Why the President’s Targeted Killings are Illegal (According to Professor O’Connell)

by Julian Ku

Kevin has done, and is doing, a very nice job of critiquing the legality of the Obama Administration’s targeted killing policy.  On the critical side, it is also worth noting the views of Mary Ellen O’Connell, Professor at Notre Dame, who has become a leading public critic of the legality of this policy.  Her basic point is that international law only permits such killings on the battlefield, and any killings off of the battlefield (as she defines it) are illegal acts of extrajudicial murder. This would be true whether or not the U.S. actor is a privileged combatant.  I think this makes sense, even if I doubt it is right.  It does show, however, that the Obama and Bush Administration’s policies as to the nature of this war is pretty close (and getting closer).  Because it is President Obama, and because he has folks like Harold Koh, Neal Katyal, and Marty Lederman to defend these views, I don’t think there will be nearly the same level of controversy as during the Bush years.


19 Responses

  1. Response…
    Yet, my draft article addresses some of her claims and demonstrates why some of Mary Ellen’s claims are not based in general patterns of practice and patterns of generally shared legal expectation about such practice, and so forth.  There are some points of agreement as well, esp. re: who should be flying the drones.  Please see  http://ssrn.com/abstract=1520717
    JJ Paust 

  2. There is no rule about “battlefields”. No such rule has ever existed or been imagined in any previous war. When Pettigrew’s North Carolinians went looking for shoes and ran into Buford’s cavalry unit,  the little town of Gettysburg was not a battlefield. Yet both forces immediately attacked each other and a new battlefield was created.

    When the Japanese tried to track down individual Coastwatchers who were radioing information about Japanese ship movements in the Solomons, there was no “battlefield”. They were targeted individuals attacked by military force, frequently large units of searching infantry. They were located deep in the jungle specifically to be as far away from the enemy and any battle as possible. Yet they were legitimate military targets.

    Every individual fighter pilot who stumbled upon a solitary enemy plane and fought a dogfight knows, if he is lucky enough to be the one who gets away, that battles can be fought between individuals, and the war just happens whenever two enemies encounter each other wherever that happens to be.

    In Mogadishu there was no battlefield when Super Six One was shot down and crashed. Gary Gordon and Randy Shughart won the Medal of Honor for attempting to secure the crash site and rescue the four crewmen from a large force of hostile Somalis, but exactly when the rescue transitioned into a battle is a bit hazy. They probably would not have survived even minutes if they had to keep asking, “Is this a battle yet?” before opening fire.

    In a war you attack the enemy any time you see them. There is nothing in international law about “battlefields” being different from any place else. If you catch the enemy sleeping in a field, you can shoot at them. When an enemy solider paddles ashore in a rubber raft, you can kill him. If you see a scout behind your lines, you can shoot at him. Nobody calls these lone encounters a “battle” and nobody remembers them except the individuals involved.

    A war involves the entire territory of both countries and all international waters. When the US went to war with Afghanistan, international law does not distinguish between Kandahar and Chicago. We may have the military power to send planes and troops to fight in Kandahar, while the enemy may not have the power to send its troops to fight in Chicago, but that is a matter of military power and not international law. The South had not been able to mount a sustained attack in the North, until they did at Gettysburg. Legally the sope of the territory involved in the armed conflict involves the entire territory of the US as well as that of the enemy. If the enemy can launch even one ship, then a naval battle is possible in international waters anywhere on earth.

    When the US goes to war, every single square inch of the US is at war. Fighting can break out at any moment anywhere in the US where the enemy is able to project armed force. There is nothing in international law, in US law, or in rational thought that distinguishes your home from the house of a village leader in Khost province.

    In WWI and WWII Belgium was neutral, until that neutrality was violated. Violation of neutrality may be itself a crime, but once that neutrality was violated both sizes poured in troops and the formerly neutral country became a battlefield. Then in WWII Operation Torch, the Allies violated the nominal neutrality of Vichy France to land troops in Algeria. In war, there are parts of the world where the war isn’t being fought, until some ships appear on the horizon and then the war comes to your hometown.

    Napoleon marched a quarter of a million men into Russia. There was clearly a war. There was clearly an invasion. What there mostly wasn’t was a battlefield because the Russians avoided direct contact with the invaders. A few battles just happened, but the campaign shows that you can have an entire war without any real fighting. There still was a lot of dying. Lots and lots of dying even without fighting. A scorched earth policy can be an effective military defense, and starving an entire army to death is perfectly legal without a single identifiable battlefield.

    There are rules about civilians and enemy combatants. There is no rule about “battlefields”. There never has been. Nor is there any logical basis for one, nor can it be derived from a misinterpretation of any real law of war. It is something people made up out of thin air to support an imaginary distinction that simple logic would discard if someone actually stopped to think about it.

    Only in the US would any large number of people be dumb enough to imagine that, by definition, war is something that is always fought overseas somewhere. The French would not be this dumb. The Russians would laugh you out of the room. The Chinese know better. In living history they have all experienced war on their own homeland. America has been fortunate enough that wars have only touched us on the outskirts (Pearl Harbor, the Philippines, Guam). We have been lucky. International law is international. There isn’t one set of rules for the US and another for the rest of the world. In an armed conflict, we may hope that our power and our luck will continue to keep us safe, but do not confuse the lack of open conflict with some imaginary law that says that the war is not allowed to touch us, or our territory is legally different from their territory.

  3. Come on, Howard.  It’s really clear from the editorial to which Julian links that O’Connell is using “battlefield” as a metaphor for armed conflict.  Her comment about crossing the border into Pakistan is just one example.

  4. “Now cross the border from Afghanistan to Pakistan. In Pakistan the United States is sending wave after wave of unmanned drones to attack far from any battlefield.”

    So on 9/10/2001 the Army of Afghanistan (or at least the government in control of the southern 90% of Afghanistan) consisted of 45,000 mostly Pashtun militia including 10,000 militia men from the Pashtun tribal areas of Pakistan. The Pashtun tribes never recognized the legitimacy of the international border between Pakistan and Afghanistan since it essentially represented the limit of British colonial expansion from their core holdings in India.

    After “the planes operation” and the US response, this force mostly dissolved. The Pakistani militia and foreign fighters crossed the border from the Afghan Pashtun tribal area to the relatively safe haven of the Pakistani Pashtun tribal area where they would be bothered by neither NATO nor the Pakistani army. From this safe haven they continued to train soldiers and infiltrate Afghanistan and attack US forces.

    This is not to say that there is no difference between the Afghan Taliban and the Pakistani Taliban. Pakistan is a big country with a lot of different groups and even the Pashtun tribes are not entirely homogeneous. However, when 10,000 Pakistani tribesmen who were fighting US forces in Afghanistan cross the border and then continue to fight the US from Pakistan, I don’t see the fact that they have removed themselves from the immediate vicinity of the “battlefield” to have any significance in international law. It is a strategic retreat and relocation of an enemy army to a relatively safer position. We cannot allow it to have any legal significance. It certainly has no effect within the broad wording of the AUMF, and as long as Pakistan makes no effort whatsoever to restrain us, the border should be as unimportant to us as it is to those who attacked us.

    If they are training soldiers to attack us, there is armed conflict even if the trained soldiers have to march many miles before they make contact with our soldiers. If we are shooting at them, using unarmed drones, then there is an armed conflict even if there are no boots on the ground. Pakistan is not meaningfully “far from any battlefield”, it’s just a different battlefield.

    It would not matter, however, if all that was happening there was to train enemy soldiers for the conflict. Such training is continuous combat function under the definitions of the ICRC and renders the military training facility a target for military attack and the people there are enemy combatants subject to lethal targeting.

    ” Combat drones carry missiles that may only be used in armed conflict. They are not the weapons of police. Outside of a battle zone or zone of armed conflict, human rights law demands that government officials use lethal force only when it is clearly necessary to save lives immediately. In general, this means that police must warn before shooting” Again, the question is whether combat is occurring, not the geography and recent history. If as you suggest, “battlefield” is a metaphor for armed conflict, then by engaging the enemy with drone aircraft we create a battlefield, but then the use of drone aircraft is proper on the battlefield we just created. The metaphor is then self-validating and there is no actual complaint about US policy.

    On the other hand, if she really believes that there is a geographical limit of the use of military force to only those locations where the enemy engages US forces directly, which has to be Afghanistan unless and until the US puts forces in Pakistan so the enemy can engage them there, and she believes that rear areas where the enemy supplies, trains, and equips its soldiers and where attacks are planned are off limits because, having positioned these areas far enough away from US troops, they are no longer in an “active battlefield” and therefore immune to military force, well that is simply an attempt to invent an irrational imaginary international rule that has no purpose except to protect the enemy, kill American soldiers, and attack US standing around the world. So I think it is important to forcefully point out that this idea has absolutely no basis in law, history, or logic.

    On June 5, 1944 the beaches of Normandy were far from any active battlefield. There had been an allied raid at Dieppe, but that was long ago and far away. No armies had fought in France since the Blitzkrieg. Even so, the British and Americans seemed to insist on violating the international law restricting the use of lethal military force to areas adjacent to an active battlefield. They continued to send fighters and bombers to attack troops, trucks, and trains far from any area of combat. It is true that these were manned planes, unlike the unmanned aircraft used today over Pakistan, but I see nothing in this theory of law where the difference between manned and unmanned aircraft is significant. Rather, the question appears to be proximity to the “battlefield” and there was no battlefield in France on that day.

    Of course, the occasional attack by a plane was nothing compared to the completely unwarranted and illegal s**t storm the allies were about to release on the peaceful countryside of Normandy (far from any battlefield) on the following day.

    Or as I like to say, in war you are far from any battlefield until the battlefield comes and finds you. Meanwhile, if there are enemy planes overhead it is probably wise to duck.

  5. The comparison with France is inapposite: France was occupied territory, so it was in the midst of an armed conflict. Vichy was a puppet state, so it was controlled by Germany, which was of course part of the armed conflict. A more sound comparison would be a case where the Allies targeted a German in Spain or Switzerland and got away by claiming that this was legitimate. This would be similar to Israel targeting enemies in Yemen, for instance; or the US targeting enemies in Pakistan.  I am aware of the two different issues here (legality under IHL and violation of sovereignty), but am a bit unclear about the interaction of the two – probably because the rules of armed conflict are predicated upon ‘traditional’ notions of state conflicts and neutrality…Can you really distinguish between the two?

  6. Well, there was the US invasion of Mexico in search of Pancho Villa, an enemy of the current Mexican government who had crossed into the US and attacked towns and military units. Then there were the various attempts to choke off supplies and units of the NVA infiltrating down the Ho Chi Min trail through Laos or Cambodia. Then there was the attack on the Ehrenfels, a German ship in the nominally neutral port of Goa (Portugal) by members of the Calcutta Light Horse. Then there was the Archangel expedition against the Bolsheviks who had taken Russia out of World War I. We were on the receiving end when protests against the British stopping and searching our ships for their naval deserters triggered the War of 1812, or on the flip side of the same sort of thing when Roosevelt aided the British with the Lend Lease program while America remained nominally neutral before Pearl Harbor.

    The problem is that international law does not distinguish between a really neutral state, a “puppet” neutral state, a secret enemy, a secret ally, and all the other grey areas. This all misses the point. There are two completely, completely separate questions. The first asks whether the targeted individuals are lawful targets of lethal force or not. Are they enemy combatants in an armed conflict or civilians?  The second then asks whether an attack on this target is prohibited because it would violate the neutrality of some third country. The Graf Spee was a legitimate military target and an attack on it would have been completely legal under the laws of war, but during the short time that it was in Montevideo harbor it was protected by neutrality.

    However, the cited editorial does not claim that the attacks are a violation of Pakistani sovereignty. If it did, that would be a matter for Pakistan to raise, not something that any American Professor at Notre Dame has any authority to raise herself in the face of effective Pakistani silence. This article argues that the presence of enemy combatants on Pakistan soil, far away from a battlefield, means that the use of lethal military force is prohibited by international law because they have ceased to be engaged in an armed conflict. That is the claim that has no basis in logic. The Army of Afganistan remained a legitimate military target after it escaped into Pakistan just as the Graf Spee remained a German pocket battleship after it entered the River Plate. WWII remained an armed conflict which did not somehow stop after the German pocket battleship entered neutral territorial waters. The Exiter, Achilles, and Ajax may not have been able to continue the battle, but that was because they could not violate Uruguayan neutrality and had to wait for the battleship to leave on its own.

    There are some who will admit that a war against an enemy state is world wide, but then they argue that war against a non-state party is local. This is a misrepresentation of the real rule clearly spelled out by the ICRC. The uniformed soldiers of an enemy country are legitimate military targets at all times, but members of the organized armed forces of a non-state party to an armed conflict are targets only while they are engaged in continuous combat function. That rule says nothing about being far from a battlefield or the relative level of real neutrality of third party countries. It can apply to Yemen as well as Afghanistan. It has a local component, but the local question is whether the target is an integrated member engaged in continuous combat function (which can be preparation for combat) and not whether there has recently been a battle or any active armed conflict close enough to authorize a Predator strike.

    Once a unit of the Taliban/al Qaeda army establishes a military training base in Yemen that reaches the level of continuous combat function, then it becomes part of the worldwide armed conflict between the US and that army. If the Yemeni government is unable to take action itself, but agrees to turn a blind eye to US activity against that enemy, then attacks on that base and individuals of any nationality participating in continuous combat function are authorized under the laws of war. There is no need for a separate, independent battlefield to be established (really or metaphorically) in Yemen.

  7. Al-Qaeda is not an army.

  8. Terminology notwithstanding, Howard has stated much of the law accurately, in my opinion.  I disagree with Kevin about Professor O’Connell’s use of “battlefield.”  She has written a draft article (probably published by now – I read it a while ago) in which it is clear that her position is that human rights law, not the laws of war/IHL, regulates armed attacks outside a zone of active conflict.  Kevin implicitly made the same claim in his earlier post, at least as I understood him.

    That is the real crux of the matter.  Domestic law aside, one’s opinion on the legality of these attacks depends upon one’s view of the territorial applicability of the laws of war/IHL.  I am of the view that members of an armed organized group in armed conflict with one or more states may be attacked wherever they are found so long as they are fairly determined to be combatant members of that group.

    With that said, I have an entirely different view of the applicability of IHL to civilians supporting such groups outside a zone of active conflict whose activity does not rise to the level of taking a direct part in hostilities.  That view is based on the gradual historical development of the protection of civilians on the battlefield.

    Based on the historical development, it appears to me that the laws of war/IHL regulating the conduct between opposing combatants applies functionally (to their “belligerent intercourse” – which includes not only armed attacks, but also other interactions such as detention, care for sick and wounded, etc.), not territorially.  Common Article 3 of the Geneva Conventions injected territoriality into non-international armed conflict only because, as a treaty, it could bind only high contracting parties.  On the other hand, the laws of war regulating the conduct of combatants toward civilians sometimes apply functionally (e.g. proportionate attacks) and sometimes apply territorially (e.g. occupation law, security threat detentions).  It was a failure to appreciate this distinction that led to some of the more outlandish, little-old-lady-in-Switzerland claims of the prior administration.

  9. Al Qaeda is a waqf, a charitable foundation established under Shari’a entitled to receive the required charitable donations or Zakat of wealthy Muslims. It was established in the 1980’s to support the foreign born soldiers who came to Afghanistan to fight the Soviet invaders. In the 1990’s it resumed activity in Afghanistan, raising money in Arab states to recruit, train, and equip foreign fighters to support the Taliban led government in its war against the Islamic Front for the Salvation of Afghanistan (aka Northern Front). It also trained a few fighters for the conflict in Bosnia and then Chechnya. Discovering that they could raise more money and recruits by emphasizing Chechnya than the Northern Alliance, they focused all their publicity on the Chechen part of their operation, even though of the 18,000 soldiers they recruited, trained, and equipped only a few dozen ever made it to Chechnya.

    Soldiers trained by al Qaeda formed the 055 Brigade (aka Ansars), which was an important part of the front line Afghan Army. The Afghan Army retained its tribal and regional unit identities just as the government remained a loose coalition of tribal leaders based on the Afghan Jirga tradition and not Western models. Bin Laden and the Shura Council of al Qaeda occupied a position in that government equivalent to that of an Afghan tribe. Although not ethnic Afghans, they were fellow Muslims and, within the theology of the Taliban, that made them “brothers”.

    So while al Qada was not an army, it was responsible for raising and supporting a component of the army, just as, for example, the Durrani would have contributed militia. Al Qaeda was also not a government, but it was a component of the Jirga and participated in military decisions.

    Alternately, in the first two years of WWII the American Volunteer Group (AVG, aka Flying Tigers) were a unit of the Chinese Army fighting the Japanese. This unit was made up entirely of Americans and operated under the somewhat independent leadership of Claire Lee Chennault. Arguably they were mercenaries, although after Perl Harbor the entire unit was integrated into the US Army. The AVG was not an army or a government, but it was part of two separate armies at two different times. The Japanese may not have recognized the AVG as lawful combatants, but the US regarded them as such.

  10. John,

    I am of the view that members of an armed organized group in armed conflict with one or more states may be attacked wherever they are found so long as they are fairly determined to be combatant members of that group.

    I am genuinely curious about a few things.  Where is your support in IHL for the idea that the occasional terrorist attack qualifies as “armed conflict”?  And where is your support in IHL for the idea that you can simply lump any alleged al-Qaeda member regardless of ideology or location into the category “al-Qaeda,” such that you can impute the attack of any member of al-Qaeda to all of the members?  And where is your support in IHL for the idea that once you have created a supposedly unified global al-Qaeda and identified a particular terrorist attack or set of attacks that supposedly rises to the level of armed conflict, that armed conflict extends world-wide such that you can attack a member of al-Qaeda regardless of location?

    Your and Howard’s approach to terrorism might make more sense than the current one (though I don’t think so), but I don’t recognize any currently-existing IHL in it.

  11. Adding to KJH’s questions, what are the implications for your views of article 1(2) of AP II?

    [The text of art. 1(2): “This Protocol shall not apply to situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature, as not being armed conflicts.”]

    Although the U.S. hasn’t ratified AP II, I still think it’s relevant, insofar as it is at least evidence that IHL distinguishes between armed attacks of varying severity.

  12. Anonsters,

    The conflict with al Qaeda is not an “internal disturbance” by any reasonable interpretation of that term, in my opinion.  The rest of the terms listed are descriptive of the types of internal disturbances to which the treaty does not apply (“such as”).

    Next, regardless of whether that language reflects universal customary international law for the threshold of violence necessary to the existence of armed conflict, the gravity of violence is relevant to whether an armed conflict exists, not just its sporadic nature.  The terms are descriptive and not exclusive (“such as”).  Historically, limited wars between nations often involved sporadic, low-threshold violence.  This was not a bar to the application of the laws of war.

    It is possible, then, to view 1(2) as a subset of non-international armed conflict to which the protocol applies for states party to it, leaving customary international law to regulate armed conflict outside its scope.  Of course, armed conflict must genuinely exist.

    Kevin, this should address some of your questions.  I do not have time to respond further right now.  I will only say that were it not for very modern notions of international human rights law, I suspect we would not be having this debate at all.

  13. Fair enough.

    Your response still doesn’t answer the question about categorizing people as al Qaeda for the purpose of exposing them to attack. You said:

    I am of the view that members of an armed organized group in armed conflict with one or more states may be attacked wherever they are found so long as they are fairly determined to be combatant members of that group.

    In your view, what makes someone a “combatant member” of al Qaeda?

  14. what makes someone a “combatant member” of al Qaeda?

    Al Qaeda has very few actual members. It may have trained and equipped as many as 18,000 combatants, but they were members of the Afghan Army, not of al Qaeda. Even KSM who conceived, planned, and commanded the planes operation on 9/11 does not consider himself to be a member of al Qaeda, having declined to take the personal oath of allegiance to Bin Laden. There is an unnamed organized armed force recruited, trained, and equipped by al Qaeda and led by people associated with it, but you should regard al Qaeda as a funding source. The armed force used to be the Army of Afghanistan and, though often referred to as “the Taliban”, it is currently an unnamed organized armed unit engaged in an armed conflict in Afghanistan, Pakistan, and elsewhere.

    I would suggest that a starting definition of becoming a “combatant member” of this enemy armed force is roughly the same as becoming a combatant member of the US Army. You go somewhere where the enemy is in control and sign up, go through basic training at a military training camp, take the oath, and then pull guard duty, peel potatoes, and dig latrines. It used to be that the uniform distinguished combatants from civilians, but with an army that does not wear uniforms making the distinction is harder to the outsider. All the other stuff (basic training, etc.) remains the same with or without uniforms.

    Actual “terrorists” are not combatants by this definition (although combatants can volunteer for terror missions). If you strap a vest around a 13 year old and teach him nothing more than how to push the button and kill himself and a marketplace full of civilians, he does not become a combatant through this process. You cannot become a combatant by watching TV and supporting terrorists any more than you can join the US army by mail order.

    Expanding the traditional definition of combatant to include a larger part of an actual terrorist network may be something that others may prefer, but it has not proven to be necessary for any of the prominent detainees that the US has held or targeted as combatants (except for the Bosnians in the Boumediene case who not plausibly be called combatants by any reasonable definition).

  15. Howard:

    What would you do with the distinction between those who directly participate in hostilities and those who do not?

  16. Direct participation in hostilities is required in order for civilians to be targeted. Members of an organized armed unit of a party to an armed conflict can be targeted if they are engaged in “continuous combat function”, which can be simply preparation or planning for combat. (See the ICRC Guidance on Direct Participation) “Continuous” in this context means that they have not withdrawn from the armed unit to go home or do something non military for a long period, it doesn’t require that they engage in combat function all day every day.

  17. Anonsters, where would you draw the line? Specifically, which of these, assuming that the US is at war with A-Q, do you consider to be engaged in direct participation in hostilities:

    1 A US army serviceman on base but off-duty
    2 A US national guard serviceman performing a non-armed supply chain role in Afghanistan
    3 A US national guard serviceman performing a non-armed supply chain role in Iraq
    4 A US national guard serviceman performing a non-armed supply chain role in a neutral third State eg Qatar
    5 A US national guard serviceman performing a non-armed supply chain role in the United States
    6 A US non-enlisted contractor performing armed escort duties
    7 A foreign non-enlisted contractor performing armed escort duties
    8 A foreign non-enlisted contractor performing unarmed supply chain services in Afghanistan
    9 A Pentagon analyst who briefs Pakistani intelligence
    10 A State Department analyst who briefs Pakistani diplomats
    11 A US non-enlisted contractor performing unarmed supply chain services in a neutral third State eg Qatar
    12 A US non-enlisted contractor performing unarmed supply chain services in the US?

  18. hm, those fancy html buttons don’t work as well as intended, will fix…

  19. Uniformed members of a regular armed force (1-5) are combatants subject to targeting at all times, unless they are rendered incapable of combat as by injury.

    Members of regular armed units of a non-state party to a non-international conflict are targets when they engage in continuous combat function (not direct participation). Every case except 10 could represent continuous combat function of training for combat, preparing for combat, or planning for combat provided that the supply chain function involved combat supplies, but 1-5 are regular soldiers and 6-12 are civilians so the rule doesn’t apply to any of them.

    6 and 7 represent civilians who might end up directly participating in combat if they are attacked by privileged belligerents. The problem here is that none of the enemy have been privileged belligerents, so an attack on a civilian guarded convoy by unprivileged belligerents may not be regarded as combat since nobody involved has combatant privilege. 6 and 7 are not directly participating in anything if they are simply armed but are not attacked.

Trackbacks and Pingbacks

  1. There are no trackbacks or pingbacks associated with this post at this time.