The MSF Airstrike Report: Better on the Facts Than on the Law

By: Professor David Glazier
This op-ed originally appeared on Just Security.

The military’s investigation of the October 2015 airstrike on the Médicins Sans Frontières (MSF) hospital in Kunduz, Afghanistan was back in the news last week thanks to highly speculative accounts that unidentified Afghans might have manipulated US forces into attacking the facility. While there is little doubt many Afghan officials harbored substantial resentment over MSF’s willing treatment of Taliban fighters, to my reading, the investigation report logically discredits this conspiracy theory.

The facts show that Afghan personnel requested the attack by providing geographic coordinates corresponding to a legitimate military objective, an Afghan National Directorate of Security facility, under the control of Taliban forces. It was only due to a series of cascading, and wholly unforeseeable, breakdowns in US communications, equipment, and procedure, that the aircrew inadvertently selected the hospital — located approximately 1,500 feet away from the intended target — as the location to be struck. To conclude that one or more Afghan officials cunningly manipulated the aircrew into attacking the wrong target based on a deliberate misdescription in the fog of a multi-party (and presumably multi-lingual) relay of information at oh-dark thirty in the midst of a multi-day battle in order to have the MSF facility struck simply defies logic. Particularly given that they’d be diverting the fire away from the site that they were tasked to assault, significantly increasing the risk to their own lives by failing to “prepare the battlefield.”



But while the investigation seems to do a decent job of developing the facts, as others have already touched upon (here, here, and here, for example), its handling of the relevant law is much less credible, particularly its consideration of the international law of armed conflict. Without providing any source information, or identifying any specific war crimes, a US Central Command summary accompanying the report simplistically concluded that none were committed because there was no intent on the part of US personnel to strike a protected medical facility; all those involved mistakenly assumed they were striking the lawful target identified by Afghan forces. Yet despite the identification of egregious violations of the effective US rules of engagement, there is no serious discussion of possible prosecution for unlawful use of force not constituting a war crime per se. These factors demonstrate real reason for concern about the current state of operational legal knowledge among deployed US forces — a matter not helped by the fact that essentially all comments by the two Judge Advocates involved in supporting and reviewing the report are redacted, as are their identities.

Missing Foundational Analysis

In assessing whether a war crime had been committed, a modern student or practitioner of international criminal law would logically begin by seeking to establish satisfaction of the requisite “chapeau elements” — the existence of an armed conflict, its classification as international or non-international, and the relationship of the conduct to the conflict. The first and last of these elements are fairly self-evident in this case; the classification of the conflict is more complex.

Although virtually all commentators agree that fighting in Afghanistan began as an international conflict (i.e., one between states), it logically transitioned to a non-international one (e.g., a conflict between at least one state and at least one non-state group) following the establishment and international recognition of the Karzai government. This matters because the law of armed conflict applicable to non-international fights is much less extensive than for the international variety, as is the set of apposite war crimes. The Rome Statute of the International Criminal Court, for example, defines 34 war crimes for international conflicts but only 19 for non-international ones. But neither the report itself, nor the Central Command summary, address this issue, and both employ terminology characteristic of international fights.

Reliance on Inappropriate References

The list of references relied upon by the report’s authors, buried on pages 118 and 119 of the 120-page document, give further cause for concern about the soundness of the legal analysis. The investigation relies upon two treaties, the First Geneva Convention of 1949 (which covers the sick and wounded in the field) and Additional Geneva Protocol I of 1977, despite the fact that both are applicable only to international conflicts. Moreover, the United States is not a party to Protocol I and has never definitively explained which provisions of that agreement it recognizes as customary international law, so it is a particularly odd reference for this investigation.

The report also relies on two US military legal manuals. Although, as Adil Haque has previously noted, the report does not conform to legal positions controversially adopted by the Defense Department’s Law of War Manual issued last summer, that volume is nevertheless listed as a resource relied upon during the analysis. So, too, is the 2011 edition of the Operational Law Handbook informally published by the Army Judge Advocate General’s School, even though it bears the disclaimers that it “is not intended to represent official U.S. policy regarding the binding application of varied sources of law” and that it “is not a substitute for official references.” Further, as the Library of Congress website notes, it is reissued annually and “[t]he current edition supersedes all previous editions.”

Neither of these manuals actually provides comprehensive guidance on the current scope of recognized war crimes — either those included in the Rome Statute, or those which may additionally be sourced to current customary international law. Indeed, although the DOD manual does individually identify a number of war crimes interspersed in its 1,200 pages, it also confusingly references an erroneous statement found in the Army’s badly outdated (but not yet superceded) 1956 law of war reference publication, which says that “any violation of the law of war is a war crime.” So it shouldn’t be surprising that serving Judge Advocates would be unclear about the contours of contemporary war crimes law.

Overlooked US Military Law

What is less understandable, however, is why the report would not explicitly consider the possibility of criminal responsibility on the part of the Ground Force Commander (GFC) for the deaths at the hospital under the Uniform Code of Military Justice. As both manuals note, longstanding US practice relies on charging American military personnel with underlying military law offenses, such as murder or dereliction of duty, rather than war crimes per se.

Careful reading of the report makes it clear that the rules of engagement (ROE) applicable to US forces in Kunduz were far more restrictive than the law of armed conflict mandates, limiting airstrikes to the actual defense of US or Afghan forces. The report makes clear that there was no permissible basis at the time for calling for fire on any target, let alone a hospital. The Afghan forces tasked with assaulting the National Directorate of Security were still nine kilometers away and pre-assault “softening” of the target — although fairly standard practice in conventional conflict — was clearly prohibited by the ROE in effect.

So, if the GFC had no legal authority to call for an airstrike (yet affirmatively misrepresented to the aircrew that they did) as the report concludes, it seems logical that this would at least constitute the “culpable negligence” for resulting deaths which constitutes manslaughter under UCMJ article 119 (b). While global public outcry over the incident has focused on generic calls for prosecution for perceived war crimes, it seems unlikely that the conduct described in the report falls within the recognized definition of any specific war crimes. But the flip side of US reliance on UCMJ offenses rather than international criminal law when disciplining our own military personnel is that military criminal prosecution may well be possible, and appropriate, even when a trial based on the law of war per se is not.

* * *

In sum, the investigating officer seems to have done a credible job in sorting out what happened, and specifically what went wrong, during the airstrike, but understandably lacked the legal knowledge to correctly assess the legal violations represented by that conduct. That information should have been provided by the supporting and reviewing Judge Advocates. Unfortunately, the extensive redactions in the legal review makes it impossible for the outside reader to assess the quality of that support.

But this incident also highlights the failure of the Defense Department to ensure that its personnel have current, and comprehensive, legal guidance on the contemporary law of armed conflict, including a clear understanding of the full scope of recognized war crimes and requisite elements of proof.