Last week, the Trump Administration decided to eliminate a section of an Executive Order requiring release of public reports on civilian casualties resulting from U.S. counterterrorism strikes. Having helped to create and apply that Executive Order during my time at the State Department, this decision seems like a good opportunity to reflect upon the whole 2016 Executive Order—“United States Policy on Pre- and Post-Strike Measures to Address Civilian Casualties in U.S. Operations Involving the Use of Force”—as well as Section 3, in particular, the one eliminated last week. What does this action on Section 3 really mean? What have the other parts of the 2016 Executive Order actually done—and not done? With both the Defense Department and Congress actively working on the issue of civilian casualties, a moment of reflection can help the government to better meet a key intent of the EO: to systematize learning and improve U.S. efforts to protect civilians.

First, the news about revoking Section 3. This section called for the United States to be more transparent regarding the civilian toll from the most sensitive U.S. operations, outside of declared areas of hostilities. This annual reporting requirement involved releasing an estimate of the civilian deaths caused by the U.S. and tallied over all areas and operations, from military actions to those of other agencies tasked with combatting terrorism. This information was aggregated across agencies and accompanied by the total number of strikes and combatant deaths. This aggregation represented the balance of a commitment to pursue transparency with a desire to preserve U.S. national security.

The decision to revoke Section 3 reduces transparency regarding a portion of the most sensitive and secretive of U.S. actions. While some have said that this move was merely a decision to reduce transparency overall, it is possible that there was a different and more specific reason. The Executive Order was drafted at a time when there was no other systematic reporting of civilian casualties. Since then, Congress has mandated that the Pentagon provide annual reporting of civilian casualties from all its operations under Section 1057 of the National Defense Authorization Act (NDAA). Section 1057 reporting includes only the Defense Department’s portion of Section 3 reporting, but the precision of information required by Congress is much higher – it includes civilians wounded as well as killed, and it is broken out by geographic areas, neither of which Section 3 reporting provided. Section 1057 is a step forward for transparency and provides a fuller account of the civilian toll from Defense Department operations. However, some significant information is lost by having to rely solely on reporting under the NDAA. For example, this reporting also is limited to Defense Department actions. Also, Section 1057 does not include a toll of estimated combatant deaths to compare against the number of civilian casualties.

When Section 3 was developed, some in the U.S. government were concerned that information provided about U.S. lethal actions would allow triangulation back to specific covert actions, including dates and locations. That concern was why Section 3 aggregated the civilian deaths to a single range of estimates encompassing all U.S. strikes.  After the addition of Section 1057 reporting, renewed concerns over triangulation could have been behind the decision to eliminate Section 3. While the Pentagon portion was largely redundant, as the Administration rightly claimed last week, the elimination of the Section 3 reporting requirement now creates a transparency gap regarding the civilian toll from any non-Defense Department lethal actions.

As Congress weighs options to address this transparency gap, it is worth considering that the Pentagon recently completed a review of its own civilian casualty accounting and found ways to improve. Such honesty and self-learning are good indicators that the Department is serious about transparency and accurate accounting of harm resulting from its actions. These are traits that should be built into any new mechanism for transparency of ALL civilian casualties, no matter which U.S. entity or agency caused them.

Taken as a whole, the Executive Order offers other good practices from the U.S. military.

For example, Section 2 contains a list of policy commitments that the United States affirms in its own operations, as well as in its work with foreign military partners. This section originated from best practices and lessons in Afghanistan. In response to a large-scale civilian casualty strike in Bala Balouk in western Afghanistan and a renewed focus on civilian casualties across the theater, the Joint Staff created a Joint Staff Civilian Casualty working group in 2009. This group, initially led by the Chairman of the Joint Chiefs of Staff, supported NATO’s civilian casualty mitigation efforts. The United States and its allies improved their ability to protect civilians in Afghanistan operations, with civilian casualties decreasing as a result.  As a reminder, the decrease in civilian deaths means lives saved due to increased care on the battlefield. Overarching lessons and best practices from this effort were captured by the Joint Staff in a 2013 report, which served as the starting point for Section 2 of the Executive Order.

Another example of inspiration from U.S. military practices in Afghanistan is Section 4, calling for monitoring of civilian casualty trends. The core idea—that militaries can monitor civilian casualties and by doing so improve efforts to protect civilians in near real time—was motivated by actual events. Following U.S. support for NATO’s efforts to reduce civilian casualties through an independent review (the Joint Civilian Casualty Study noted above), NATO sent monthly updates of its civilian casualty database to the Joint Staff, which in turn analyzed emerging trends. Have civilian casualty numbers gone up or down? Where? How?  In January 2011, analysis revealed new increases in civilian casualties for two particular types of operations. This analysis was sent to the staff of General David Petraeus (then Commander of NATO forces). NATO made operational adjustments in response. Continued monitoring showed that these increases were quickly reversed – again, lives were saved. This experience showed that militaries can rapidly respond and adapt to reduce civilian casualties. Section 4 calls for the U.S. government to similarly monitor trends in U.S. operations and make adjustments as necessary to better protect civilians.

After the revocation of Section 3, Congress is discussing legislation that will address the resulting gap in civilian casualty reporting. Filling that gap could include reporting of civilian casualties both outside of declared areas of hostilities, as Section 3 was designed to provide, but also civilian casualties from non-Defense Department actions inside areas of active hostilities.

Congress should also reflect on actions that reinforce other sections of the EO. For example, a congressionally mandated, independent review to identify key factors leading to civilian casualties in operations outside of declared areas of conflict, including lessons and best practices that can improve the conduct of those operations, would help to fulfill the policy commitments in Section 2. Likewise, Congress could call for the administration to fulfill its Section 4 commitment to monitor civilian casualty trends and make necessary adjustments. Given the sharp rise in the civilian casualty rate in U.S. operations in Iraq and Syria in 2017 – a year after the Executive Order was created – it does not appear that this commitment is being met. If the Administration is not going to meet this policy requirement on its own, Congress could create an NDAA amendment that calls for periodic reports from this monitoring and review function, say every three months, to help ensure that the U.S. government meets the requirement.

This is a good moment to reflect not only on what was lost in last week’s decision, but also what more is possible. In response to the loss of Section 3, more can be done than just pushing for the release of numbers. We can push for trying much harder.