What you need to know about U.S. airstrikes and civilian harm in Afghanistan

Does the media give you what you need to know about critical national security issues?  If you think it does, you may be disappointed in a story (“As U.S. air war in Afghanistan surged, investigations into civilian harm plunged”) that appeared in the Washington Post (online) last Friday. The entire tenor of the article about investigations into civilian harm wrongly suggests that the U.S. military isn’t sufficiently concerned about civilian deaths, and that somehow airstrikes have been counter-productive.  Let’s unpack this, and consider some vital information the Post inexplicably omits from its story.

The Post says that since 2018 there has been a “surge” in the “air war” and makes this charge:

In the same period, allegations of civilian casualties reviewed by the Pentagon doubled. But the number of in-depth investigations into those allegations dropped by half. Hundreds of charges of Afghan civilian deaths and injuries as a result of airstrikes received only an initial assessment.

The article also says:

When asked about the drop in investigations, a U.S. military spokesman said the assertion that Resolute Support conducted fewer investigations is “inaccurate.” “Resolute Support investigates every claim of civilian casualties of which we are aware, either through direct reporting, field reporting or reported on social media,” he said.

However, the Post dismisses those efforts by indicating too few “in depth” investigations take place, and that the team the reporter evidently thinks are the only assets involved in investigating incidents is too small.  The story quotes a critic as insisting that “the civilian toll in Afghanistan is surely greater as a result.”

In my view, what really matters is not whether or not the investigations the U.S. is conducting please a Washington Post reporter or the critics she elects to cite, but rather the degree to which the U.S. efforts to limit civilian harm in the midst of a bitter conflict with a ruthless enemy are ultimately effective.  I find the data shows that they are, but you decide for yourself.

The law

Before we look at the data, let’s clarify what the international law of armed conflict (LOAC) requires.  The Post references civilian casualties as “missteps” but that’s misleading as the LOAC accepts that civilian casualties are inevitable, and permits the conduct of attacks even when it is known civilians will be harmed.

The reason for this is obvious: if a belligerent was shielded from attack if even one civilian would harmed, that would give an even greater incentive for a depraved belligerent to use human shields to create a ‘legal’ fortress from which to wreak malevolence.

In fact, so long as feasible precautions have been taken, the LOAC prohibits attacks where civilian harm is foreseen only when such harm is expected to be “excessive in relation to the concrete and direct military advantage anticipated.”  Military forces may choose, as they often do, to employ more restrictive rules of engagement for political, strategic, or other reasons, but the law does not require doing so.

Yes, there has been a surge in the “air war”

Yes, there has been a “surge” in the “air war” at least in terms of “weapons releases” as the graphic (left) that the Post provides illustrates.

Unfortunately, the Post article doesn’t relate that America’s use of force in Afghanistan is aimed at reducing civilian deaths by eliminating the main cause of them: enemy forces.

That’s right.  Enemy forces are causing the majority of civilian deaths.  For years, the UN Assistance Mission in Afghanistan (UNAMA) has documented that the overwhelming number of civilian casualties typically have been caused by “anti-government” forces, not U.S. or coalition forces (see its reports found here)

Civilian harm has actually declined

The most recent UNAMA report (“Afghanistan: Protection of Civilians In Armed Conflict Midyear Report: 1 January—30 June 2020”) tells you something the Post overlooks: “[t]he overall number of civilian casualties represents a 13 per cent decrease as compared to the same time period last year, and the lowest figure since 2012.” (Emphasis added).  Here’s a graphic from that report (red markings are my adds):As you can see, the number of civilians killed or injured since 2017 has steadily declined as the air war “surged.”

Would you get the facts that civilian casualties are down from the Post’s report?  You decide.  So yes, the U.S. may be dropping more bombs, but my view is simply this: when you kill the fighters doing most of the killing of civilians, you are going to save civilian lives.

Also, it is important to understand that airstrikes (U.S., coalition, and Afghan) cause only a fraction of the civilian deaths in Afghanistan, something else you wouldn’t get from the Post article.  See for yourself in the graphic at the right from the recent UN report. (Red circle is my add.)

If you don’t trust the U.S. government or the UN, you may want to consider the reports from the UK-based Bureau of Investigative Journalism.  I think even a cursory review of its website shows it is no friend of U.S airstrikes in Afghanistan or elsewhere.  Yet take a look at what they show as of this writing (Sept 9):I added the boxes in red so as to facilitate an analysis of the maximum effects of airstrikes.  Do your own calculations, but mine show over 91% of the persons killed in airstrikes were not civilians, and if you divide the number of airstrikes (14083) by the maximum number of civilians reported killed (878) I believe you’ll find only one civilian death for every 16 airstrikes.

Again, decide for yourself but in my experience this is a phenomenal level of accuracy given the military situation in Afghanistan. It reflects a military force making an extraordinary effort to avoid civilian harm in extremely difficult circumstances.  Indeed, U.S. and coalition forces are facing a brutal enemy who not only doesn’t wear uniforms, but also hides himself among civilians and uses human shields.

How reliable are ‘eye-witness’ reports?

In decrying the adequacy of the U.S.’s investigations, the Post celebrates eye-witness testimony and indicates it is insufficiently relied upon in coalition’s inquiries, but does so without alerting readers to the growing body of scientific evidence questioning its reliability.

As just one example, in a 2018 article for the Association of Psychological Science (“Myth: Eyewitness Testimony is the Best Kind of Evidence”) Stephen L. Chew of Stanford University says that “[e]yewitness testimony is more fallible than many people assume.”  He points out:

According to the Innocence Project, 358 people who had been convicted and sentenced to death since 1989 have been exonerated through DNA evidence.  Of these, 71% had been convicted through eyewitness misidentification and had served an average of 14 years in prison before exoneration.

Moreover, it is common-sense that the “witnesses” to airstrikes are often in areas where a savage enemy is operating, and that makes even genuine witnesses extraordinarily vulnerable to being forced to slant or even fabricate testimony.  Add the reality that portraying themselves as airstrike victims might lead to financial compensation–no small matter in a country considered among the poorest in the world – and you have a formula for information that could very well be untrustworthy.

Since the enemy causes far more civilian harm than the U.S., do these tragic losses really have the strategic effect the Washington Post and others seem to think?  

The Post asserts that “[c]ivilian deaths at the hands of U.S. forces have been a powerful rallying cry for anti-American and anti-Afghan government sentiment during the two decades of war.”  Maybe so, but what the Post doesn’t tell you is that civilian deaths “at the hands of U.S. forces” is only a small percentage of the civilian deaths, and are far overshadowed by those caused by the Taliban and other anti-government forces.  The chart at the right from the most recent UN report is typical (at the right; the blue and red outlining are my adds) .

Given the far greater responsibility of anti-government forces for civilian deaths, we should ask ourselves why aren’t these loses more of a “powerful rallying cry” against the Taliban and other terrorists than they are?   Are the Afghans being misinformed or misled by faulty data or enemy disinformation?

Do Western-thinking people really understand how Afghans deal with civilian casualties in an armed conflict?   Or are such outsiders prone to “mirror image” their own views onto an ancient culture that may see things quite differently for a variety of reasons?

If civilian casualties, per se, really had the “powerful” effect the Post (and, frankly, others) seem to think, wouldn’t the Taliban, Al-Qaeda, ISIS, and other “anti-government” elements been driven from Afghanistan long ago by the tens of millions of Afghans who are not their followers?  After all, as noted above, the UN has documented that the overwhelming number of civilian casualties typically have been caused by these “anti-government” forces, not U.S. or coalition militaries.

This isn’t at all to suggest that somehow Afghan people don’t grieve deeply for their losses as they certainly do; rather, it is simply to say that civilian casualties don’t necessarily have the effect on the counter-insurgency effort in that particular society in the way that Western-thinking reporters and others may suppose. In short, it’s a much more complicated issue than the Post portrays it.

Americans may not be able to control how Afghans choose to react to the relatively small percentage of civilian casualties caused by U.S. airstrikes, but what they can continue to do is to eliminate those who threaten the country’s future and America’s security.

Concluding observations

If someone read the Post article in isolation, he or she might think the post-strike investigatory process was the only or even the main way that civilian harm is mitigated in the U.S.’s conduct of airstrikes in Afghanistan and elsewhere.

In truth, reducing the risk to civilians involves a myriad of factors including specially-developed munitions (e.g., here), extensive intelligence-gathering processes, deployed military lawyers, disciplined rules of engagement (which are typically more restrictive than what the law requires), and a complex targeting process that carefully assesses the danger to civilians, and much more.   (If you want to really understand how key parts of the U.S. military’s airstrike process actually works, check out Hunting the Caliphate).

Finally, it is worth noting that refraining from conducting airstrikes does not, ipso facto, mean civilian lives are saved.  Recall 2009 when Gen Stanley McChrystal took command in in Afghanistan and restricted airstrikes  in the hopes of limiting civilian casualties.  The result?  When he was relieved a year later, deaths from airstrikes were down, but overall civilian deaths were up a startling 31% at that point for 2010.  Civilians are not protected when airstrikes that might have killed those who ultimately cause civilian harm don’t take place.

Again, the grim reality of war is that sometimes the best and perhaps only way to protect the lives civilians is to kill the most profligate killers of civilians, and airpower has done that effectively.

Foregoing otherwise lawful airstrikes in such circumstances may allow U.S. forces to avoid censure for any civilian casualties that might have occurred from the strike, but it also creates what I call the moral hazard of inaction in war as it shifts the risk to civilians on the ground who are now in peril of harm at the hands of the terrorists who escape elimination.

No moral person wants civilians harmed in armed conflict, and a free, informed, and unbiased press can provide the public and policymakers useful critiques of U.S. military operations when harm does occur.

But such critiques are not helpful when they omit vital data needed to make reasoned judgments of a difficult issue where so many lives are at stake.

Still, remember what we like to say on Lawfire^®: gather the facts, examine the law, evaluate the arguments – and then decide for yourself!