The United States’ Use of Drones in Pakistan

Editors Note: We feature below a discussion between a group of leading United States academics on the US’s targeting of Taliban and Al Qaeda targets  in Pakistan. Each of the discussants is a  leading writer on international law, and on the use of force in particular.  We are delighted to post this discussion on EJIL:Talk! As usual, readers are invited to post their comments below.

The discussion kicks off with remarks by Professor Jordan J. Paust , Mike and Teresa Baker Center Professor at the University of Houston Law Center. His initial remarks were originally prepared in response to a request from the media for clarification regarding certain issues arising from US use of drones in Pakistan. The other discussants are Professor Mary Ellen O’Connell (Notre Dame Law School); Professor Leila Sadat (Washington University School of Law, St. Louis); Professor Tony D’Amato (Northwestern University School of Law); Professor Geoffrey Corn (South Texas College of Law); Professor Ken Anderson (American University, Washington College of Law); and Professor Timothy Waters (Indiana University at Bloomington).

I. Professor Jordan J. Paust 

In general, proportionate use of armed force against non-state actors who are directly engaged in ongoing armed attacks against U.S. military in Afghanistan are permissible acts of self-defense under Article 51 of the United Nations Charter [see my article in 35 Cornell Int’l L.J. 533, 538-39 & ns. 17-18 (2002)].  As such, they are not “assassinations” within the meaning of the laws of war, which would otherwise be impermissible and war crimes, or unlawful “arbitrary” killings within the meaning of human rights law.  Such Article 51 targeting could also occur in time of peace against non-state actors who are engaged in an ongoing process of armed attacks against the United States, U.S. military, or U.S. nationals abroad, and this would not mean that the U.S. is at “war” or in an “armed conflict” with the non-state actors that do not have the status under international law of an “insurgent,” “belligerent,” “nation,” or state, and we would not be at war with any other entity that was not attacked (e.g., the state in which the targetings occur).  For example, these are not acts of war against Pakistan.  Further, under Article 103 of the U.N. Charter, the right of self-defense against an armed attack and ongoing processes of armed attack by non-state actors trumps inconsistent international agreements.  However, all states also have an obligation under the U.N. Charter universally to respect and observe human rights.

However, there has also been targeting of those who, according to generally shared expectations within the international community, are not directly involved in the armed attacks (or taking a direct part in armed hostilities either), e.g., drug lords who merely finance al Qaeda or the Taliban.  I know that some disagree and argue that those who finance armed attacks can be considered to be taking a direct part in hostilities (bankers beware), but this is definitely a rare minority viewpoint that has not been accepted by the international community.  These acts of targeting would be illegal and the label “assassination” can apply within the context of an international armed conflict such as that occurring in Afghanistan and, as expanded in given instances by the United States, in other countries.  They might not be “arbitrary” within the meaning of human rights law, but there is little authoritative guidance in that regard and reasonable people might disagree.

I note also that any time U.S. military personnel equipped for combat are engaged in an armed conflict, it is in the interest of the United States to recognize that the conflict is of an international character in order for our soldiers to have “combatant” status and “combatant immunity” for lawful targeting during an international armed conflict as well as prisoner of war status.  Under international law, we cannot be at “war” with al Qaeda, but we have been involved in an international armed conflict with the Taliban when we internationalized a belligerency (to which all of the customary international laws of war had already applied) between the Taliban and the Northern Alliance on October 7, 2001.  Contrary to nonsense mouthed by the Bush Administration, human rights law applies during any armed conflict as well as in times of relative peace.

 II. Response from Professor Mary Ellen O’Connell (Notre Dame Law School):

I am in the process of completing a draft book chapter on combat drones. The final version should be posted on SSRN by October 9.  The analysis is at variance with regard to some of your points to the reporter.  Most significantly, it includes reference to the new ICRC guidance on direct participation in hostilities.  The guidance makes clear that individuals may not be targeted without warning outside a situation of armed conflict.  That position is consistent with both IHL and human rights law.

III.  Professor Paust’s Reply:

 I can’t believe that if the United States is attacked from non-state actors in Mexico that the U.S. must warn the non-state actors before responding under Article 51 of the U.N. Charter (which, under Article 103, would override inconsistent treaty law except for the human rights duties of all members under Articles 55(c) and 56) — and what human rights?  The right to life, for example, is set forth as merely a right from being “arbitrarily” deprived in ICCPR, art. 6 — and apparently with respect to persons who are under a state’s effective control.

 IV.  Questions and remarks from Professor Leila Sadat (Washington University School of Law, St. Louis):

 I wonder if the last point is too broad.  What does it mean to say that the “United States is attacked by non-state actors from Mexico?”  If drug lords open fire in U.S. territory and harm or kill U.S. civilians, that, in my view, is not an “attack on the United States” nor does it create an armed conflict – nor does it invoke Article 51 of the Charter, nor does it give the U.S. a right to send armed drones to Mexico to execute suspected drug lords.  I think such a use of drones, if leading to death, would violate international human rights law (and common decency, but that is a subject for another day). 

 So perhaps it is more useful to be quite careful in delimiting permissible and nonpermissible uses of drones (I agree with Mary Ellen on this one, and think there are more limits than Jordan’s statement suggests) so that nonspecialists don’t think that any violence by so-called “non-state actors” gives a state the territory of which has presumably been violated a right to respond with cross border death squads, be they machines or human beings – or, to put it another way, would it have been legal (or a good idea) for the UK to use drones against suspected Irish “terrorists?”  The McCann case (European Court of Human Rights) seems to say no.

 V.  Reply of Professor Paust:

Perhaps, and I take the point that one has to decide what constitutes an “armed attack”.  I had in mind an al Qaeda attack on the United States on 9/11 when communicating with the media person.  What if there was a Hezbollah rocket attack from Mexico on a U.S. military base in west Texas.?

VI.  Professor O’Connell’s Reply:

It is a conflicts of law question.  IHL and human rights law restrict states to police rules for the use of lethal force in peacetime. Moreover, as of May, the drone attacks were resulting in about 50 civilian deaths for every one intended target.  The numbers look better since then but we are investigating whether that is really the case or the reporting has changed.  The paper I referred to has an appendix confirming the 50-1 number, which initially came from David Kilcullen.  The final version of the paper will discuss why the numbers of civilian deaths have been so high.  One reason could be the involvement of the CIA in these attacks.  The CIA, to my knowledge, are not trained in LoAC/IHL.  Does anyone have different information on that point?

Regarding the attacks from Mexico, I am in accord with Leila, and as to any further hypotheticals, you might find the ILA Use of Force Committee’s Initial Report on the Definition of Armed Conflict helpful.

VII.  Remarks of Professor Geoffrey Corn (South Texas College of Law):

I echo Jordan’s response, which I believe calls into question (properly) the conclusions of the ICRC report.  Moreover, CIA General Counsel’s office received LOAC training from the ILAW Department at the [U.S. Army] JAG School.  Whether that training is provided to operatives I don’t know, nor do I know for sure if Drone strikes are conducted exclusively by CIA personnel, or are conducted by joint CIA/DOD teams.  If the latter, I have no doubt that the individuals involved are well versed on both the LOAC and the ROE.

 VIII.  Remarks of Professor Tony D’Amato (Northwestern University School of Law):

Regarding the exchange among Mary Ellen, Jordan, and Geoffrey:  Isn’t there an inconsistency in your comments on warning?  If Mary Ellen is right that drones are inherently inaccurate weapons, what’s the point of warning the targets?  Shouldn’t you be warning the innocent bystanders?  Yet how can you identify the innocent bystanders?

IX.  Professor Paust’s addition to Professor D’Amato’s questions:

And aren’t the drones “smart” weaponry — far more selective, proportionate than Bill Clinton’s use of missiles against al Qaeda in 1998? 

X.  Remarks of Professor Ken Anderson (American University, Washington College of Law):

 I’m actually on the road and unable to really focus as I’d like on this — I have a feeling, though, that my views are complicated and probably too heterodox from many responding here for me to really propose in-line edits – instead I’ll just refer anyone who might be interested over to a book chapter coming out in a policy book from Ben Wittes/Brookings Press on counterterrorism proposals to Congress.  I have a chapter there on targeted killing — a not finalized working version is up at SSRN.

 XI.  Response from Professor Paust:

I think that this topic has sparked real interest and certainly some disagreement — e.g., what is an “armed attack”?  is a “warning” necessary?  is “consent” from the state from which the armed attack is emanating necessary? are drones inherently indiscriminate (or merely their use in certain cases)?  Do human rights apply to those, for example, who are not within a state’s effective control [see ICCPR, art. 2]?  Do some even argue that Article 51 self-defense is not permissible unless it is against a state or “belligerent” or “insurgent” and the laws of war (or some of them) would apply?  I had thought that the Caroline incident was a well-known example of self-defense pertaining with respect to non-state actor attacks (although the British and U.S. disagreed about the test regarding particular means of response against a U.S. vessel in U.S. waters, the U.S. arguing that even self-defense measures against attacks that had already occurred and were ongoing had to be necessary, etc.) and that the Security Council recognition of the propriety of a U.S. use of force in self-defense after 9/11 was further evidence of generally shared expectation that non-state actor armed attacks can trigger a right of self-defense under Article 51 of the Charter.

XII.  Questions from Professor Timothy Waters (Indiana University at Bloomington):

Jordan, I was curious about one point in your initial comment, although I realize I am not seeing it in its full context. You mention that an attack against a non-state actor (like Al Qaeda) would be legal under Article 51 and “we would not be at war with any other entity that was not attacked (e.g., the state in which the targetings occur).  For example, these are not acts of war against Pakistan.” But why not? Unless we are imputing the acts of the non-state actor to the state, or advancing a theory that the state’s inability or unwillingness to halt the attacks somehow voids its sovereignty, why would we conclude that this was not an attack against it, because clearly, if we are launching missile strikes into Pakistan (even if aimed at particular non-state actors within), ‘that entity’ is being ‘attacked,’ even if our intention is to target a non-state actor within.

I suppose one could argue that an otherwise legitimate act of self-defense cannot, by definition, constitute an illegitimate act of war, or even that it cannot be an attack for purposes of triggering the targeted state’s own Article 51 rights (because legitimate defense cannot give rise to a right of defense), but given the complexities of evidence, it seems rather too much to say that the state attacked would be obviously wrong to treat the landing of missiles on its territory as a casus belli. Certainly it is interesting to see the theory of self-defense (which is a function of state sovereignty) deployed to convert, not only any civilians harmed in the attack, but also the legal edifice of Pakistan’s sovereignty, into a form of collateral damage.

 Related EJIL:Talk! Links

http://www.ejiltalk.org/usnato-targeting-of-afghan-drug-traffickers-an-illegal-and-dangerous-precedent/

http://www.ejiltalk.org/clearing-the-fog-of-war-the-icrcs-interpretive-guidance-on-direct-participation-in-hostilities/

http://www.ejiltalk.org/afghanistan-accedes-to-additional-protocols/

http://www.ejiltalk.org/what%e2%80%99s-in-a-name-the-gwot-redefinition-accomplished/