Earlier this month, the New York Times reported that U.S. airstrikes in Baghuz, Syria on March 18, 2019 killed scores of civilians including many, if not mostly, women and children. The Times noted that U.S. Central Command for the first time publicly acknowledged the strikes (including 80 casualties) after the Times reporters sent the command their findings in the week before the NYT published its article.

What follows is the full statement by Centcom spokesman, Capt. Bill Urban. I have identified and annotated the 20 most significant elements of the statement.

Readers may also be interested in an earlier article at Just Security: Brianna Rosen, Luke Hartig, Tess Bridgeman & Ryan Goodman, Questions on the Baghuz Strikes (Nov. 15, 2021)


Statement by Capt. Bill Urban, CENTCOM spokesman

In mid to late March of 2019, U.S., coalition and Syrian Democratic Forces had isolated the final remnants of the ISIS proclaimed territorial caliphate in Baghouz, Syria. The ISIS pocket included thousands of fighters and family members including women and children. The remaining fighters including some women and child combatants[1] along with many ISIS family members, including some who were likely held against their will, decided to make a determined stand in an area that included buildings, tunnels and cliffs. Multiple entreaties to ISIS to allow family members to depart the area were rebuffed, and thousands of family members remained in the area of the fighting.

[1] In favor of Centcom’s position here: The International Committee of the Red Cross recognizes that “children below the lawful recruitment age may lose protection against direct attack” when they directly participate in hostilities. Also, academics and other independent experts have documented ISIS’s use of women and child soldiers as part of the fighting force.

However, when Centcom states that these were “women and child combatants,” a key question is: What are the criteria that leads DOD to determine that a woman or a child is a combatant? What were the women and children doing at the time that DoD classified them as combatants?

In contrast, the DOD Inspector General analyst who reviewed all the video footage of that day saw a “large group of women and children huddled in a place where there doesn’t seem to be a lot of action.” If some women and children in a group are armed but clearly huddled away from the action, as if avoiding the fight, would DOD consider them combatants?

On March 18, 2019, five days before the end of the battle, as many as 200 hundred ISIS fighters launched a determined counterattack on SDF and coalition forces that started around 4 a.m. local time, lasted numerous hours and resulted in more than 30 SDF casualties. After more than 6 hours of fighting and dozens of strikes, numerous U.S. manned and unmanned strike platforms had departed the area due to having expended all of their ordinance. ISIS fighters had inflicted high casualties on the SDF through small arms, rocket propelled grenades, indirect fire, improvised explosive devices and suicide bombings.

After 10 a.m. local time, an SDF position under heavy fire and in danger of being overrun called for defensive airstrikes on ISIS fighter positions. The SDF fighters reported the area clear of civilians, and the nearest U.S. special operations forces on the ground also reported that they could observe no civilians in the area.[2] The only remaining U.S. unmanned aerial vehicle (UAV) in the area at the time of the strike was only capable of providing standard definition full motion video (FMV) as high definition-capable UAVs that were on station earlier had departed following the expenditure of their ordinance. The remaining U.S. UAV had also expended all of its hellfire missiles and was unarmed at the time of the call. The U.S. UAV on station was unable to discern[3] any civilians in the area at the time of the SDF request. The only remaining armed strike aircraft on scene at the time of the request for air support was a U.S. F-15 with precision guided 500- and 2,000-pound bombs. With the area reported clear of civilians, and SDF positions under fire, the special operations strike cell authorized two defensive strikes on ISIS positions with two 2,000-pound precision guided munitions and one 500-pound precision guided munition[4] from the F-15. The strikes were determined to have taken out at least 16 active ISIS fighters that were engaged in attacks on SDF positions.

[2] In favor of Centcom’s position here: The New York Times does not adequately include Centcom’s explanation that “SDF forces reported the area clear of civilians, and the nearest U.S. special operations forces on the ground also reported that they could observe no civilians in the area.”

That said, with respect to the latter wording, did the U.S. special forces report they observed no civilians in the specific area of the strike? Or did they report they were not in a position to observe civilians? The Times did say that an independent videographer’s “footage shows that ground troops may not have been able to see the group of civilians.”

Also, is it a common practice for the U.S. military to rely on local partners for these purposes, and how much confidence does DOD place on assessments by the SDF compared to its own pre-strike intelligence? Notably, the recommendations at the end of the Centcom statement (see [17] below) do not include changing how the U.S. military relies either on local partners’ calling in strikes or local partners’ assurances of no civilian casualties. Is that because the U.S. military’s rules already included cautions against such reliance, and those cautions were not followed in the March 18, 2019 strikes?

[3] The key words here: “unable to discern.” Does this mean the UAV video resolution was of such poor quality or involved such poor visibility that it could not identify whether civilians were present (and the special operations task force relied on it anyway in taking the strikes)? Or did the U.S. special forces know women and children were present but assessed them as likely or presumptively combatants?

An important antecedent question: What is the definition of “civilian” used by the special operations task force? And, what is the target engagement criteria the special operations task force used to determine whether individuals at the location were combatants?

Also importantly, this UAV video footage was conspicuously missing, at least from the materials Centcom provided to the DoD inspector general’s office. According to the New York Times: “Central Command was slow to turn over evidence,” according to Eugene Tate, an evaluator who worked on the case for the inspector general’s office, and “Mr. Tate obtained video from several drones flying over Baghuz that day, but could not locate the footage from the task force drone that called in the strike.”

[4] Centcom has subsequently issued a correction stating that the strikes involved three 500-pound munitions and no 2,000-pound munitions. The more recent statement attributes the error to the Public Affairs office: “CENTCOM Public Affairs shop misidentified the munitions used.”

Several hours after[5] these two strikes, a coalition UAV operator,[6] who was operating a coalition UAV in the area with high-definition full motion video reported to U.S. special operations forces that the UAV had observed possible civilians in the area at the time of the strike. As the U.S. strike cell was unaware of the presence of the coalition UAV,[7] U.S. special operations forces did not have access to their feed at the time of the strike. In response to this report, U.S. special operations forces leaders initiated[8] a civilian casualty credibility assessment which determined[9] that the report of civilian casualties was credible and warranted a 15-6 investigation.

[5] [6] & [7] The language used to describe the “coalition UAV operator” and “coalition UAV” and the elapsed “several hours” invites scrutiny. Among other things, this description gives the appearance of a largely separated entity from U.S. forces. The New York Times refers to the high-resolution UAV as a “U.S. military drone.” More fundamentally, the Times reports:

“At the U.S. military’s busy Combined Air Operations Center at Al Udeid Air Base in Qatar, uniformed personnel watching the live drone footage looked on in stunned disbelief.”

In the minutes after the strike, an alarmed Air Force intelligence officer in the operations center called over an Air Force lawyer in charge of determining the legality of strikes. The lawyer ordered the F-15E squadron and the drone crew to preserve all video and other evidence, according to documents obtained by The Times. He went upstairs and reported the strike to his chain of command, saying it was a possible violation of the law of armed conflict — a war crime.” (emphasis added)

[8] On what date was the civilian casualty assessment initiated?

[9] On what date was the civilian casualty credibility assessment concluded?

The civilian casualty credibility assessment and the 15-6 determined[10] through the study of all available evidence including coalition high-definition video, that several locations unobserved by the U.S. UAV[17] with men, women, and children likely received lethal effects from the strikes. The investigations concluded that at least 4 civilians were killed and 8 were wounded from these strikes.[11] The investigations were unable to conclusively[12][13] characterize the status of more than 60 other casualties that resulted from these strikes.[13] The reason for this uncertainty is that multiple[14] armed women and at least one armed child were observed in the video, and the exact mixture of armed and unarmed personnel could not be conclusively determined. Likely, a majority of those killed were also combatants[15] at the time of the strike, however, it is also highly likely that there were additional civilians killed by these two strikes.

[10] On what date was the 15-6 investigation concluded?

In its initial Nov. 15, 2021 article, the Times reported:

“Two and a half years later, on the military’s website for its campaign against the Islamic State, known as Operation Inherent Resolve, the military still publicly lists the case as ‘open.’”

In a follow-on article on Nov. 17, 2021, the New York Times reported:

A task force investigated the strike and acknowledged that four civilians were killed, but it also concluded that there had been no wrongdoing in the unit. In October 2019, the task force sent its findings to the military headquarters in Baghdad, as well as to the Central Command headquarters in Tampa, Fla.

But the command in Baghdad failed to review and close the inquiry, and Central Command did not follow up and remind the Baghdad command to do so, Capt. Bill Urban, the Central Command spokesman, said ….

As a result, senior military officials in Iraq and Florida never reviewed the strike, and the investigation technically remained open until the Times investigation.

“Should we have followed up? Yes,” Captain Urban said in a telephone interview, blaming “an administrative oversight.” (emphasis added)

This is also a striking admission by Centcom. Are there other cases in which the command in Baghdad similarly engaged in a failure to review and close an inquiry into credible civilian casualty reports?

[11] On what date exactly did the DOD assess 4 civilians were killed and 8 were injured in the strikes?

This date may be especially important because DOD failed to report the March 18, 2019 strike in its annual civilian casualty reports to Congress submitted around May 2020 and June 2021. Notably, the New York Time reports that in May 2019, “the task force finished up a civilian casualty report on the strike that month and determined that four civilians were killed.”

Did Centcom leave the report “open” for over two years with the result (whether intentional or unintentional) that the civilian casualties were never reported to Congress?

[12] What does “conclusively” mean? Did DOD have a high or even moderate confidence assessment of the status of the 60 people killed that falls short of conclusive? Or is this a radical uncertainty?

[13] The failure to assess individuals as either civilian or combatant is unusual. When the Office of Director of National Intelligence, for instance, provided post-strike civilian casualty assessments in 2016, it stated that when it cannot be determined if an individual was a combatant, the person will be placed in the civilian category: “The assessed range of non-combatant deaths includes deaths for which there is an insufficient basis for assessing that the deceased is a combatant.” That said, the ODNI formulation appears to be inconsistent with the DOD standard. DOD requires proof by a preponderance of the evidence (more likely than not) that an individual was a civilian. It would appear that the DOD – if it applies its own standard to the March 18, 2019 strikes – may need to report the 60 cases in which it has uncertainty as combatants rather than civilians. This exercise may point to more fundamental problems with DOD’s post-strike standards of proof for assessing civilian casualties.

[14] The use of the word “multiple” can be deceptive. In the New York Times’ “The Daily” podcast, Tate, the analyst at the DOD inspector general’s office who reviewed all the available footage, said:

Dave Philipps (NYT reporter): “And what he sees is this large group of women and children huddled in a place where there doesn’t seem to be a lot of action.”

Eugene Tate: “Yes, I see people with weapons. Not a lot. I see, like, two people with weapons.” (emphasis added)

[15] This is an astonishing statement. Centcom needs to show its math here. On what evidence and calculation does Centcom assess that a majority of those killed were likely combatants?

The problem may lie in how DOD/Centcom defines and infers someone is a combatant. Specifically, pre-strike, what are the permissible target engagement criteria for determining an individual is a combatant? Post-strike, what are the factual inferences DOD/Centcom uses to determine whether an individual satisfies its definition of combatant or civilian?

The 15-6 investigation concluded that these two strikes were legitimate self-defense strikes in support of SDF forces under fire, that they were proportional due to the unavailability of smaller ordinance at the time of the request,[16] and that appropriate steps were taken to rule out the presence of civilians at the time of the strike. To prevent unintended casualties in the future, the investigation recommended requiring high-definition video for similar strikes in the future, and the requirement for the strike cell to coordinate with any coalition surveillance assets in the area at the time of the strike, and those recommendations were implemented.[17] Finally, in accordance with the findings, the investigating officer[18] determined that no disciplinary actions were warranted.

[16] This formulation does not match any recognizable definition of “proportionality” under the law of targeting or the rules on self-defense, including publicly available standing rules of engagement. It is such a misstatement of the concept of “proportionality” that it raises the question whether the statement was properly legally vetted by Centcom legal staff.

[17] This statement is tantamount to an admission of wrongdoing on two-levels. First, it admits these procedures were not adopted in the many years of targeting operations before March 2019. Second, it acknowledges that it is feasible to make these procedures a requirement even for self-defense strikes.

What’s more, the U.S. government recognizes that the laws of war require as a minimum that “feasible precautions must be taken in conducting an attack to reduce the risk of harm to civilians.”

In sum, by Centcom’s own admission, the procedures were feasible precautions, and they were not taken in conducting the March 18, 2021 strikes.

Also, if these procedures were feasible why were they not adopted for U.S. military operations beforehand?

[18] What was the rank and position of the investigating officer and what was their relationship, if any, to the special operations task force that conducted the strikes?

We abhor the loss of innocent life and take all possible measures to prevent them. In this case, we self-reported and investigated the strike according to our own evidence and take full responsibility for the unintended loss of life. It is important to understand that ISIS decided to put their own families at risk when all avenues of escape were afforded to them. It is also important to note that some women and children, whether through indoctrination or choice, decided[18] to take up arms in this battle and as such could not strictly[19] be classified as civilians. Ultimately, the determined effort of U.S., coalition and SDF forces culminated the final territorial defeat of ISIS on March 23, 2019, but it did not come without a significant cost of U.S. and partner casualties or without the regrettable loss of civilian life.

[19] “Indoctrination and choice” omits that some women and children were forced to take up arms for ISIS. Does Centcom consider “indoctrination” a form of voluntary or involuntary enlistment? If involuntary, does it affect Centcom’s targeting decisions or civilian casualty assessments?

[20] This notion of being “not strictly” classified as civilians is highly unusual. Centcom should explain this terminology and classification.


The full text of the statement, without annotations, is below.


Image: The exterior of U.S. Central Command at MacDill Air Force Base in Tampa, Florida. The facility is the headquarters for military operations in the Middle East and central Asia.