The United States spent years turning lessons from Iraq and Afghanistan into a formal framework for mitigating civilian harm, then stepped back from an international effort it helped create. On September 10, allied nations gathered at NATO headquarters in Brussels for a two-day summit of the International Contact Group on Civilian Harm Mitigation and Response, with no U.S. delegation in the room.
The absence followed a much larger rollback. Defense Secretary Pete Hegseth dismantled much of the Pentagon’s Civilian Harm Mitigation and Response program, with dedicated staffing falling from nearly 200 people to about two dozen. A program that had been presented as a model for allied militaries was suddenly operating at a fraction of its previous size.
The international group dates to 2023, when the United States and the Netherlands formed a loose coalition to exchange practices and briefings on civilian harm. An official 2024 Defense Department readout lists Austria, Australia, Belgium, Denmark, Finland, Germany, the Netherlands, Norway, the United Kingdom and the United States among participants. ProPublica reported that the group has met roughly half a dozen times since 2023 and continued under U.K. and Dutch leadership after the United States stopped participating.
The American program did not emerge in the abstract. It followed years of civilian-casualty incidents, investigations and reviews from the post-9/11 wars, particularly Iraq and Afghanistan, as military officials tried to turn lessons from those conflicts into enduring procedures rather than relearn them operation by operation.
The Pentagon itself made the strategic argument explicit. Its 2022 Civilian Harm Mitigation and Response Action Plan said protecting civilians was critical to long-term battlefield success and created new institutions and processes intended to improve learning, investigations, reporting and operational decision-making.
The plan called for a Civilian Protection Center of Excellence, standardized civilian-harm reporting and data systems, improvements to assessments and investigations, training across the joint force, and closer integration of civilian-harm considerations into cooperation with allies and partners. Advisers were also embedded with regional commands to help translate those principles into operations.
Top military commanders reportedly objected when Hegseth moved to cut the program. Their objections did not stop the reductions.
The disagreement is not simply humanitarian versus military. The Pentagon’s own 2022 plan treated civilian protection as part of operational effectiveness, arguing that tactical successes can produce strategic failure when the civilian environment is not adequately protected.
That matters because removing dedicated staff does not eliminate the underlying responsibilities. It reduces the institutional capacity devoted specifically to collecting data, identifying patterns, advising commanders and turning findings into lessons that can be applied in later operations.
The Dutch experience helps explain why the international coalition formed. In 2015, a Dutch airstrike in Hawija, Iraq, targeted a car-bomb factory and ignited a nearby munitions depot. At least 70 civilians were killed and hundreds were wounded. Dutch responsibility was not publicly revealed until 2019, prompting a national scandal and accusations of a cover-up.
That episode was part of the context in which the Netherlands and the United States began coordinating more closely on civilian-harm mitigation. Both militaries were trying to turn damaging experiences from earlier operations into systems that could reduce the risk of repeating them.
The current handoff is therefore less a transfer of an American invention than a shift in who is carrying a shared body of work. NATO adopted its own Protection of Civilians policy in 2016, six years before the Pentagon’s CHMR Action Plan.
Andrew Hyde, a senior fellow at the Stimson Center who works on U.S.-European relations and NATO, told ProPublica that the alliance has continued developing civilian-protection work without American participation and remains open to U.S. re-engagement. He also noted that NATO’s policy predates the American model even as the alliance has incorporated elements of the newer U.S. approach.
The U.K. now faces its own institutional gap. A 2026 review of the British Ministry of Defence’s civilian-harm framework found that the U.K. does not have a single coherent CHMR policy and recommended stronger systems for tracking allegations, investigations, open-source analysis and post-harm response.
That makes the shift in leadership complicated. The United Kingdom and the Netherlands can keep the contact group active, but several of the participating governments are still developing or formalizing parts of the systems the Pentagon spent years building.
There is also a diplomatic cost. Former Defense Department official Jenny McAvoy told ProPublica that the U.S. decision to acknowledge lessons from civilian-harm cases helped officials in other countries argue for reforms of their own. American participation gave allied governments political cover to say their militaries could improve too.
There is a competing view, and it deserves an honest hearing. Hyde has argued that the international effort could become more durable if it is no longer dependent on the political direction of a single U.S. administration. A framework that continues through an American withdrawal may ultimately become more genuinely multilateral.
That is the optimistic read. Its test will be whether participating governments sustain staffing, expertise and political attention after the country that once supplied much of the momentum has stepped back.
Denmark and Norway, along with several other allies, have participated in the contact group according to official Defense Department records. Participation matters, but attending a coordination forum is not the same thing as reproducing the personnel, data systems, training infrastructure and institutional processes that the American CHMR program was designed to connect.
What the Pentagon cut is therefore worth describing precisely. CHMR was not a single diplomatic office. It was an operational framework that included a center of excellence, advisers, reporting and data processes, training requirements, assessment mechanisms and policy structures intended to keep civilian-harm lessons inside the institution.
The Pentagon has maintained that civilian protection remains part of operational planning and that the department continues to fulfill its civilian-harm responsibilities. The question is not whether every civilian-protection function vanished overnight. It is how much capacity remains after dedicated staffing fell so sharply.
Institutional knowledge is expensive to accumulate and easy to scatter. The people who worked inside dedicated CHMR roles carried experience in interpreting strike data, assessing allegations, advising commanders and coordinating with allied counterparts.
A future administration can rebuild offices and restore billets. Rebuilding accumulated judgment, working relationships and familiarity with years of cases is slower.
That is the larger institutional issue. Structures created through repeated operational learning can be reduced much faster than the expertise behind them can be regenerated, particularly when experienced personnel move into other jobs or leave the work entirely.
The empty American seat in Brussels was not, by itself, the end of civilian-harm mitigation among NATO allies. The group met without the United States, and the U.K. and the Netherlands continued the work.
What changed was the American role. A country that had spent several years encouraging allies to formalize civilian protection was no longer at the table when those allies compared how to carry the work forward. Re-entering that conversation later is possible. Recovering the institutional capacity surrendered along the way may be harder.