Earlier today, Missy Ryan at the Washington Post reported on a major examination of civilian deaths in military operations underway at the Pentagon, including an internal study on civilian casualties that was completed in April but not previously released or even known to the public. The Chairman of the Joint Chiefs of Staff ordered the internal review in late 2017 at the behest of former Secretary of Defense James Mattis following several troubling media reports regarding a rise in U.S.-caused civilian casualties and serious failings of existing investigation and response mechanisms.

The internal review, however, was quite narrow in scope and, as a result, does not answer some of the biggest questions about the principal causes of civilian casualties–or their reported increase under this administration. Nor does the report resolve the broad disparity between the Pentagon’s publicly reported civilian casualties’ figures and the estimates of NGOs like Airwars that systematically track casualty reports.

But the report’s recommendations, which you can read here, do provide ample basis for the development of a more comprehensive Defense Department policy on civilian casualties, an effort that is underway now at the Pentagon. This review also validates the need for more in-depth internal research into some of the questions that this review leaves unresolved.

Below we analyze some of the key takeaways from the report’s findings and recommendations and then provide a roadmap of what to expect next given the looming statutory reporting deadlines on civilian casualties and the Pentagon’s new effort to craft a comprehensive Department-wide civilian casualties policy.

(In the interest of full disclosure, the two of us participated confidentially in the NGO advisory panel and NGO roundtables mentioned on page 1 of the report. Our participation was in the form of providing recommendations and raising issues of concern. We did not participate in the drafting of the study or the selection of its findings and recommendations and it should not be assumed from our participation as advisors that we concur with the study’s scope, methodology, findings, or recommendations.)

Scope and Organization of the Report

The newly released report summarizes the findings and recommendations of a study carried out by the National Defense University’s Institute for National Strategic Studies “to examine civilian casualties (CIVCAS) that resulted from US air or artillery strikes in the US Central Command (CENTCOM) and US Africa Command (AFRICOM) Areas of Operation (AOR) from 2015-2017.”

The report is then further limited to five categories of analysis (or tasks): A) Guidance, Intent, and Oversight; B) Internal Reporting Procedures; C) Reconciliation and Verification of External Reports; D) Investigations; and E) Response, including Solatia. Within these categories the study addresses more specific “sub-tasks,” which provide much needed clarification of the “purpose” of each task. For each of these categories the report contains both findings and corresponding recommendations.

All told the report is only about two dozen pages, not counting the redacted appendix containing the Secretary of Defense Level Rules of Engagement. Given the bevy of complex issues that must be addressed to develop an effective and comprehensive civilian casualties policy across the Department, the length of the report alone makes it clear that this first study is just the start of what is needed to produce real results.

Key Positive Takeaways

  • The report calls for proactively seeking additional sources of information when assessing internal reports and incorporating civilian casualties in battle damage assessments (Recommendation 4).
  • Recommendation 5 suggests standardizing the civilian casualties assessment processes within the Combatant Commands. The same recommendation includes a proposal for “civilian casualty review boards” to provide feedback and lessons learned to pilots, analysts, and ground force commanders. (Such a body could be valuable, but should not, as the report suggests, take the place of formal investigations, especially given the fact that assessment reports seem to have replaced the administrative investigation as a means of preliminary inquiry already.)
  • The report calls on the Joint Staff to develop standard, but flexible, protocols for responding to known cases of civilian casualties (Recommendation 9). The report rightly acknowledges that the range of options available in such cases includes, but extends beyond, solatiapayments, to apologies and explanations, in-kind offerings, and clearing of the family name of victims.
  • The report’s first two recommendations call for clarifying guidance and doctrine related to the risk of civilian casualties in “by, with, and through” activities, which many groups, including our own, have called for in the past (oddly the body of the report neglected to include any analysis or justification for these recommendations, unless it has been redacted).
  • The report also sheds new light on the sequence of steps taken to assess internal and externally-sourced reports of civilian casualties, to include the initial assessment and the credibility assessment process.

Gaps and Shortcomings Meriting Further Review

  • The report contains a “finding” that NGOs are frustrated by a perceived lack of transparency (Finding B.3.) – with no qualitative assessment of the validity of that view (i.e., there is no actual finding on this issue).
  • Many findings are entirely redacted, such as the section on civilian casualties in urban environments (A.6.), or lack the detail needed for the public to evaluate some of the more significant claims, such as the presence of clear commander’s intent related to civilian casualties in military guidance.
  • A lack of emphasis or detail regarding the benefits and means available for engaging local groups, survivors, and witnesses directly or through intermediaries to corroborate external reports or in the response to civilian harm.
  • Too few recommendations for ways to improve communication between the military and outside sources in order to lead to more accurate outcomes, despite the report’s acknowledgement that the standards used to evaluate external information are too strict.
  • The report contains almost no analysis from Afghanistan, and does not include any “leading practices” or “lessons learned” from any theater from the last 17 years of war.

And Perhaps the Most Interesting Set of Findings

Footnote 1 of the main report makes explicit that the “report does not attempt to explain causality for the observed increase in civilian casualties between 2015 and 2017, or to explain the gap in civilian casualty numbers between the US military and NGOs.”  Yet even within these parameters, the study group could not entirely avoid these most important controversies, and some of the findings could create necessary space for addressing them.  For example:

  • One of the report’s most important findings is the conclusion that “US military standards for verifying third party allegations vary significantly, and some may be construed as restrictive” (Finding C1). This conclusion, while brief and incomplete, provides some official validation of one of the most consistent grievances of outside groups: that the military has set the threshold for considering external reports of civilian casualties too high. The corresponding recommendation (Recommendation 7) points to the need to broaden the geographic and temporal parameters used to verify external reports of inquiry (a longstanding request by Airwars) and calls for a new classification system that allows for a category of “disputed” cases that would add much-needed nuance to aggregated estimates that rest on the binary classifications of “confirmed” and “non-credible.” The implications of this finding extend far beyond just the important interest served by recognizing the value of external sources of information. A more accurate range of estimates could also change operating assumptions about the presence of civilians in future engagements, perhaps saving lives in the process.
  • The report attempts to resolve a main source of discrepancy between internal and external civilian casualty estimates, by finding that the Positive Identification (PID) procedures used in targeting decisions did not result in a meaningful increase in civilian casualties (finding A.4.). Over-classification of noncombatants as combatants remains a top concern of outside groups and experts and, interestingly, this is the only issue for which the report notes that there was internal dissent among the Defense Department’s study group members. In a footnote, the study explains that two study group members challenged this finding in part on the basis of the interview-dependent methodology of the study and the fact that those being interviewed would not perceive a problem because they would not be aware of the misidentification.

What to Expect Next  

Readers interested in this issue will want to watch out for a major report due to Congress this Friday under Section 936 of last year’s National Defense Authorization Act. This provision, detailed here, requires the Pentagon to develop a comprehensive civilian casualties policy (which is reportedly underway), to appoint a senior civilian official (which it did back in November), and to provide Congress an update on the Pentagon’s efforts at the 180-day mark. Following this 180-day update, the Pentagon has another report coming due on May 1 on the number of U.S. military-caused civilian casualties.

These two upcoming reports to Congress will be important bell weathers for what to expect from the Pentagon’s new civilian casualties guidance on preventing and responding to civilian casualties and just how serious the Pentagon is about doing, as former Secretary Mattis has urged, “everything humanely possible” to avoid harm to civilians.


IMAGE: A U.S. Air Force MQ-1B Predator unmanned aerial vehicle (UAV) returns from a mission to an air base in the Persian Gulf region on January 7, 2016. (Photo by John Moore/Getty Images)