The Pentagon claims the data it releases about the progress of its wars are accurate, but it turns out they’re cooking the books.
- By Micah ZenkoMicah Zenko (@MicahZenko) is a senior fellow with the Center for Preventive Action at the Council on Foreign Relations and is the author of Red Team: How to Succeed by Thinking Like the Enemy.
On March 11, 2016, Col. Steve Warren, spokesman for Operation Inherent Resolve (OIR), was asked about the daily news releases published by the U.S. military to describe the progress of the war against the Islamic State. The statements list the number of airstrikes conducted by the coalition during the previous 24 hours, their locations in Iraq or Syria, and an estimate of the Islamic State targets damaged or destroyed. One “strike” is defined as “one or more kinetic events that occur in roughly the same geographic location to produce a single, sometimes cumulative effect.” They omit artillery fire used for counter-fire, or for fire support on behalf of maneuvering ground forces.
During the news conference 11 months ago, reporter Lolita Baldor asked Warren directly: “The daily strike press releases that we get. Can you give us a sense of whether or not all airstrikes are included in those press releases?” Warren replied:
We do everything we can to report every single strike on those releases. We’ve missed one or two for admin errors. … But our standard is that, if a bomb falls in Iraq or in Syria, it makes it to that press release … whether it’s a high-value individual, or a bunker, or a tactical unit or a chemical weapons facility.
Warren often repeated the claim that official releases included “every” strike. On Oct. 28, 2015, he stated: “every single time we do an air strike, it gets listed on our daily air strike release.” On Dec. 22, he proclaimed: “We list the number of airstrikes conducted every day on our daily strike releases.” On May 27, 2016, during his final news conference as OIR spokesman, he declared: “We release the actual number of strikes that we conducted every day. … I would ask you to go check out that website. It’s a terrific website. … You will see the exact number of strikes conducted that day.”
Since the counter-Islamic State campaign began in August 2014, U.S. civilian and military officials have emphasized that being honest about the conduct of the war would be a priority, in order to ensure public support in the United States and in the greater Middle East. Indeed, compared with the paucity of information provided by Russia for its operations in Syria, or from the Saudi-led coalition for its bombing of Yemen, the U.S.-led coalition has offered extraordinarily detailed, up-to-date data. Moreover, in my experience at least, the OIR public affairs office has been responsive to requests for additional or clarifying information.
U.S. officials’ vows that the public strike data was accurate and comprehensive, and the willingness of the OIR press office to assist researchers and journalists, makes Andrew deGrandpre and Shawn Snow’s bombshell report in the Military Times on Sunday so startling. In a piece appropriately titled “The U.S. Military’s Stats on Deadly Airstrikes Are Wrong. Thousands Have Gone Unreported,” the reporters uncovered a glaring shortcoming in the military data for OIR, as well as for airstrikes in Afghanistan.
There are three consequential findings from the Military Times reporting. First, U.S. Army strikes conducted by armed drones and attack helicopters are excluded from the overall totals from both OIR and Afghanistan. (This may be because the Army “doesn’t view airstrikes as independent action,” as brilliant airpower thinker Air Force Brig. Gen. Clint Hinote tweeted, though for a military that considers itself an integrated “joint force,” this does not make sense to me.) The Army does not independently publish its own data; DeGrandpre and Snow managed to acquire it from Afghanistan only because the military command in that country gave it to them.
What they discovered is that the publicly released data for strike sorties in Afghanistan last year (615) was 43 percent lower than the actual number of strikes (1,071). Second, none of the military officials interviewed appeared to know how long this undercount has been occurring, or what Department of Defense policy was for all armed services to report their airstrikes. Third, the military data published for airstrikes in other theaters, including Syria, Iraq, Yemen, and Somalia, may be artificially low for similar reasons — assuming the Army is conducting its own strikes in those theaters. For example, U.S. Africa Command reported 495 strikes on behalf of the Government of National Accord in Libya last year, a number that may be an undercount.
Cumulatively, this calls into account the integrity of the information the Pentagon has been releasing about American airpower, the most heavily utilized military element of its current wars. Sunday’s report quotes an Air Force official who claimed the coalition “as a whole, which is all 20 nations and the U.S. branches,” was fully represented in the OIR public releases. DeGrandpre and Snow observe, sadly, “It’s unclear whether this statement was intentionally misleading, or simply indicative of widespread internal ignorance, confusion or indifference about what’s contained in this data.”
Such internal uncertainty means that other data the U.S. military has provided to Congress and the American public cannot be taken at face value. The data include the number of troops deployed, the number of supporting military contractors, U.S. troop casualties and fatalities, missions being conducted, rebel groups being armed and equipped, financial costs, civilian casualties, and overall progress in the campaign to destroy the Islamic State. Given the published undercount of something as discrete and measurable as a clearly defined “strike,” military officials may — even unknowingly and with good intentions — be misrepresenting additional military activities.
As someone who has relied on Pentagon data — albeit with skepticism — to evaluate U.S. military operations, Sunday’s findings cause me to reconsider some of my earlier analysis in light of this new information. For example:
- Comparing the number of civilian fatalities from (overwhelmingly) drone strikes in Yemen, Somalia, and Pakistan and the (overwhelmingly) manned airstrikes in Iraq and Syria. (Taking new information into account, manned strikes are even more precise compared with drone strikes than I had thought.)
- Determining the number of civilian fatalities per weapon released in Afghanistan. (U.S. airstrikes in Afghanistan can now be seen as slightly more precise.)
- Comparing the Islamic State air war with recent U.S.-led bombing campaigns. (The Islamic State air war should now be considered more robust than was previously believed.)
- Assessing what percentage of overall coalition strikes are being conducted by the United States. (Apparently, the United States is shouldering more of the burden than was thought.)
- Totaling the bombs dropped by the United States in 2015 and 2016. (We know now that there were clearly more bombs dropped in both years.)
The undercounting of U.S. airstrikes in official public releases is a serious matter. It undermines the credibility of military messaging, casts doubt on other military data, and inhibits any ability of the Pentagon to establish norms for transparency that other militaries might follow. If Army attacks will not be included in the official data, then the military should state explicitly that it is not comprehensive. The Pentagon can count and promote its data however it sees fit, provided they are clear and transparent.
Meanwhile, the Senate and House Armed Service Committees should look into this matter and hold public hearings with relevant military officials to learn more about the airstrike discrepancy. The Pentagon’s Office of Inspector General should try to determine the root cause of the strike undercount, and see if any officials and staffers have been willfully misleading Congress and the public. Finally, Secretary of Defense James Mattis should direct the chairman of the Joint Chiefs of Staff and his senior aides to establish and publicize an official policy for what kinetic activity should be considered an airstrike. Without such clarity, it is difficult to know whether to believe the military at all.
Photo credit: ALEX WONG/Getty Images