington refused, and Riyadh turned to Islamabad. The American guarantee behind this deal is thin.
Terms. Do the terms favor the militarily dominant party or the weaker one? The dominant party conceded, across reconstruction, sanctions, funds, and withdrawal.
Hedging. Are the dominant power’s allies buying their security elsewhere? The Saudi-Pakistan pact says yes.
Cost. Does the dominant power make the side payments that close the deal, or collect them? Washington pays, underwriting reconstruction financing past $300 billion and the relief that comes with it.
Code those six for this war and the same answer comes back. Settlement occurred outside Washington, and on the one count in which Washington acted, it acted as the party that pays. Another analyst, working only from the public record, codes it the same way. Run the same six against the 1991 Gulf War, and the answers flip. That war ran through the Council, which authorized “all necessary means” in Resolution 678 before the first shot was fired, and Washington convened the coalition, dictated the ceasefire at Safwan, guaranteed the Gulf that followed, and was paid for the war instead of paying for it. This one ran around the Council. It opened without an authorizing resolution, and the body’s one binding action condemned Iran’s retaliation while omitting the strikes that began the war. The forum that gated escalation in 1991 was bypassed at the start and deadlocked at the end, and the convening it once did moved to Muscat, Geneva, and Islamabad. The same six questions distinguish the two wars.
The sixty-day window the memorandum opens is the index’s first live run, and it began to move within days. Iran refused to appear at the signing ceremony set for June 19 in Switzerland, citing Israel’s strikes in Lebanon, and a senior Pakistani mediator flew to Tehran to bring the delegation back. The talks reconvened at the Bürgenstock resort on June 21, not as a meeting between Washington and Tehran but as a four-party session with Pakistan and Qatar at the table; even then, Iran refused to enter the room while the cameras stayed, denying Washington the photograph. Convening and brokerage held in the live run as they had in the war.
The central term gave way next. Three days after agreeing to reopen the Strait of Hormuz, Iran declared it closed again over the same Lebanon strikes, and Washington denied that Iran controls the strait and said the oil kept moving. The dispute is the point. The deal’s signature deliverable now turns on whether Washington can make its own ally stop firing, and through the first week it could not. The clause it turns on is the memorandum’s first, the end of fighting on all fronts that names Lebanon, and the army breaking it is the one party that never signed. Israel’s defense minister said the army would not withdraw from Lebanon, where its strikes had pushed the death toll past four thousand, and the ally that set this war’s beginning was now shaping its end. The other lines wait behind it: the frozen funds Iran ties to the clock, the enriched uranium whose fate is unsettled, Netanyahu’s election due by late October, and a supreme leader four months into the job holding a truce he says he never wanted. The index does not predict which line breaks first. It reads the breaks as they land, and the strait broke inside three days.
As of June 23, the live run still reads the same way. The first round of technical talks closed at Bürgenstock with a road map toward a final deal inside sixty days, and the Lebanon clause became its own proving ground, with Iran, the United States, and Lebanon standing up a deconfliction cell to hold the all-fronts ceasefire, which Iran’s foreign minister called the first real test. Trump called the strait totally open while Iran’s lead negotiator flew to Muscat to work with Oman and held that Tehran runs it. The next round moves to Washington on June 23, the first the United States will host, after every round that built the deal ran somewhere else. By the same day, Israeli and Lebanese delegations were meeting in Washington over a partial Israeli withdrawal from southern Lebanon, the clause the deal turns on now under direct negotiation.
The Function That Moved
The United States could start this war on a morning of its choosing and could not end it without help from Pakistan, Qatar, and a memorandum that paid Iran to stop. That fact is larger than any tally of who came out ahead. The capacity to begin a war and the capacity to end one have come apart, and Washington holds the first.
The win, the defeat, and the wash all miss what moved. What the war left behind is a redistribution still in motion, the diffusion of the capacity to end wars toward the middle and regional powers that now hold it.
That is not an American defeat in the way a lost war is a defeat. It is the kind of change Hedley Bull had in mind when he described great powers as the managers of international order, holding functions that can be lost one at a time. The United States has lost one of them here, in plain sight, and has not said so out loud. It could open this war in a single night and could not, at the end, bring the other side into one room to close it. The power to end wars has moved, and the next war will show how far.
Christopher Sweat is Co-Founder and CEO of GrayStak. GrayStak is a media and political risk company. Our real-time Domestic Political Volatility Index (DPVI) measures the velocity and volatility of U.S. domestic politics. We cover political and institutional risk across the United States, centered in Washington, D.C., New York, and Chicago.
GrayStak holds no position.