The Empire That Isn't

Five years ago yesterday, the last American aircraft lifted out of Kabul. The New York Times marked the anniversary with an analysis arguing that the fears behind the war were unfounded, that the stakes in the Gulf are low, and that the lesson of Afghanistan is learning how to quit. Today’s Long View is my answer. Because the argument it answers will be everywhere this fall, this edition goes to every reader, free and premium alike. The daily continues this week, and the Boston stop of the Fall Speaker Series is tomorrow.

The Empire That Isn’t

Five years after Kabul, the Times says the lesson is learning how to quit. The evidence reads the other way: America does not police an empire, it preempts chaos, and every time it stops, the bill arrives.

Five years ago yesterday, the last C-17 climbed out of the Kabul airfield and the United States was out of Afghanistan. The New York Times marked the anniversary with an analysis built on two claims. The fears that kept three presidents in the war “have proven to be unfounded,” because Afghanistan “has not become a terror haven or launch pad for attacks in the United States.” And in the Gulf today, “the stakes are much lower, and so too are the costs of quitting.”

The piece hands its frame to Andrew Hoehn of RAND. “So much of what we’ve been involved in these last several decades is really policing the empire,” he says. And then: “We’re not fighting existential conflicts.”

The first quote concedes what the restraint school usually denies: that there is an order out there worth keeping, and that somebody keeps it. Hoehn has described maintenance and called it futility.

The second quote is not an indictment either. It is the outcome the forward posture purchased. American conflicts stayed non-existential because America fought them early, far away, and at manageable scale. No one in 1938 could boast that the free world’s conflicts were non-existential; the existential one was already assembling, precisely because nobody had gone forward to meet it.

One note on the bench: the piece’s expert chorus runs from RAND to the Middle East program director of Defense Priorities, a restraint advocacy shop, so the sourcing carries the conclusion.

The deeper problem is older than this article. Deterrence shows its costs and hides its products. The deployments, the budgets, and the funerals are visible. The tanker that transits, the war that never starts, and the bomb never built are not. Every generation rediscovers the invisible half and calls the visible half waste. The name for what America does at its best is not policing. It is preemption, and the difference is the subject of this essay.

The noun fails first. Empires tax subjects and take territory. America garrisons allies who pay to host its soldiers and grow rich beneath the umbrella. South Korea has hosted roughly 28,500 American troops for seven decades while a war-ruined peninsula became a top-ten economy, and that economy now sells tanks and howitzers to America’s NATO allies. If that is policing an empire, the beat produced Samsung.

The verb fails next. A policeman is reactive; he arrives after the crime. He is neutral among the residents on his block, and he is bound to procedure. American power at its best is none of those things. It is anticipatory, openly partial to its friends, and concerned with outcomes rather than process.

The cleanest proof is a paired test from a single six-month stretch. On September 14, 2019, Iran gutted the Abqaiq and Khurais oil facilities in Saudi Arabia, the heart of the kingdom’s processing capacity, and Washington chose no military response. The region read the silence and began to hedge. On January 3, 2020, Qassem Soleimani died at Baghdad airport, and the world war the commentariat promised lasted one calibrated missile barrage against a forewarned base. Sparing the system invited more. Punishing the man deterred.

The article’s empirical anchor is that single sentence: no haven, no launch pad. It fails in three cuts.

First, the haven. Eleven months after the last flight out, al-Qaeda’s emir, Ayman al-Zawahiri, was living in a downtown Kabul safe house tied to the leadership of the Haqqani network, the faction that runs the Taliban’s interior ministry. United Nations monitors have documented the Taliban’s hospitality to al-Qaeda ever since.

Second, the launch pad, a claim that survives only on its qualifier. ISIS-K, run from the Afghanistan-Pakistan theater, killed about a hundred mourners at Kerman in Iran and 145 concertgoers at Crocus City Hall outside Moscow. It built the Vienna concert plot that aimed, in the CIA’s words, to kill “tens of thousands.” And it reached inside the United States: an Afghan evacuee arrested in Oklahoma City in 2024 for a plot timed to election day, an ISIS-K suspect arrested miles from the nation’s capital last December, and the Abbey Gate planner, who reached an American courtroom in 2025. “No attack in the United States” was also true on September 10, 2001.

The strongest counter-brief, from the International Crisis Group, argues that ISIS-K is in retreat inside Afghanistan. Split in two, the claim answers itself. Suppression inside the country is Taliban self-interest. External planning migrated across the theater and kept producing bodies. A haven that exports is not a haven eliminated.

Third, the costs were real, and the concession belongs in the Times’ own figures: more than 2,400 American dead across the war, 868 of them after bin Laden was killed. But those figures indict a particular way of fighting. Goals shrunk to casualty minimization. A commander who expected to be taken for a fool. Analysts likening the mission to a critical-care ward whose doctors hoped the patient would die on someone else’s shift. That is a portrait of half-war, of fighting not to lose. Carter Malkasian tells the Times that the American military “fights to win.” Washington kept assigning it missions in which winning was undefined.

And the choice was never between a twenty-year occupation and an empty airfield. By January 2021, the American presence was down to 2,500 troops, and it went roughly eighteen months without a combat death. That residual position was traded away at Doha, in a deal negotiated with the Taliban over the elected government’s head. The disease was aimless occupation. The cure preemption prescribes is defined objectives and decisive means, with no nation-building.

The pattern generalizes. Iraq, 2011: withdraw, watch the Islamic State rise, return in 2014 at far higher cost. Libya, 2011: decapitate and leave, and Qaddafi’s scattered arsenals armed the Sahel for a decade. Syria, 2013: leave the red line unenforced, and Russia entered the Levant two years later while the refugee wave remade European politics. Quitting is also a policy, and it has a death toll of its own.

Kabul fell in August 2021. Russia massed on Ukraine’s border that fall and invaded in February 2022. Correlation is not causation. Price signals are price signals.

The word carries 2002 baggage, so it needs clearing first. Preemption here is not preventive invasion. No armored columns rolling toward a foreign capital, no occupation authority, no ministry-building. It is a standing architecture that runs on two rails.

The first rail is commercial: alignment with America must pay, quickly and visibly. The Abraham Accords built trade that grew through even the Gaza war. The corridor announced in 2023 from India through the Gulf to Europe, the technology and investment compacts signed across the Gulf in May 2025, sanctions relief sequenced for a Damascus that behaves, and American financing for the Lobito rail corridor carrying Congolese and Zambian minerals to the Atlantic are the same sentence written in four regions: side with Washington and doors open.

The second rail is personal: defiance must ruin the men who choose it, not the peoples they rule. Soleimani at Baghdad airport. The exploding pagers, and then Hassan Nasrallah under the rubble of his Beirut headquarters. United Nations sanctions snapped back in September 2025 after the Europeans triggered the mechanism Tehran assumed was dead. And the first night of this war reached the very top of the Islamic Republic: Ali Khamenei did not survive it, and every palace that matters has learned the price.

The signature move is the migration of punishment from populations to palaces. Blanket embargoes and occupations immiserate societies. Preemption prices bad behavior to the individuals who order it. That is why this, and not retreat, is the anti-imperial position. Empires punish peoples. Preemption punishes princes.

The rails only work together. Carrots without sticks read as tribute; that was the 2015 nuclear deal, which paid rent on a problem it did not solve. Sticks without carrots read as siege, and siege is what actually produces the rally-round-the-flag effect the Times worries about. Hold both rails and the audience is no longer a nation but a cabinet, every member of which can compare the price of alignment with the price Soleimani paid.

“Empires punish peoples. Preemption punishes princes. That is why the forward posture, not the retreat, is the anti-imperial position.”

Applied to the file the Times says should scare America home, the doctrine reads the history differently. The Islamic Republic has held power for nearly half a century, and its longevity is not proof that pressure fails. It is the compounding bill for four decades of managing what should have been defeated: the 444 days of the embassy seizure; the Beirut barracks bombing of 1983, unanswered; Khobar Towers in 1996, unanswered; the 2015 deal, paying rent; Abqaiq, met with silence.

The exceptions prove the rule. Operation Praying Mantis in 1988, launched after an Iranian mine nearly sank an American frigate, put half of Iran’s surface navy out of action in a day. Baghdad airport in 2020. The twelve-day war of June 2025, when American bombers struck Fordow, Natanz, and Isfahan in a single night. The first night of this war. Force applied briefly and decisively, and none of it produced the promised quagmire. This regime is a problem to be defeated rather than managed, and every decade spent managing it has raised the price of the defeat.

Six months into this war, the Times counsels the Afghanistan reflex: victory looks bleak, the public is souring, wind it down. Its own witnesses argue against it. General Dan Caine, the chairman of the Joint Chiefs, concedes only that “air power has got limits.” Thomas Mahnken calls reopening the strait “a doable task.” Secretary Hegseth, who told the country “This is not Iraq. This is not endless,” says the Navy can hold the blockade “as long as we need to.” A sea blockade is not a counterinsurgency. There are no villages to hold and no ministries to build, attrition runs against the blockaded party, and nobody has proposed occupying Iran. The Afghanistan analogy requires an occupation that no one is proposing.

This week’s evidence says the squeeze is working the way sieges work, slowly and in one direction. The ceasefire that the June 17 Islamabad Memorandum tried to extend lapsed on August 17. Last week CENTCOM finished clearing Iran’s mines from the approaches; over the weekend it struck the IRGC launch teams on Larak Island that were preparing to fire mine-laying rockets back into the strait, the first American strikes in a month, and Tehran’s answer, ballistic missiles aimed at a base in Jordan, was intercepted in Jordanian airspace early this morning. Merchant transits with no Iranian connection ran above their spring average in mid-August even as the blockade strangles Iranian trade, with 64 merchant vessels redirected, three disabled, and two boarded by CENTCOM’s own figures. The transit tolls Tehran collected in the spring, reported at up to $2 million a vessel, have collapsed into a ransom note: the strait reopens, says Iran’s security council, when America ends the war and pays compensation.

The preemptive endgame is concrete, and readers of this newsletter know its name, the Hormuz Mandate. No tolls in Hormuz, ever, because a transit fee normalized is extortion institutionalized. Verifiable dismantlement of the nuclear program. The blockade held until both. And beside those sticks, a standing, published offer of commercial reintegration for an Iran that stops exporting revolution, so that both rails stay visible to every colonel and cleric weighing succession politics in Tehran. The permanent guarantee remains what this newsletter has always said it is: a different government, won by the Iranian people, with America backing them and pricing their jailers.

The President’s instinct on the oil is right: America does not need Iran’s barrels, and he has refused to let Tehran’s blackmail set his clock. The catch is the global market. Hormuz carries roughly a fifth of the world’s oil and much of its liquefied natural gas, so a tolled or closed strait taxes every American at the pump no matter where the barrels land. The stakes are not low. They are diffuse, which is a different thing, and diffuse stakes are the kind a republic notices only after it stops paying the insurance.

Syria is the control group, the file that shows what fills a vacuum when America merely watches. Assad fell in December 2024 to decisions made in Idlib, Moscow, and Ankara. Washington abstained, and abstention did not produce Syrian self-determination. It selected which foreign capital would hold the lease.

The summer’s evidence: Turkish outposts stand across northern and eastern Syria, none relinquished since the regime fell. Turkish warships called at Latakia in July, the first such port call since the civil war began. Ankara is reorganizing the Syrian armed forces and buying up the coast at Latakia and Tartus. A NATO ally that bought Russian S-400s and was expelled from the F-35 program for it now runs a client on Israel’s border, and Israeli jets have already struck the Abu al-Duhur airfield to keep Turkish combat drones off it. Jerusalem is drawing the line that Washington has not drawn.

As this newsletter’s Turkey war file argued last week, that collision is avoidable, but only by an America that sets terms. The preemptive alternative is on the table now: sequence the sanctions relief and the reconstruction money against verifiable behavior, keep the eastern presence and the Syrian Democratic Forces as leverage, and price sole-patron capture of Damascus as unacceptable. “Not our problem” purchased a competitor’s writ. It did not purchase neutrality.

The regions critics call peripheral run the experiment cleanest, because America is most absent there. In Sudan, United Nations investigators concluded in February that the Rapid Support Forces committed genocide at El Fasher. By the monitors’ figures, at least 60,000 people may have been killed since the city fell last October, a figure still unconfirmed, with 150,000 residents unaccounted for, more than thirteen million Sudanese displaced, and famine confirmed. Around that horror runs a bidding war: the UAE, Egypt, Türkiye, Russia, and Iran all arm or fund the parties while eyeing the Red Sea coast. The United States holds more leverage over more of those bidders than any other actor, and it has declined to preside. So the auction proceeds without a floor.

In the Sahel, the French decade failed, and then the American residue was expelled: the Air Force handed over its last Niger base in August 2024 and the surveillance coverage went dark. The vacancy was let to Russia’s Africa Corps, a protection racket that kills more civilians than the jihadists it was hired against and whose fighters lost their largest battle, at Tinzaouaten on the Algerian border, in 2024. Al-Qaeda’s local franchise, JNIM, has meanwhile graduated from ambushes to strangulation, blockading western Mali and cutting the fuel lines from Senegal and Côte d’Ivoire that carry most of the country’s supply, with the coastal states next in line and Europe’s migration politics directly downstream. The Sahel now accounts for more than half of the world’s terrorism deaths. Readers know the rule here as the No-Vacuum Doctrine: every American withdrawal is an entry for somebody worse.

Then the tell. Even allies have priced in American absence. On August 7, Saudi Arabia, Türkiye, and Pakistan signed a mutual defense pact at Mecca that names no threat and, by its framers’ own account, exists because they no longer assume Washington. The pact has a Pakistani bomb in it, and Tehran is openly discussed as a future member. Vacuums do not stay empty. They get organized, sometimes by friends, on terms that dilute American leverage, which is why the Forum answered the pact with a case for an alliance worth assuming.

The affirmative offer for Africa is the first rail written as contracts rather than garrisons: corridors, minerals, and security assistance with conditions attached, set against Moscow’s blood-for-gold model, which is failing on its own terms.

Six objections deserve answers, in ascending order of difficulty.

1. Their survival, our discretion. The article’s sharpest point: the Taliban fought for survival, Iran’s rulers do too, and America’s stakes are lower, “and so too are the costs of quitting.” The asymmetry is real, and it is exactly why preemption aims at the survival calculus of rulers rather than the endurance of nations. The Taliban were an insurgency with a Pakistani sanctuary, swimming in a sympathetic population, cheap to sustain and hard to target. A state regime fighting for survival has palaces, oil terminals, a navy, a praetorian guard that must be paid, and a succession problem. Its survival can be priced, and the first night of this war set the price at the very top. Deterrence works on men with addresses.

2. The public opposes it. The figures are real: Quinnipiac in late July found six in ten voters opposed to the war, Reuters/Ipsos in late August put support at 31 percent, and an AP-NORC survey found about two in three saying the war has not been worth it. But opinion follows outcomes and prices. The war is unpopular because it is stalemated, not stalemated because it is unpopular. Gallup found a majority of Americans calling Korea a mistake in early 1952; the commitment outlived the poll and built the ally that now arms NATO’s eastern flank. The task is shortening the distance to a visible outcome, the strait open, the tolls dead, pump prices down, not converting a poll into a policy.

3. The money belongs in the China pivot. The arc from Hormuz to the Gulf of Guinea is the China file. Beijing opened its first overseas base in Djibouti in 2017 and a naval pier at Ream in Cambodia in 2025; it holds Belt and Road ports on both African coasts, bids for the Gulf’s computing power, and locks up African minerals. Retreat from the arc funds no pivot. It forfeits the board the pivot is supposed to contest.

4. Pressure rallies Iranians to the regime. A regime that has carried out at least 916 executions since January, in what Amnesty International calls an escalating campaign of arbitrary executions against its own protesters, is not a nation rallied. It is a regime at war with its people in the middle of a war. Preemption’s design keeps the quarrel with the rulers, which is why the pairing of the rails matters: siege alone would hand the regime the flag it wants to wave.

5. The forever-war slope. Preemption refuses occupation by design, and its combat record is brief and closed-ended: 1988, 2020, 2025, the first night of this war. The standing elements, the fleets, the bases, the sanctions architecture, have run for seventy years without becoming Vietnam. The slope runs the other way. It is the half-wars of undefined objectives that never end.

6. Barkhane, the hardest one. France was vigorous in the Sahel for nine years, Operation Barkhane, 2013 to 2022, and failed anyway. The honest answer is that Paris ran one rail: all garrison and kinetics, no commercial offer a Sahelian government could sell at home, and a former colonizer’s static on every frequency. Barkhane proved that presence without a prosperity rail fails, which is the doctrine’s own claim. Preemption is not presence for its own sake. It is presence priced into an offer.

Hoehn deserves one concession: Americans have asked about quitting every war since 1945. That is the burden of the power whose absence is the variable everyone else plans around. The question now has a five-year answer from Kabul and a six-month answer from the Gulf, and the two answers agree.

The Times closes its anniversary piece at a Pentagon ceremony for the Afghan war’s veterans, and it notes that “no one mentioned the country’s defeat.” Here is the mention. The defeat was chosen at Doha and completed at Abbey Gate, and the world has been pricing it ever since. The way to honor its dead is to refuse the sequel.

The conflicts stayed non-existential because somebody kept them that way. That is not the patrol log of an empire. It is the maintenance file of an order, and the maintenance is the cheapest thing the republic buys.

This special edition is free for every reader. The Sunday Long View, the full archive, the quarterly threat briefings, and reply access are for premium members: $12 a month or $100 a year. Every MEF donor reads free; if your invitation has not arrived, write [email protected].

The American service members killed in Afghanistan after Osama bin Laden’s death in May 2011, by the Times’ own figures. That is the price of a decade of half-war, of fighting not to lose. It is the case for defined objectives and decisive means. It is not the case for absence.

Yesterday’s premium Long View, The Anti-Woke Diaspora, profiled the twenty exile voices this fight runs through, and Friday’s The Next Six Months maps the war this essay defends; the daily continues this week. New readers can orient with the Field Guide; the MEF Dispatch, our summary of new articles and analysis from meforum.org, is one click: Subscribe to the MEF Dispatch.

Boston opens the series tomorrow with the money war; Philadelphia on Thursday asks what victory requires; New York on September 9 follows the tentacles home.

Boston · Tuesday, September 1 · Bankrupt the Revolution: The Money War Against Tehran

Philadelphia · Thursday, September 3 · Finish the Job: What Victory over Iran Requires

New York · Wednesday, September 9 · The Tentacles of the Outlaws: Iran’s Reach into American Life

All three run 12:00 to 1:30 p.m. Seating priority runs with annual giving. To host an evening around any stop, write Nikki Ellis ([email protected]) or me ([email protected]).

Clear this morning. When it should not be, write [email protected]; corrections run here, in print.

The argument you just read is the argument the Forum carries into Washington, into the press, and onto three stages in the next nine days. The researchers who keep that case sourced and airtight are funded by readers. If this essay armed you, arm them back. And forward this edition to one person who read the Times piece and nodded.

Reply question: which vacancy costs America most, Damascus, Khartoum, or Bamako? Replies land in my inbox; the sharpest answers run in the next mailbag.

Middle East Insider · a project of the Middle East Forum