Stay Until the Children Grow Up in Peace

Section 1The question before us

There is a way of using military force that costs more lives than either fighting properly or staying home. It is the half measure: enough force to break a state, not enough to rebuild it; enough presence to be resented, not enough to be useful. The United States has chosen it repeatedly, and the cost is not paid by the people who choose it.

This paper takes no position on whether any particular war should have been begun. That is a separate question, decided by others, and litigating it has become a way of avoiding the question that actually determines outcomes. Our subject is what the country owes a place once its soldiers are standing in it.

1.2   Once committed, the debate is over

A great deal of American energy is spent arguing about whether a war was justified. That argument matters enormously before the decision and very little afterward.

Once American forces have removed a government, the country that was there is gone and something will replace it. That is now true regardless of whether the invasion was wise, regardless of who voted for it, and regardless of what any citizen believed at the time. Whether you supported the war becomes irrelevant the moment it begins. What replaces the old order will be built either by us and the people who live there, or by whoever is willing to fill the vacuum — and the second group is never short of volunteers.

So there is only one direction from that point, and "we should not have gone" is not a policy. It is a position in a debate that ended when the first order was given. The obligation created by breaking a state is not contingent on having been right to break it.

1.3   Our own share of the failure

This is not a partisan failure and it cannot honestly be described as one. Five presidents of both parties presided over Iraq and Afghanistan and none of them made the investment that Japan, Germany, and South Korea received. A failure that survives that many changes of party is not a failure of administrations. It is a failure of the country.

The honest accounting is that Americans wanted these wars cheap. We were willing to fund the deployment and unwilling to fund the schools. We demanded results on a timetable no society has ever met and lost interest when they did not arrive. We asked our soldiers to hold ground for twenty years while declining to build anything behind them worth holding.

That impatience is ours. It belongs to the voters who rewarded whoever promised the quickest exit, and to those of us who advocated for these commitments and then stopped paying attention to how they were carried out. Until we say so, our advice on the next one is worth what our attention to the last one turned out to be worth.

Section 2The doctrine we already had

Nothing proposed here is new. The country worked it out after Vietnam and then declined to follow it.

In November 1984, Secretary of Defense Caspar Weinberger set out six tests for the use of American force: vital national interests must be at stake; the government must be willing to commit the resources needed for victory; political and military objectives must be clearly defined; deployments must be continually reassessed; the support of the American people and Congress must be secured through candor; and force must be a last resort.1

Colin Powell, Weinberger's military assistant at the time and later Chairman of the Joint Chiefs, added two conditions: a clear exit strategy and international support. Powell's emphasis fell on decisiveness. Speaking after the Gulf War, he put it plainly: if war becomes necessary, you must do it right, go in massively, fight so as to keep casualties to a minimum, and go in to win.7

The 1991 Gulf War is generally regarded as the doctrine applied. It was set aside during the later Clinton years to permit humanitarian intervention, and rejected outright by the civilian leadership of the Defense Department before 2003.8

2.2   Where we depart from Powell

Honesty requires noting that this paper does not simply restate the doctrine. It amends it at one point, and the amendment is smaller than it first appears.

Powell's sixth condition was a clear exit strategy. We do not disagree that there must be one. We disagree about the unit of measurement.

An exit strategy measured in military objectives — the enemy army destroyed, the capital taken, the regime removed — is satisfied long before the country is safe to leave. Iraq met that standard in three weeks. The exit condition that actually matters is generational: a cohort has grown to adulthood without war, has something worth defending, and is prepared to defend it. At that point the mission ends, because the peace can hold itself.

This is not an open-ended commitment and we reject that characterization. It is a condition-based one, and the condition is checkable. Are the children in school? Do the courts function? Does the next generation regard the peace as theirs rather than as something imposed on them? Those questions have answers, and a citizen can evaluate them without a security clearance.

Japan, Germany, and South Korea are the proof that the exit occurs. American forces remain in all three, but the mission ended decades ago. What ended was the occupation; what continues is an alliance those countries want. That is what a successful exit looks like — not the withdrawal of troops, but the moment the presence stops being an occupation and becomes a partnership. We have never reached that moment anywhere we have gone since.

Section 3What the historical models actually show

Japan, Germany, and South Korea are the cases most often cited, and they do support the argument. American forces have remained in all three for generations. All three are prosperous democracies and reliable allies. Where the United States stayed and invested, the outcome held.

One thing should be said plainly before the argument goes further, because the story is usually told badly in both directions. The work was done together.

Germans, Japanese, and Koreans rebuilt their own countries, at a cost in labor and privation Americans did not bear. The occupation of Germany was a four-power undertaking. The Marshall Plan ran through sixteen European governments and a body those governments created themselves, which became the OECD. Britain, France, Canada, Australia and others carried a share of all of it. Anyone who tells the story as America single-handedly repairing the world is not describing what happened.

But America was a great deal more than a bank with an army attached, and the opposite error is just as false. The generation that produced the Marshall Plan produced a design: aid as grants rather than debt, so that recovery did not begin under a mortgage. A requirement that the recipients jointly determine their own needs, which forced former enemies to sit in one room and plan together. Counterpart funds, so American money pulled local money alongside it rather than replacing it. Conditions requiring participating countries to open trade with one another rather than with us. A four-year sunset with declining annual funds, so the program was built to end. In Japan, a constitution, land reform, and the extension of the franchise to women.

None of that is banking. It is a theory of how a free society is put back together, and it was thought out in advance by people who had every excuse not to bother. Consider what it took to extend that hand: the United States offered this to Germany and to Japan three years after those countries had killed hundreds of thousands of Americans. That is not a boast about strength. It is a fact about character, and it belongs to a generation that deserves the credit for it.

So the claim of this paper is not that America fixes countries, and not that America merely funds them. It is that America once knew how to make it possible for a country to fix itself, worked shoulder to shoulder with people who did the fixing, and has not done that in a long time.

But a paper that stops there is not making an argument; it is selecting evidence. The differences between those cases and Iraq are substantial and we will state them.

They were states before they were occupations. Germany and Japan had unified national identities, functioning bureaucracies, industrial economies, and legal traditions. The occupation rebuilt institutions that had existed and could be restored. Iraq's national identity was contested along sectarian and ethnic lines that predated the invasion and survived it.

They surrendered unconditionally after total war. Both societies had been comprehensively defeated and knew it. There was no organized insurgency during either occupation, because the will to resist had been exhausted by the war itself. Iraq's regime collapsed in three weeks and its army dispersed rather than surrendered.

A common enemy made the American presence welcome. Soviet power gave Germans, Japanese, and South Koreans a reason to want American troops on their soil. That alignment of interests cannot be manufactured.

Korea is not an occupation at all. The American presence there follows an armistice along a defined border against an external adversary. It is a garrison, not a transformation, and it tells us little about governing a hostile population.

What follows from this is not that the models are useless. It is that they specify conditions. Where a defeated society retains usable institutions, accepts its defeat, and perceives a shared threat, a long American presence can secure a durable peace. Where those conditions are absent, duration by itself buys time and nothing else. That is a more demanding standard than "stay until the job is done," and it is the one the evidence supports.

Section 4Iraq: what actually happened

4.1   The decisive errors, May 2003

The invasion itself was conducted with overwhelming force and the regime fell in weeks. What followed was not a failure of duration. It was a failure of design, and it can be dated.

Paul Bremer arrived in Baghdad on 14 May 2003. On 16 May he issued Coalition Provisional Authority Order Number 1, barring Ba'ath party members from public employment and removing the top tier of Iraq's administrative class. On 23 May he issued Order Number 2, dissolving the Iraqi security forces — approximately 385,000 in the armed forces, 285,000 in the Interior Ministry, and 50,000 in presidential security units.2

The order was issued against the advice of the U.S. Army, the CIA, and prewar planning by the State Department and the Army War College, all of which had recommended retaining the Iraqi army for internal security. Jay Garner's team on the ground, which had been negotiating with Iraqi officers willing to serve under coalition authority, regarded the order as undoing their work.3 American psychological operations had dropped leaflets before the invasion promising Iraqi soldiers a role in the new army if they did not fight; by mid-May, some 137,000 had registered to return.

The effect was to create, in a single week, a large population of trained, armed, unemployed men with a grievance. By the autumn of 2003 the insurgency — composed substantially of former regime elements and disbanded soldiers — was fully under way.3

This matters for the paper's thesis in a way that strengthens rather than weakens it. The problem was never only that we left. It is that the institutions required to hold the country were dismantled by us at the outset, which meant that no length of stay would have been sufficient without rebuilding what had been destroyed on purpose.

4.2   The withdrawal, and who decided it

In November 2008, after nearly a year of negotiation, the United States and Iraq concluded a Status of Forces Agreement. Signed by President Bush and ratified by the Iraqi parliament, it required American combat forces out of Iraqi cities by June 2009 and all U.S. forces out of the country by 31 December 2011.4

Efforts to negotiate a residual force after 2011 failed, principally over Iraqi unwillingness to grant American troops legal immunity — a precondition of U.S. basing everywhere. James Jeffrey, the American ambassador to Iraq from 2010 to 2012 and an advocate of keeping troops there, wrote afterward that the failure had roots reaching well beyond the administration then in office.9

We set this out because the partisan version of the story is false in both directions, and because a paper arguing for strategic patience cannot itself be an exercise in blame allocation.

4.3   What followed

After the withdrawal, sectarian governance in Baghdad accelerated. Competent commanders in the Iraqi security forces were replaced with politically loyal ones; training and combat capability degraded accordingly.10 ISIS, an offshoot of al Qaeda in Iraq, rebuilt in the space this created, took Fallujah and parts of Ramadi in January 2014, and took Mosul that June. The Iraqi army could not mount a counter-offensive.5 American forces returned in 2015.

Section 5What "stay" actually means

The word does most of the work in this argument and it is routinely misheard. To stay does not mean to garrison. A garrison holds ground. It does not build a country, and twenty years of holding ground is not patience — it is a long way of doing nothing.

To stay means two things at once. It means keeping the peace, so that ordinary people can rebuild without warlords, militias, and men with guns deciding who eats. And it means investing in what a society actually runs on: schools, universities, courts, hospitals, ministries that function, police who are paid, churches and mosques and synagogues and temples left alone to serve their people, trade with neighbors, and open movement between our countries and theirs.

5.1   We did not spend too little. We spent it on ourselves.

The common defense of American conduct is that the money was there. It was. Total U.S. assistance to Iraq through fiscal 2006 was roughly a fifth more than the assistance given to Germany between 1946 and 1952 in constant dollars, and about double the amount given to Japan.11

The difference is not the sum. It is where the sum went, and the point holds inside a single set of books. The Special Inspector General for Afghanistan Reconstruction accounts for approximately $148.2 billion appropriated for Afghan reconstruction between 2002 and 2021. Of that, roughly $88.8 billion — about 60 percent — went to the security sector.6 Nearly two dollars in three of the money designated for rebuilding a country was spent on men with rifles.

Afghan reconstruction appropriations, 2002–2021

Category	Approximate
Security sector	$88.8B
Governance, development, humanitarian, and all other	$59.4B
Total appropriated for reconstruction	$148.2B
Of which lost to waste, fraud, and abuse	$26–29B

Source: SIGAR final report, 2025. The waste estimate spans 1,327 documented cases.6

Reconstruction was the smaller account. Separately from it, the Department of Defense spent on the war itself — troops, equipment, logistics, operations — and those are distinct appropriations rather than the same dollars counted twice. Estimates of the combined total over twenty years run to roughly $2.3 trillion, with figures near $946 billion cited for direct spending of which the overwhelming share went to U.S. military outlays.12

We use the conservative figure. The argument does not need the larger one. Even confined to money Congress designated for rebuilding Afghanistan, three fifths went to security forces that dissolved in days. That is not an investment in a country. It is the cost of an American army being in a country, which is a different thing, and it purchased what one would expect: a security apparatus with no society underneath it capable of sustaining it.

5.2   What the successful cases actually received

The Marshall Plan was not simply larger. It was built differently. It committed roughly 5 percent of American GDP and 13 percent of the federal budget across four years. More than 90 percent came as grants rather than loans. It required recipient governments to match dollar aid with local currency deposits, ensuring the money stimulated domestic activity rather than replacing it. It was conditioned on removing trade barriers between participating countries, and the body created to administer it became the OECD.11

Read that list again as a description of intent. Grants rather than debt. Local partnership rather than direct administration. Trade integration rather than dependency. A four-year sunset with declining funds, which forced the building of things that could stand on their own. It was designed to produce a country that would not need us — and then the bases stayed anyway, because by then both sides wanted them there.

Germany, Japan, and South Korea did not become allies because American soldiers were present for a long time. They became allies because their own citizens built something worth defending, and because the terms of American help were designed for exactly that — to work beside them rather than over them. Counterpart funds required local money beside American money. Trade conditions required them to open to one another rather than to depend on us. The bases are a consequence of the peace those countries made, not its cause.

5.3   The cost of the half measure

Set the counterfactuals aside and observe only the sequence. The United States broke a state, dismantled the institutions that held it together, remained for eight years without rebuilding them, departed on a published timetable, and returned four years later to fight an organization that had grown in the vacuum.

Every stage of that was more expensive than doing it properly. The insurgency that Order 2 helped create killed American soldiers for years. The organization that filled the vacuum killed Iraqis in numbers that dwarf the invasion. The return in 2015 cost money and lives that a sustained investment would have cost anyway, and bought a worse outcome. Nobody saved anything.

Section 6And it pays for itself

Everything above rests on the argument in Section 3, which is a moral one and stands on its own. Peace built properly would be worth building if it never returned a dollar.

It happens to return a great many. We set this out second and deliberately, because a case that leads with returns has conceded that the question is whether the investment pencils out — and invites the reply that we should abandon the allies who do not pay. That is not our position. But if a reader will not build a school for the moral reason, he should at least be shrewd enough to build it for the economic one.

6.1   Prosperous allies are customers

A country that becomes prosperous buys American goods. A country left broken does not, because it cannot. The prosperity is theirs, earned by them; the trade is ours to gain or lose depending on whether we helped make it possible or walked away before it was.

U.S. two-way goods trade, most recent full year

Country	Total	Per person
South Korea	$180B	~$3,460
Germany	$217B	~$2,580
Japan	$209B	~$1,685
Iraq	$7.9B	~$172
Afghanistan	$0.046B	~$1.10

Sources: U.S. Trade Representative country summaries; Bureau of Economic Analysis. Figures are goods only and rounded; Iraq's total is roughly 80 percent American purchases of Iraqi oil.13

American goods trade with Afghanistan in 2025 totaled $46 million. With Japan it is on the order of two hundred billion. The ratio is not a rounding difference; it is roughly four thousand to one. And the Iraqi figure flatters us, because most of it is oil we buy rather than goods we sell — American exports to Iraq run under $2 billion against roughly $73 billion to Japan.13

Per head the gap is starker still and does not depend on which countries a reader chooses to compare. American trade runs to roughly $3,460 a year for every South Korean and about $2,580 for every German. For every Afghan it is one dollar and ten cents.

We do not claim that all commerce with Germany and Japan is attributable to the occupation; some would have occurred regardless. We claim something narrower and harder to dispute: a country that becomes prosperous and friendly buys from us at a scale a country that becomes hostile and poor cannot approach, and the difference over seventy years exceeds anything spent on the rebuilding.

6.2   Allies pay to host us

The presence is not a gift we extend. Japan contributes roughly half and South Korea roughly 40 percent of the cost of the American military presence in each country, through direct payments and in-kind support. Japan's own government puts its annual contribution near $5.4 billion across the Special Measures Agreement, land rents, and facility construction. The Government Accountability Office has audited these arrangements in detail.14

6.3   The bases are cheaper than the alternative

Removing them would cost more than keeping them. If the United States gave up its carrier battle group base in Japan, maintaining the same Pacific presence would require several additional carrier battle groups — an estimated $50 billion. Analysis of South Korea suggests the United States may spend less stationing those troops there than it would basing them at home.14

Set that beside the alternative history. A hostile or Soviet-aligned Japan and Germany would have required a defense posture costing many multiples of the occupation, indefinitely, with no host paying half.

6.4   They finance our debt

Japan is the largest foreign holder of United States Treasury securities — roughly $1.1 to $1.2 trillion, approximately 13 percent of all foreign holdings of publicly held federal debt.15 Germany repaid its occupation loans.

Read that against Issue Paper No. 1. The country we defeated, occupied, and rebuilt is now the single largest foreign financier of the American deficit. Whatever was spent in Japan between 1945 and 1952 has been returned to the Treasury many times over, by a creditor who exists in that form because of what was built there.

6.5   What the comparison does not show

Vietnam is the case that complicates this and we will not hide it. American trade with Vietnam is now among the largest of any partner, despite a war that ended badly and an occupation that never happened. Commerce, it turns out, can resume without any of this.

What it took was roughly twenty years of no relationship at all, following economic liberalization Vietnam undertook itself. And what it produced was a trading partner rather than an ally: Vietnam hosts no American bases, buys little American defense equipment, and finances little American debt. Japan and South Korea produced both. The lesson is that commerce is recoverable and alliance is not, which is an argument for finishing the job rather than against it — but it is a narrower argument than the trade figures alone suggest.

Section 8What the standard should be

Before American forces are committed to transform a state rather than merely to destroy a capability, five conditions should be satisfied. Any one of them failing should be sufficient to decline.

Section 9The argument you can carry

The paper compressed to what a person can remember and repeat.

And the one that is ours rather than theirs. Five presidents of both parties presided over these commitments and not one of them built what Japan and Germany received. That is not a partisan failure; it is a national one. Americans wanted these wars cheap — willing to fund the deployment, unwilling to fund the schools, impatient for results on a timetable no society has ever met. The soldiers did what was asked. The country did not.

Section 10Conclusion

The men and women who fought in Iraq and Afghanistan did what was asked of them. They took the ground, and in Iraq they took it twice. What they were not given was a country willing to build anything behind them.

A generation is the honest unit of measure, and it is not a figure of speech. It takes that long because hatred is extinguished by children who grow up without it and then defend the peace they were raised in. That is also the exit condition. Not a date on a calendar and not the fall of a capital, but the moment a country's own young people regard the peace as theirs and are prepared to keep it. There is no shorter route, and any policy promising the result on a faster schedule is promising something no one has ever delivered.

What we have done instead is build tomorrow's enemies. We remove a government, decline to replace it with anything that works, leave on a published timetable, and then act surprised when what fills the vacuum comes looking for us. Then we do it again somewhere else, and the world watches us do it, and draws the obvious conclusion about what an American commitment is worth.

That is not the reputation this country once had. America was known for a time as the nation that comes to protect people who need protecting, stays until the work is finished, and does the work alongside them. Germans and Japanese and Koreans rebuilt their own countries — and they did it inside a design Americans had thought out in advance, funded with grants rather than debt, and offered to enemies three years after the shooting stopped. That reputation was an asset worth more than any weapons system, and we have spent it.

It is a strange thing that the country which invented Superman so often abandons his ethos. The character was never about the strength. It was about what a person with that much power owes the people who do not have it, and about staying until everyone is safe rather than until the fight is won. We have the strength. What we have mislaid is the second half.

The test should be applied before the first order is given, and it is simple. Are we prepared to go in with everything, name the objective plainly, fund the building as seriously as the fighting, and stay until the children there grow up in peace? If the answer is yes, then go, and do not stop short of it. If the answer is no, say so honestly beforehand. What we owe the people we send — and the people whose countries we enter — is that the question be answered before the fact rather than discovered afterward.