In the war on terror and ensuing related conflicts, the United States often describes strikes killing large numbers of civilians as tragic, isolated mistakes, when it acknowledges them at all. Yet comparing and contrasting such strikes across different administrations can help us detect important institutional patterns as well as individual choices. Both the Kunduz hospital strike under the Obama Administration and the Minab school strike under the Trump Administration are important focal points through which we can begin to assess both the meaningful differences between the two administrations’ approaches to mass civilian casualties, but also the similarities. Of course, the Minab school strike is still being actively investigated, and a different approach to accountability, redress, and reform is not yet foreclosed. With the recent Defense Department Inspector General’s stern recommendation to the Pentagon to implement its legal obligations on civilian harm risk mitigation, along with increasingly vocal Congressional and public cries for accountability, as well as the possible expansion of strikes to Mali, looking at the past can help us understand paths for the future.
As I previously explained, the 2015 strike on the Doctors without Borders trauma center in Kunduz, Afghanistan, is one of the definitive incidents of civilian harm of the War on Terror. More recently, the February 2026 strike on the Shajareh Tayyebeh Elementary School in Minab, Iran, has emerged as another significant event. Although they arose in different conflicts under different administrations, both incidents illuminate enduring policy and legal questions about how the United States identifies lawful military objectives, responds when those assessments prove incorrect, and reforms its targeting practices after catastrophic failures.
Comparing the two incidents reveals both striking similarities and meaningful differences. Part I begins with a focus on the similarities, whereas Part II will continue with differences and recommendations. The similarities reveal the persistence of target misidentification issues that transcend presidential administrations and I anticipate another similarity will be the absence of significant backwards looking accountability. The differences, so far at least, regard executive leadership on accountability issues and institutional capacity to allow military mistakes to serve as an opportunity for reform. The current administration seems significantly less interested in transparency, accountability, and organizational learning than the Obama administration, which, to be clear, often fell short on these issues as well. And the Iran strike also implicates, in a way the Kunduz strike did not, and as with so many contemporary problems, the willingness of the legislative branch to demand adherence to already enacted legislative reforms.
So, to begin with, the similarities. First, the status of those injured and killed in each strike is an easy legal determination. No serious person contests that those killed are properly classified as civilians. No evidence suggests that civilians were directly participating in hostilities. In other words, both the trauma center in Kunduz and the school in Minab were protected sites with the civilians inside them engaged in purely civilian activities. Nor, in some sense, should the locations have posed particularly difficult intelligence gathering issues. The Doctors without Borders trauma center had been operating since 2011, and its coordinates ought to have been well known to those engaged in targeting in 2015. Publicly available satellite imagery shows Iranians had converted the site into the Shajareh Tayyebeh Elementary School by October 2015, despite its location on what had been a base for the Iranian Revolutionary Guard Corps (which would be a permissible target when it served as such). Obvious evidence of its civilian character includes exterior walls painted pink and blue to designate girls’ and boys’ areas as well as football pitches and playgrounds.
Second, both attacks hit their intended targets, even though neither strike architect intended to target civilians. No munition failed to strike where it was aimed. Instead, in both cases, the military improperly identified the target. In Afghanistan, the military meant to strike the National Directorate of Security headquarters building, which was thought to be occupied by fighters, in order to assist the Afghan military conducting a clearing operation. That building, however, was actually located several hundred meters away. In Iran, the U.S. meant to hit the Iranian military base, a legitimate target. The Defense Intelligence Agency had labeled the building in question as a legitimate target based on its original usage, but presumably did not know the specific facility had been transformed into a primary school. This improper identification raises distinct legal and policy questions surrounding the intelligence, verification, and decision-making processes.
Thus, both incidents raise legal questions about compliance with the Geneva Convention’s requirement of constant care and feasible precautions. To take a step back, the laws of war require parties engaged in armed conflict to distinguish between combatants and civilians. Intentionally targeting civilians is unlawful. Targeting combatants and military targets is, assuming no other violations, lawful. But merely intending to hit a military target is insufficient to satisfy the laws of war. Nor, however, does the law require perfection in selecting targets. Instead, under Article 57 of Additional Protocol I, those engaged in planning or deciding upon strikes must take constant care. That requirement encompasses doing everything feasible to verify that objectives “are neither civilians nor civilian objects and are not subject to special protection but are military objectives”
Beyond the legal questions, the strikes also raise related policy questions of whether and how much to exceed or go beyond the Geneva Convention to protect civilians. States are, of course, always able to do more than what is required, whether that means something more substantial than feasible or adding different protections in other parts of the use of force chain, such as weaponeering or other methods to reduce civilian casualties, such as engaging in improved after-attack assessments to identify systematic failures or weaknesses.
Third, in both cases, the intended targets were not merely civilian objects, but ones given specific mention in the Geneva Conventions and their Additional Protocols, reflecting a judgment that both are institutions essential to civilian life and flourishing. The Geneva Conventions provide special protections to medical facilities under Article 18 of the Fourth Convention, and schools are presumed civilian objects under Article 52 of Additional Protocol I. Such enhanced protection reflects a judgment that states ought to be particularly careful not to strike such locations absent reliable information that they have become legitimate military objectives. For functioning medical facilities, the standard is so high under Article 19 of the Fourth Convention that even if such facilities are used to commit “acts harmful to the enemy,” opposing parties may not attack until they have provided due warning. For schools, under Article 52, even in cases of doubt as to whether it is making an effective contribution to military action, it must be presumed to retain its civilian status.
Fourth, both strikes resulted in extensive civilian casualties, making them focal points for evaluating each administration’s approach to civilian casualties with substantial coverage in traditional media sources, social media, and legal commentary. In Kunduz, the United States ultimately acknowledged 42 killed and 229 wounded. In Minab, the United States has not yet confirmed the number killed, but Iranian authorities say 120 children and 36 others have been confirmed killed. The tremendous scale of civilian harm renders each event not merely a mistake but, representative or not, a highly salient referendum on U.S. practices. Hundreds, if not thousands, of civilian casualty events from Afghanistan are only known to the American public via aggregated information referenced in popular news sources, if they are known at all. In Iran, the attacks are so recent and the opacity of administration so great (which is to damn the Obama administration with the faint praise as better on this score), that similarly most civilian casualty incidents are unknown to the public and likely policy makers as well.
The prominence of these incidents reflects more than the significant number of casualties involved. They prompt broader questions about the rules of engagement, the quality and existence of legal advice, the reliability of intelligence, the pace of decision-making and operational tempo, and the sufficiency of accountability practices. and executive decision-making. Even if these incidents might be outliers in their scope, are they similar, less high-profile mistakes occurring and/or likely to recur? Of course, while salient events are helpful for provoking such conversations and potentially reform, we should be careful not to overgeneralize. We must learn, rather than presume, whether such incidents represent significantly likely replicable institutional failures and what exactly needs to be changed.
Both incidents may have involved repeated failures to recognize and correct the initial targeting mistake. I say may as the official U.S. investigation in Minab has not been concluded, and the results have not been released. In Kunduz, in the official U.S. account, the U.S. missed several opportunities to stop the attack after it began. To take just two, it took 17 minutes from receiving a phone call from Doctors without Borders notifying U.S. forces it was under attack until headquarters, and the commander realized the mistake, and at no point did those directing the ongoing strikes take the lack of hostile activity as a red flag that they might be aiming at the wrong target. In the Minab strike, a CNN report concludes commanders ignored warnings that their intelligence was outdated and declined to revet the strike list. Moreover, at least one report suggests that enough time lapsed between the first strike and the latter two for surviving students to be moved to the prayer hall, perhaps time where the military could have realized its’ mistake, though such evidence has not yet been disclosed.
Another likely, but not yet preordained conclusion, is the lack of criminal accountability. In the wake of the Pentagon’s Kunduz investigation, sixteen members of the U.S. military including a two star general received administrative punishments. While administrative punishment can cause significant career consequences in the military, none faced criminal proceedings or criminal punishments. As we are at the beginning rather than the end of the investigation into the Minab strike, I cannot say with certainty that no criminal prosecutions will follow. But given some of the institutional changes to be noted in Part II, I’d take that bet six times a week and twice on Sunday.
All of these similarities raise questions about not just the specific strikes themselves, but also about pervasive issues surrounding target identification and verification, and the capacity for short-term reassessments. In both cases, the military’s unwillingness to change course after the initial determination led to significant and avoidable civilian casualties. In Part II, I will address some of the more troubling differences between the two strikes and what reforms I suggest ought to flow from them.