Costs of War Can’t Be Assessed Without Official Civilian Casualty Estimates

In a recent essay, retired Major General Charles J. Dunlap Jr. argues that the release of official estimates of the numbers of civilians harmed in U.S. counterterrorism operations  is more likely to “confuse” a public when it comes to assessing the “propriety of the use of force.” Yet Dunlap—who is not alone in his view—is himself misleading when attempting to taint the practice of tracking and assessing civilian casualties by association with Defense Secretary Robert McNamara’s controversial use of enemy combatant casualty numbers as a metric for success in Vietnam. As Luke Hartig clarifies in his own response to Dunlap’s piece, counting the number of enemy combatants killed as a metric of military success is not the same as counting the number of civilians harmed as a way of assessing the costs of war.

While the exchange between Dunlap and Hartig has been helpful in identifying some areas of consensus on transparency, we feel a clarification is needed to ensure that the discussion of civilian casualty estimates is not centered on “body counts” but focuses on the many policy, legal, and humanitarian benefits of having a more complete assessment of war and its costs, measured in part by its toll not just on but of civilian casualties. In this post, we highlight some of the issues neglected in Dunlap’s analysis, namely: the right of the public, and impacted communities, to demand official civilian casualties estimates; the benefits that accrue to the government with greater transparency; the relationship of civilian casualty counting with civilian harm mitigation; the continued value of the civilian casualty reporting requirement rescinded by Trump; and the need for more transparency. In so doing, we hope to contribute to the “robust debate on the parameters of transparency around U.S. counterterrorism operations” that both Dunlap and Hartig favor and that seems more important than ever at a time of increasing secrecy.

The U.S. Public—and Civilians and Communities Directly Impacted—are Entitled to an Accounting of Civilian Harm in U.S. Wars and Counterterrorism Operations

An unfortunate feature of Dunlap’s essay is its reliance on the age-old “the military knows things you don’t” argument, and the related, dangerous implication that some issues are the exclusive preserve of the military and civilians have no place commenting on them. In a democracy, information is not the preserve of those in power, but something those in power must justify withholding from the public they serve. The public is, at minimum, entitled to a basic but accurate accounting for the government’s use of lethal force in its name.

Dunlap argues that the disclosure of civilian casualty figures to the U.S. public will “likely cause more misunderstandings than coherent conclusions.” He argues that the U.S. public would effectively need access to classified information and expertise to judge civilian casualties against military advantage as part of an analysis of proportionality under international humanitarian law. Yet—and it almost seems strange to have to state what seems so obvious—the public and their representatives are perfectly justified in demanding information about civilian casualties, and they are capable of evaluating and forming opinions on such information as a broader policy and humanitarian matter. These concerns have particular resonance in the context of U.S. counterterrorism operations since 9/11, where, for many years, policymakers and officials have responded to demand signals for information with a public narrative of “clean” war.  Claims of zero “collateral death[s]”, and adjectives like “precise” or “surgicalconvey the sense that the US is able to employ lethal force with exceedingly minimal or no associated costs. The acknowledgment and release of official civilian casualty estimates from 2016 onwards was an important, if insufficient, step in giving the public a truer picture of the civilian harm associated with such operations. Dunlap also ignores the relevance of a general picture of civilian harm to an assessment of the legality of the decision to go to war and the overall war effort (ius ad bellum). This assessment requires a balancing of harms inflicted and prevented in line with the just war tradition cited as part of the philosophical foundation for the modern law of war in the Department of Defense’s Law of War Manual to which Dunlap refers.

Regardless of the legality of specific incidents, the overall harm to the civilian population is something that citizens of governments engaged in war have been concerned about on many occasions, with good reason. Dunlap argues that other factors—such as control of territory, levels of violence—may be more relevant for determining what “the military wants to achieve.” Fair enough, but surely citizens are entitled to judge military success, and the entire military strategy, by the civilian costs of these operations alongside other metrics. The impact of war on civilians is and should remain one of the primary bases upon which policy decisions about war are made, and the costs and benefits of war measured. Without official estimates of civilians harmed it is difficult for the public and their representatives to participate in debates about war. It is almost impossible to evaluate a claim that lethal force is necessary when the public lacks critical data (as we explain further below, the U.S. government and military regularly release information on the purported benefits of lethal force). It’s also more difficult to encourage legislative action that enables appropriate redress for those injured or the families of those killed in U.S. operations without a more comprehensive, and complete, articulation of the harm caused.

Military leaders as experienced as General, and former Secretary of Defense, James Mattis have indicated their understanding of the value of lower civilian casualty counts as a measure of success. In an August 2018 press briefing discussing civilian casualties resulting from Saudi actions in the war in Yemen, Mattis said, “And if what we’ve done in the past had reduced the loss of innocent life, then I would not want to stop doing that and think, ‘There, we took care of that problem,’ and watch that number go up.”

For these reasons, members of Congress continue to demand official estimates of civilian casualties from the government. Other countries have also addressed the importance of counting and publishing official civilian casualty estimates. For example, the independent inquiry into the UK’s involvement in the Iraq war (commonly known as the Chilcot Inquiry) devoted an entire chapter of its thoroughly researched report to civilian casualties, concluding “that a Government has a responsibility to make every reasonable effort to understand the likely and actual effects of its military actions on civilians” and that “greater efforts should have been made in the post‑conflict period to determine the number of civilian casualties and the broader effects of military operations on civilians.” The Chilcot Inquiry went on to emphasize, “As well as serving to minimise the effect of military action on civilians, such [civilian casualty] assessments and estimates will also enable the Government to address criticisms of the human cost of military operations.”

Dunlap also appears to overlook a key constituency when evaluating the compilation and release of official civilian casualty estimates. Most importantly, transparency in relation to civilian casualties is critical to those directly affected. Too often, the nameless, faceless civilians are deprived of their agency in discussions about the value of information about civilian harm in war. We talk about the cost of war in terms of the American public, but too rarely in terms of those who have actually paid with lives and loved ones. The release of official civilian casualty estimates is important in ensuring the legitimacy and credibility of claims by the U.S. military that it takes “stringent precautions,” goes to “extraordinary lengths to reduce the likelihood of civilian casualties,” and takes “many chances to avoid civilian casualties at all costs.” Releasing official estimates of civilian casualties allows civilians and communities some basis upon which to judge these claims. Even more importantly, the release of official civilian casualty data is an initial and incomplete first step toward accountability for civilian harm: Survivors and families of those killed in incidents involving the United States consistently call for more—including acknowledgement of strikes, explanations for what has happened, and compensation. Survivors and families of civilians killed in countries where the U.S. military is active frequently complain that they are forgotten and treated without dignity: A “body count” as Dunlap terms it, is really the bare minimum to show that civilian lives are registered, counted, and to some degree acknowledged. Hence, the widespread acclaim for the investigative report, The Uncounted.

The Government Also Benefits from Greater Transparency

The U.S. government has an abiding interest in disclosing accurate information about civilian casualties to ensure that it retains the credibility it needs to counter public claims about the harm it causes, something the military seems to generally, if inconsistently, recognize. U.S. military communications doctrine, for example, calls attention to the importance of using public messaging to bolster perceptions of legitimacy, and more generally points to the release of accurate information as an effective means of countering misinformation. As Hartig points out, the U.S. military seems “inclined towards dialogue,” and in circumstances where the military acknowledges strikes and releases information about civilian casualties, it becomes much easier to credibly defend its decisions on when and how it uses force. But the government as a whole loses ground, not only with American domestic audiences but also in the places where it is operating, when there is reason to doubt its claims—or the absence of them. As Abdulrasheed Al-Faqih explains, in relation to U.S. strikes in Yemen, “Acknowledgment of accidental civilian deaths can be a vital step toward preventing further acts of violence. … Al Qaeda and similar jihadist groups stand ready to capitalize and exploit these feelings of discontent and injustice.”

What’s more, greater transparency and attention to civilian harm also serves to bolster the military’s oft-repeated claim that it prioritizes reducing civilian casualties, not only for legal reasons but also “as a matter of policy;” and not only for the benefit of the public debate and accountability, but for its own operational benefit. As Army Techniques Publication 3-07.6 on Protection of Civilians states:

“Minimizing and addressing civilian casualty incidents supports strategic imperatives and is also at the heart of the profession of arms. Excessive civilian casualties call the legitimacy of U.S. operations into question and may assist the enemy… Effective response will enhance the Army units’ legitimacy and build trust among the civilian population and other actors.”

Without the requirement for all agencies to fully or publicly reckon with civilian casualties, the government may have less incentive to do so internally, depriving policy makers and military leaders of the ability to assess the net effects of civilian casualties on strategic and policy objectives.

Civilian Casualty Counting and Civilian Harm Mitigation

A related concern that Dunlap articulates is that crude civilian casualty numbers don’t take into account the nuance required in making a proportionality assessment in war. Dunlap states:

“Without considerably more information, a percentage comparison of a combatant body count with a civilian one doesn’t even tell you if operators complied with the law. The proportionality rule in the law of war is something to be applied prospectively, not in hindsight. It speaks in terms of permitting strikes on otherwise lawful targets by a ‘reasonable’ commander so long as the incidental civilian losses are not ‘excessive in relation to the concrete and direct military advantage anticipated.’”

To a certain extent, Dunlap is right: Overall civilian casualty figures do not give clear answers to the question of whether or not a specific attack was proportionate under the law of war. At the same time, post-incident strike-specific information on the number of civilians actually harmed may be relevant, if not definitive, for a standalone assessment to help verify what incidental harm was “reasonably foreseeable” at the time.

And civilian casualty figures do have broader relevance for other legal, policy and operational questions related to the prevention of civilian harm. Dunlap fails to discuss how civilian casualty tracking may inform the U.S. and other militaries’ obligations to take “constant care” to “spare the civilian population, civilians, and civilian objects,” and to “take all feasible steps to avoid, and in any event minimize, incidental loss of civilian life, injury to civilians and damage to civilian objects.” As Ryan Goodman and Larry Lewis have noted, “better estimating the number of civilian casualties post-strike should improve pre-strike decisions,” a conclusion that separately the U.N. Secretary-General has made.

By systematically tracking and assessing numbers of civilians harmed in operations, militaries can analyse this data to improve targeting and collateral damage estimation processes in the future so that civilian harm is better mitigated. Mitigation is not possible unless we have a better understanding of how, why and where civilians actually die. Aggregate civilian casualty data can tell you where to look: If there is a spike in civilian casualties at a particular moment in time, or a particular place (or particularly low civilian casualties, which may suggest mitigation measures are working better), then ask: Why?

A key to this process is that public release of such information facilitates an important dialogue with the military to help them analyse and improve their practices. Sahr MuhammedAlly explained in the context of Afghanistan (in findings that one of the authors can validate from his past experience tracking civilian casualties for the United Nations Assistance Mission in Afghanistan):

“The dialogue [between the International Security Assistance Force (ISAF) in Afghanistan] and external and independent organizations was critical for ISAF leadership to re-examine its own incident reporting and engage on recommendations to reduce civilian harm. The tracking data and analysis was used to formulate recommendations to ISAF leadership and to influence recommendations for pre-deployment training on civilian casualty mitigation for the troop-contributing countries.”

A 2013 report by Larry Lewis—produced for the Joint and Coalition Operational Analysis Division, which is responsible for lessons learned in joint operations—identified the importance of civilian casualty tracking and analysis as one of the key lessons from the experience in Afghanistan, explaining that:

“Tracking of CIVCAS was a key enabler for senior leaders in understanding the root causes of CIVCAS, reducing casualties, and mitigating the effects of CIVCAS incidents in Afghanistan. Reliable numbers for CIVCAS in Afghanistan were only available starting in 2009—the eighth year of the conflict—and getting these reliable numbers required the establishment of reporting procedures, policies, and a staff element dedicated to this function. Even so, there were growing pains in attaining consistent reporting and tracking.”

Public scrutiny of the government record can assist with ascertaining the most accurate possible accounting for civilian casualties, which in turn assists the government in its ability to prevent future casualties. Publishing official estimates of civilian casualties—especially with other information such as more detailed descriptions of the methodology employed and information on specific strikes—can also help surface discrepancies between the methodology employed by the military and that of independent organizations and lead to improvements on both sides.

One need only look at Operation Inherent Resolve (OIR), in which the coalition’s civilian harm tracking cell often relied heavily on information provided by Airwars, which reviewed and collated external reports of harm from local communities, media and regional monitors and cross-referenced dates and geo-locational details of claims with the coalition’s own strike data.  In an examination of 191 incidents deemed credible by OIR, 58 percent of civilian casualties that had been confirmed by the coalition were provided by Airwars or other outside organizations. As a former senior International Security Assistance Force officer told the Center for Civilians in Conflict (with whom one of the authors works) in a 2014 report, “We needed third-party validation [of civilian casualty data]… Military tracking and non-military tracking are both important.”

If anything, the fact that the U.S. government retained section four of the Executive Order, which calls for periodic interagency consultation on trends and opportunities for improvement, suggests that the government recognizes the value in tracking and analysing harm. This process is undoubtedly made more effective through the release of more information to organizations like Airwars.

Why Trump’s Rescission of the Civilian Casualty Reporting Requirement Still Matters—Despite Other Reporting Requirements

Almost as if he were arguing in the alternative, Dunlap continues to defend the rescission of the EO provision requiring civilian casualty reporting on the basis of its redundancy with an otherwise mostly thorough set of reporting requirements, by selectively quoting expert Dr. Larry Lewis (who, lost in Dunlap’s analysis, also used his essay to discuss the gaps created by the change, and why they matter). We do not address this issue in detail here except to underscore what Lewis himself has explained:

“Section 1057 [of the National Defense Authorization Act which requires reporting by the Defense Department on civilian casualties] 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.”

The US government as a whole cannot claim a record of transparency in its reporting for civilian casualties if it maintains a policy of withholding information about those that occur at the hands of one agency versus any other.

More not less transparency: Civilian Casualty Figures are Not in Isolation (and Could be Better Supplemented)

Dunlap is particularly concerned that the release of civilian casualty figures “in isolation” does “not seem to be working” in “educating the citizenry about military operations,” and in a subsequent post he has since emphasized this point. Yet the public do not get civilian casualty figures “in isolation.” The U.S. government also releases a range of other information put out about what the U.S. military is doing in particular theatres, including on the progress of campaigns, its views on the successes and challenges of the mission, and its justification for the use of military force: press releases and briefings, posture briefings to Congress; and release of policy documents. Notably, when releasing civilian casualty estimates earlier this year, the Department of Defense itself opted to give additional information “to help give context, such as information regarding the objectives, scale, and effects of these operations.” Removing official civilian casualty counts will not counter “misunderstandings” as Dunlap implies, but only further obscure the full picture of what is going on.

“Body counts alone do not provide other vital information the public would need to make an appropriate assessment,” writes Dunlap, before concluding that it is important that we get “complete, decision-quality information to the public for their assessment of America’s efforts to battle the evil of international terrorism.” We agree—years of endless and spiralling wars and counterterrorism operations since 9/11 demonstrate the need for a better and more comprehensive way to assess the success or otherwise of these engagements. As Hartig points out, the solution isn’t less transparency, but more. The Pentagon is always welcome to provide context and information in its justifications, assuming that the public is rational and smart enough to make their own judgements and to arrive at its own conclusions. And we, and many others, have consistently argued that there is a need for greater transparency: The release of more detailed civilian casualty data, information on specific strikes, the specific criteria used to evaluate external reports of civilian casualties, and the results of investigations into allegations of civilian casualties data, for example, are all important improvements that can be made. 

About the Author(s)

Daniel R. Mahanty

Director of the U.S. program at the Center for Civilians in Conflict (CIVIC). He previously served 16 years at the U.S. Department of State. Follow him on Twitter (@danmahanty).

Alex Moorehead

Lecturer-in-Law at Columbia Law School, Director of the Counterterrorism, Armed Conflict and Human Rights Project at Columbia Law School's Human Rights Institute. Follow him on Twitter (@apmoorehead).