US Drone Strikes in Pakistan: Can it be Legal to Target Rescuers & Funeralgoers?

Written by

A week ago, the Bureau for Investigative Journalism (BIJ), in conjunction with the Sunday Times (of London) published a report into US drone tactics in Pakistan. The report states that since Barak Obama came into office US drone strikes in Pakistan killed between 282 and 535 civilians. The core of the recent report was that some of these civilians were killed in follow-up strikes which delibaretely targeted those who had gone to help victims of previous strikes or were killed in deliberate strikes on funerals and mourners. In a separate piece, “A Question of Legality“, the BIJ examines whether this US tactis is lawful under international humanitarian law (and international human rights law). In that piece I am quoted as follows:

Professor Dapo Akande, who heads Oxford University’s Institute for Ethics, Law and Armed Conflict, believes that under LOAC the killing of civilian rescuers is problematic: ‘The question is, can rescuing be regarded as taking part in hostilities, to which for me the answer is clearly “No”. That rescuing is not taking part in hostilities.’

The BIJ piece (and my quote) has generated a keen debate on other international law blogs as to legality of the (alleged) US tactic of attacking rescuers and funeral goers. Bobby Chesney, at Lawfare agrees with what I say on direct participation in hostilities but argues that this only matters “if we assume that a person must be directly participating in hostilities in order to be targeted lawfully in that context.”. He explains that if one agrees with the position taken by the ICRC in its Interpretive Guidance on the Notion of Direct Participation in Hostilities Under International Humanitarian Law, then a member of an organized armed group who has a continuous combat function is not a civilian and can be targeted at any time (i.e on the basis of his status alone), subject to questions of proportionality. That person does not need to be taking a direct part in hostilities to have no immunity from attack.

I agree with Bobby Chesney’s assessment. He is right that the first question is whether the people who have been targeted are civilians. It is only if the answer to that question is in the affirmative that one gets to the question of direct participation in hostilities. I did make this point to the BIJ in my interview with them but I think the quote they chose to go with was one which they felt made the point about the illegality of killing civilian rescuers most strongly. I have gone over the transcript of my interview with BIJ and this is what I said on this point:

On the one hand, if we’re talking about people who are known to be militants (and this is the big question, whether they are and how the US knows and all of that) then you can say, well, these people under LOAC can be targeted just because they are militants. That would be fine. On the other hand, it can be argued that the law says that if you’re a civilian and you take a direct part in hostilities, then you can also be targeted. So you’re not a militant,  you’re just a local, but you take a direct part.

So then the question is, can rescuing be regarded as taking part in hostilities, to which to me the answer is clearly ‘No.’ That rescuing is not taking a direct part in hostilities. And so if a person is not a militant, the fact that they are coming to recue and help, that’s not taking a direct part in hostilities.

The first para above was intended to make, in simple terms, the point that members of organized armed groups can be the object of an attack (even in a funeral or a rescue situation) on the basis of that status alone. However, that would only be stage one of the analysis as all that would have been satisfied is the principle of distinction (which requires those conducting attacks to distinguish between civilians and combatants).  If the attack causes civilian casualities or damage to civilian objects, one would then proceed to analyse whether the principle of proportionality is satisfied.

Kevin Jon Heller in a post on Opinio Juris takes a different view and has argued that the principle of distinction does not permit the U.S. to intentionally attack one member of an organized armed group who is attending a funeral along with a number of civilians.  In his view such an attack is a clear violation of the principle which states that the civilian population as such shall not be the object of attack (Art. 51(2) Additional Protocol I, 1977). Kevin notes Article 50(3) of API which states that the  “[t]he presence within the civilian population of individuals who do not come within the definition of civilians does not deprive the population of its civilian character.”

I disagree with Kevin that an attack which has as its object the killing of a member of an organised armed group (lets call him “combatant” for short) fails to respect the principle of distinction because the combatant is in the company of civilians. The attack may well be unlawful because of disproportionate civilian casualties or loss but that is a different point. In the comments to Kevin’s post, as well as in a post by Jens Ohlin at Opinio Juris and on his own LieberCode blog,  there was much discussion of the provisions of the Rome Statute and of what it means to intend to attack civilians for the purposes of the crime of intentionally directing attacks against the civilian populations. I would like to leave the questions of international criminal law to one side and just address the IHL point. Kevin’s argument assumes that any attack on a group of civilians is a breach of prohibition of attacking the civilian population. I don’t think that can be right for two main reasons.

First of all, Article 51(2) which prohibits attacks on the civilian population and individual civilians does not simply prohibits attacks but rather makes it illegal to make civilians and the civilian population ” the object of the attack”. To me this emphasises that what is at issue is the question of who the attacker wishes to affect, who he is aiming his attack at. It seems to me that in the scenario under discussion, where the aim is simply to kill the combatant, it is the combatant that is the object of the attack. The civilians are not. The use of the words “as such” after civilian population in Art. 51(2) AP II also suggests that a violation would require that the attacker wishes to target the civilian population, knowing that they are a civilian population and making them a part of his objectives. Indeed, the ICRC commentary to the provision states that:

“By using the words “directed” and “as such” it emphasizes that the population must never be used as a target or as a tactical objective.”

I am impressed with the extensive discussion of the what intent means in international criminal law and comparative criminal law but I’m not sure this is relevant to the application of the IHL norm which underlies the ICL rule. In short I agree with Jens Ohlin when he says that “the relevant question is how the concept of intent, or the idea of “directing an attack” against civilians, is understood by IHL lawyers, and in particular how the concept was understood by the negotiators of the Additional Protocol and any other relevant treaty or convention.” I also agree with the Jens that it would be strange if the ICL rule was broader than the IHL rule.

Secondly, if one were to adopt the Kevin’s approach it is difficult to see what meaningful role the principle of proportionality would play in the context of attacks against people. If it is ipso facto unlawful (under the principle of distinction) to launch attacks against groups of people, the vast majority of whom are civilian but which also include combatants, then we don’t need the principle of proportionality  when considering attacks against persons. We would only analyse proportionality when considering attacks against objects. However, this is not only contrary to received wisdom but also seems contrary to the structure of API where the proportionality principle is included in Art. 51 (which deals with attacks against civilians and the civilian populations ) and not in Art. 52 (which deals with attacks against civilian objects).

Of course, none of this is to say that attacks by the US or others on rescuers or funeral goers is lawful. They may well be unlawful if the object of the attacks are civilians not taking a direct part in hostilities or would be unlawful even if directed at combatants or civilians taking a direct part but where the attack results in excessive civilian loss or casualties.

Print Friendly, PDF & Email


No tags available

Leave a Comment

Comments for this post are closed


Kevin Jon Heller says

February 12, 2012

I have posted a lengthy reply to Dapo at Opinio Juris:

John C. Dehn says

February 12, 2012

Thanks, Dapo, for this post. I read the post to which you responded the same way you did. The revised claim in the above linked response to this is interesting and somewhat more defensible, but ultimately -- in my view -- wrong.

Intentionally "directing an attack against" the civilian population as such or making the civilian population "the object of an attack" is -- in the terms your interlocutor uses -- conduct. It is only by ignoring this mens rea requirement for the conduct, and instead making the presence of civilians not taking a direct part of hostilities or the civilian population as a "circumstance" that one can reach a different conclusion. That approach does not square with the language we are referencing.

The ICTY case law (linked in the response) does not change the requirement that an attack be specifically directed at the civilian population or civilians. The relevant sections of the opinion said:

Para. 105 (citing Kunarac) "The use of the word “population” does not mean that the entire population of the geographical entity in which the attack is taking place must have been subjected to that attack. It is sufficient to show that enough individuals were targeted in the course of the attack, or that they were targeted in such a way as to satisfy the Chamber that the attack was in fact directed against a civilian “population”, rather than against a limited and randomly selected number of individuals."

Para. 106 (citing Kunarac) "the expression “directed against” is an expression which “specifies that in the context of a crime against humanity the civilian population is the primary object of the attack”"

Para. 114 "If he [a victim] is indeed a member of an armed organization, the fact that he is not armed or in combat at the time of the commission of crimes, does not accord him civilian status.

None of this supports the idea that an attack against a combatant in the presence of some civilians is an attack against protected civilians, and even less so that it is an attack against the "civilian population if some civilians are present at the site of a discrete attack against a combatant, even if the combatant is not then in combat.

I also have trouble reading the grave breach provisions of Protocol I without reference to ALL of the relevant substantive provisions of the Protocol. The requirement that civilian population or individual civilians not taking a direct part in hostilties be the specifically intended object of the attack is the key to those provisions. This requirement is not altered by the slightly different language of Article 85. Article 85 must be read in light of them.

jpaust says

February 13, 2012

Dapo: I agree, and another question to consider in context would be whether a civilian is actually engaged in "rescuing" or also something else which might be a factor before we conclude that that person is or is not directly participating in hostilities. The status question is another, and I agree with you and Bobby that if we accept the ICRC's CCF approach and change of status, that the person is targetable in any event, but there is always the additional question of proportionality.
With respect to funerals and the BIJ conclusion of "deliberate" strikes on "funerals and mourners," one would have to ask exactly who was there? Was the funeral of an important member of the Taliban or al Qaeda and were some of the mourners also members, and were they discussing plans and operations at some point during the funeral? And more generally were some civilians killed voluntary or coerced human shields. And note the number of "civilians" killed -- from 282 to 535 (with a difference of 253), almost indicating a lack of real knowledge of numbers killed, who were they (e.g., members, those with CCF or DPH otherwise), what exactly were they doing? How important was the target (e.g., the top leader of the Taliban who was in Pakistan -- which is relevant to conclusions about proportionality)? Were there alternatives? and so on.
I have thought it important also to use a movie camera when considering whether a person is a DPH (if one does not use the change of status or ICRC CCF test) instead of a snapshot. For example, when a civilian directly participates in hostilities by shooting enemy combatants (and is surely a DPH by definition, per hypo) but pauses for a 3-hour dinner and thereafter picks up the gun, a movie camera would allow recognition that the DPH person actually did not revert back to an ordinary civilian or actually change from DPHing over time. Similarly, what if that DPH took an 8-hour pause to sleep, and thereafter picks up the gun again, and we can identify this pattern over a period of weeks, with a few other interruptions? A movie camera would allow recongition that the DPH person has not actually reverted to an ordinay ciivilian and, over time, is still DPHing (and, e.g., is targetable while sleeping). See, e.g., 39 Denver J. Int'l L. & Pol'y 569, 572 & n.16 (2011), available at
and 19 J. Transnat'l L. & Pol'y 237, 271-72 & n.90 (2010), available at
--which address the DPH standard and movie camera or process approach also in connection with self-defense targetings outside the conext of war.