Lack of robust assessment of civilian casualties enables terrorist propaganda, radicalization, and decimates US legitimacy – threatens national security
Lewis 19 – Dr. Larry Lewis, head of the CNA’s Center for Autonomy and AI, previous Senior Advisor to the Assistant Secretary of State responsible for human rights, lead analyst and co-author for the Joint Civilian Casualty Study in support of GEN Patraeus, subject matter expert on the subjects of joint and coalition interoperability, combat identification, and fratricide, Ph.D. in physical chemistry from Rice University (“Promoting Civilian Protection during Security Assistance: Learning from Yemen”, May 2019, https://www.cna.org/reports/2019/05/IRM-2019-U-019749-Final.pdf) FGY
Information gap: preventing an evidence-based approach During the US advisory effort to the coalition in 2015–2016, a key question was: is this assistance having a positive effect? And even more broadly, could the US determine how its support, in terms of bombs, aircraft, intelligence, and refueling, was being used? To what degree was US assistance contributing to the disaster on the ground in Yemen? How accurate were the findings of the JIAT? As the US wrestled with policy decisions, having this information would have enabled more of an evidence-based approach to US strategy in Yemen. However, there is no established assessment process to help answer these questions regarding security assistance to partners. Operational end use monitoring It is worthwhile to note that the US has some processes in place to address accountability when defense equipment, including weapons, are sold or transferred to foreign militaries. This process, mandated by the Arms Export Control Act, is called end-use monitoring (EUM).51 As part of the foreign military sales process, the receiving nation must agree to certain conditions to enable EUM. These conditions include: The equipment must only be used for the purpose intended; No equipment can be transferred to a third party without the prior approval of the US; The receiving nation must protect the equipment to US standards; and The receiving nation must allow monitoring processes to verify the conditions above.52 Several programs conducted under the EUM process include Golden Sentry and Blue Lantern. These US end-use programs focus on verifying equipment presence and security.53 This matches their intended purpose: they were developed to protect against unauthorized exports and technical espionage involving sensitive military equipment. These programs do not address how US equipment are used operationally. For example, in the Vignette on page 2 where the Saudi-led coalition struck a school bus full of children, the US government had no idea that a US-provided bomb was used until third parties performed open source analysis and disclosed that fact. While the US has no process for capturing how its equipment and weapons are used operationally – what we refer to as operational end use monitoring – this information is relevant for the US. For example, understanding how partners use US-provided weapons should inform policy and assistance decisions. This information is also relevant for foreign policy and national security, since the theme of US-provided weapons killing civilians is a frequent and powerful subject of messaging and terrorist propaganda, influencing US relations with other countries and contributing to radicalization and support to terrorist groups that can then threaten the US homeland and its interests.54 Three sources of information If a government wanted to conduct operational end-use-monitoring, how can this information be obtained? Governments providing security assistance can leverage three different types of information for operational end use monitoring. They are: Partner forces. Getting information from partner forces is invaluable for helping to understand the operational effects of assistance. For example, the Saudi-led coalition created a spreadsheet with mission reports documenting basic details on every strike conducted in the campaign, including the type of aircraft and weapon used. It is also helpful to know the rationale and intent for particular strikes. For example, if there are allegations that non-combatants were targeted, what was the intended target? What types of intelligence were used in that strike? And what is the partner force assessment for what happened? While such access is not guaranteed, this information was at times made available to the US when it specifically requested it. And greater access could be made a requirement for future assistance in order to strengthen monitoring efforts. Government information. In Yemen, the US sometimes also had intelligence or other information that could be useful in helping to determine what was happening on the ground. This was not frequent, since the conduct of partnered militaries like the Saudiled coalition was not an intelligence requirement for the Intelligence Community. It is possible that making this a priority would enhance the value of intelligence in this operational end use monitoring process. Open source information. Another rich information source for determining the outcome of operations is using open sources. This can range from civil society groups that do on-the-ground investigations to open source investigators leveraging social media. While militaries and governments are often dismissive of these types of sources, considering military data to be more reliable and accurate, the reality is that a complete picture of operational outcomes requires a merging of military and open source data. This is not limited to the Saudi-led coalition: a recent study by the US military determined that external allegations, based on open source data, were the source for the majority of the civilian casualties confirmed by the US-led coalition in Iraq and Syria. Militaries lack the full picture, and open source information can help complete that picture.