Power Wars Blog by Charlie Savage

[Updated with video, verbatim Q&A] Brian Egan’s ASIL speech and questions about rules for drone strikes in tribal Pakistan and for assessing collateral damage

Yesterday, Brian Egan, the Obama administration’s recently confirmed State Department legal adviser (and former National Security Council legal adviser), gave a speech at the annual American Society of International Law conference about “International Law, Legal Diplomacy, and the Counter-ISIL Campaign.” Egan’s speech was an excellent distillation of the Obama administration’s argument for why it thinks everything it has been doing in counterterrorism strikes, against the Islamic State and otherwise, has complied with international law – something that some scholars dispute when it comes to certain novel questions about imminence, sovereignty, and other issues. (Marty Lederman has a good write-up summarizing its key points at Just Security.) I happened to be sitting near the microphone, so asked him two quick questions during Q&A — one seeking clarity about the status of tribal Pakistan for battlefield targeting rules, and one seeking clarity about how the government makes collateral damage assessments. This post explains the issues, and why one of his answers was clear but the other was not. I can’t find video of the speech posted anywhere, but I’ve embedded some Tweets.

Update: Here is the video; our exchange starts just after the 43 minute mark.


In one part of his speech, Egan talked about the May 2013 Presidential Policy Guidance (PPG), which imposed tighter standards on counterterrorism strikes away from hot battlefields, like like requiring near certainty that there will be no civilian casualties – a higher limit than international law requires under the law of armed conflict or self defense. Egan noted that this does not apply in places where there is a zone of active armed conflict:

The phrase “areas of active hostilities” is not a legal term of art—it is a term specific to the PPG. For the purpose of the PPG, the determination that a region is an “area of active hostilities” takes into account, among other things, the scope and intensity of the fighting. The Administration currently considers Afghanistan, Iraq, and Syria to be “areas of active hostilities,” which means that the PPG does not apply to operations in those States.

It has long been my understanding that the PPG did not apply to tribal Pakistan – the FATA region along the border with Afghanistan, which the Pakistani government does not really control. That’s important because the overwhelming majority of CIA drone strikes have taken place there, including its campaign, since 2008, of “signature” strikes aimed at groups of men who look like militants but whose identities are not known, raising the risk of civilian deaths. The purpose for that campaign is not just to take out high level Al Qaeda leaders, but also run-of-the-mill insurgents who stage attacks on U.S. troops across the border in Afghanistan.

So I asked Egan whether he could clarify if tribal Pakistan counted as an extension of Afghanistan, where the PPG does not apply, or as a separate place that is not an area of active hostilities, where the PPG does apply?

Egan clearly indicated that the border region, on the Pakistani side, was still part of the battlefield, so PPG limits did not apply there.

Update: Now that the video is available, here is the exact exchange:

Q) One, on your clarification of the ‘zone of active hostilities’ and where the PPG does and does not apply, does tribal Pakistan count as an extension of Afghanistan for the purpose of the PPG not applying, or does it count as a ‘not hot battlefield’ where the PPG does apply?

A) With respect to the first, the question of ‘area of active hostilities,’ when I mentioned Afghanistan, I think sometimes others have referred to the Afghanistan-Pakistan border region as being part of what we talk about with respect to ‘Afghanistan.’


In another part of his speech, Egan talked in some detail about the various factors the government uses to decide whether a potential target would be a lawful object of attack, as opposed to a civilian who is not taking part in hostilities. Along the way, he made the comment I’ve bolded here so you can see it in context:

In many cases we are dealing with an enemy who does not wear uniforms or otherwise seek to distinguish itself from the civilian population. In these circumstances, we look to all available real-time and historical information to determine whether a potential target would be a lawful object of attack. To emphasize a point that we have made previously, it is not the case that all adult males in the vicinity of a target are deemed combatants. Among other things, the United States may consider certain operational activities, characteristics, and identifiers when determining whether an individual is taking a direct part in hostilities or whether the individual may formally or functionally be considered a member of an organized armed group with which we are engaged in an armed conflict. For example, with respect to membership in an organized armed group, we may examine the extent to which the individual performs functions for the benefit of the group that are analogous to those traditionally performed by members of State militaries that are liable to attack; is carrying out or giving orders to others within the group to perform such functions; or has undertaken certain acts that reliably indicate meaningful integration into the group.

This is a reference to controversy over an assertion first reported by my colleagues Scott Shane and Jo Becker in a deeply reported 2012 article they wrote in the New York Times about Obama and drone strikes:

…Mr. Obama embraced a disputed method for counting civilian casualties that did little to box him in. It in effect counts all military-age males in a strike zone as combatants, according to several administration officials, unless there is explicit intelligence posthumously proving them innocent. …

This counting method may partly explain the official claims of extraordinarily low collateral deaths. … The C.I.A. accounting has so troubled some administration officials outside the agency that they have brought their concerns to the White House. One called it “guilt by association” that has led to “deceptive” estimates of civilian casualties.

“It bothers me when they say there were seven guys, so they must all be militants,” the official said. “They count the corpses and they’re not really sure who they are.”

After that article came out, various administration officials pushed back, saying that this part was simply wrong. But last October, when The Intercept published some leaked documents about JSOC drone strikes (so they involved the military, not the C.I.A.), one of the most important details  seemed to dovetail with it. The document was a presentation about a 13-month campaign in northeastern Afghanistan from 2012 to 2013 called Operation Haymaker. There had been 56 kinetic strikes, killing 35 targets (“JP,” in the slidedeck, short for”jackpots,” lingo for a specifically intended target) and 219 other people. All 219 other people killed because they were in the strike zone were deemed to have been combatants (“EKIA,” meaning enemies killed in action). None of the 219 were deemed civilian casualties/collateral damage or deemed to be status unknown.

While Egan’s speech seemed to frame the issue as being about ex ante decisionmaking — if we can target X, may we also target that other guy Y because he is an adult male nearby? — I had understood the issue to be more about after-the-fact assessments. I pointed that out as well as the leaked drone strike data, and asked Egan if it were not the case that all dead adult males in the vicinity of a target are presumed to have been combatants in the absence of additional information about who they were, then how do collateral damage assessments work instead?

A lot of people would really like to know the answer to that.

Egan reiterated that the government did not simply presume that any adult male near a target in a strike zone to be a combatant, instead taking many factors into play.

But he didn’t specify what those factors are.  We can guess at some obvious ones – for example, did the drone, looking down at the site, see the mystery guy holding a weapon? did he seem to be following the target’s orders? But the core of issue, which remains unknown, is what the default presumption is when there is no information other than that there is an extra dead man down there, identity unknown. So his answer to that one — probably through no fault of his own, since a lot of this stuff is classified and it was a public setting — was not really satisfying.

Update: Here is the verbatim exchange from the video:

Q) And secondly, on the military-aged males thing. Now I haven’t written about that but my colleagues at the Times first wrote about that in 2012, and I immediately began hearing push-back from people in positions not unlike you’ve held in recent years that that wasn’t correct. But then it seemed to have been confirmed in the leaked drone papers published by the Intercept. And my understanding is it’s less about well if someone if someone is a military-aged male near a target then they are themselves a valid target, an ex ante calculation, but it’s more about after the fact, how do we calculate collateral damage? If it’s the case that some military-age males are dead near the strike zone and that is all that is know about them, are they counted as ‘collateral damage’ by virtue of that, or are they counted as ‘enemies killed in action’ by virtue of that? And I would note that the drone papers showed that in 13 month period of JSOC targeting, there were 35 targets killed and 219 bystanders killed as part of those targeting and all 219 were categorized as enemies killed in action.

A) Secondly on the issue of military-age males. It is not the case that a military-age male is considered a combatant solely by virtue of his being a military-age male. And I think that would apply both to assessments prior to a strike and assessments taken, after-action reports after a strike. So there are a variety of factors that are looked at in both cases. But it’s not that one’s military-age male status on its own would suffice.