Search This Blog

Wednesday, December 21, 2011

ICC Decision on Complicity for Collective Crimes

The International Criminal Court issued a decision on December 16 that offered their first major interpretation of the enigmatic Article 25(3)(d) of the Rome Statute.  The provision is notorious for the conflicting interpretations it has generated.  I am completely confident that no one can be completely confident that his or her interpretation is 100 percent correct.  That’s why I titled one of my law review articles about it Joint Criminal Confusion.  There are no easy answers here.

That being said, I do believe that there are correct and incorrect interpretations of Article 25(3)(d), and the ICC Pre-Trial Chamber got much of it right in its decision in Prosecutor v. Callixte Mbarushimana.

For those of who you aren’t Article 25 junkies and don’t have the provision memorized, here’s the full text of Article 25(3)(d):

In accordance with this Statute, a person shall be criminally responsible and liable for punishment for a crime within the jurisdiction of the Court if that person… (d) In any other way contributes to the commission or attempted commission of such a crime by a group of persons acting with a common purpose. Such contribution shall be intentional and shall either: (i) Be made with the aim of furthering the criminal activity or criminal purpose of the group, where such activity or purpose involves the commission of a crime within the jurisdiction of the Court; or (ii) Be made in the knowledge of the intention of the group to commit the crime…

First, the court rejected an interpretation of Article 25(3)(d) that had been favored by Judge Cassese, i.e. that Article 25(3)(d) only applied to “outside contributors” who assist a collective crime from the outside, but who aren’t themselves a member of the criminal group.  I argued previously in Joint Criminal Confusion that the outside contributor interpretation was wrong and that it was based on a misreading of a nearly identical provision taken from an international convention on the suppression of terrorist bombings.  What’s at stake in this debate?  If Article 25(3)(d) only applies to outside contributors, then some other mode of liability must deal with insiders, i.e. members of the criminal collective.  That mode of liability could have been Joint Criminal Enterprise, which is applied by the ICTY.  Since the ICC rejected the outside contributor interpretation, I don’t think they are likely to adopt JCE anytime soon.

Second, the court concluded that there is a threshold requirement for what counts as a contribution that can generate liability under Article 25(3)(d).  The ICC concluded that there needs to be at least a significant contribution, though the Chamber declined to hold that there needs to be a substantial contribution, which the Chamber took to be a higher level of contribution.  In reaching this holding, the court cited this paragraph of mine from an early article on the Rome Statute and collective crimes:

Many members of the community may provide contributions to a criminal organization despite the fact that they disapprove of the group’s criminality. Merchants sell food, water and clothing to criminals; they sell cars and gasoline and repair their vehicles; they rent them office space, apartments and houses. These services are no doubt contributions to criminal organizations, since, without them, a conspiracy could not continue. Furthermore, these services may well be performed knowing of a gang’s criminal goals. However, these contributions are best viewed as commodities because they are readily available on the open market. (Of course, the sale of firearms or explosives is another story.) But if one merchant does not sell the gasoline, another merchant will. However, because this contribution is ‘intentional’ in the basic sense and is made ‘knowing of the group’s intention to commit the crime’, the merchant is criminally liable for the whole conspiracy under the Rome Statute.

Obviously, I’m pleased that the court decided that providing mere commodity services is insufficient to establish vicarious liability for someone else’s criminal endeavor.  I do, however, have one semantic quibble.  The Chamber assumes, as other courts have, that a “significant” contribution is lower than a “substantial” contribution.  I find this a curious use of these phrases, because I would have expected the opposite ordering.  To my ear, a significant contribution is greater than a substantial contribution, but other readers may feel differently.  In the end, it doesn’t really matter as long as everyone is clear which one is the lower category and which is the higher category.

This won’t be the end of the ICC’s jurisprudence on collective crimes.  It’s just the beginning, because closing arguments in Katanga and Chui are scheduled for early 2012, with certainly a Trial Chamber judgment soon after, I would presume.

Sunday, December 18, 2011

Unwiling or Unable

The central jus ad bellum debate emerging from the War on Terror is how to justify the use of force against a non-state actor (NSA) located within the sovereign territory of another state.  The standard move (and the US position) is to view the use of force as permitted as an exercise of self-defense sparked by an attack committed by the NSA.  The 9/11 attacks are the paradigmatic example.

The problem, of course, is what to do about the sovereignty of the state on whose territory the defensive force is applied.  First, if the state consents, the problem disappears.  Second, if the actions of the NSA can be attributed to the state on whose territory they reside, the problem also disappears.  The real question is what happens if attribution is not possible. 

One view is that the defensive use of force is permitted if the host state is unwilling or unable to stop the NSA from using its territory as a staging ground for attacks.  Ashley Deeks has a new essay expounding on this standard.  Under her view, the host state is liable to attack if it is unwilling or unable to prevent its territory from being used by the NSA -- a standard with historical roots in the law of neutrality which permits attacks against co-belligerents who refuse to declare their neutrality or who are unable or unwilling to prevent their territory from being used by belligerents.  Kevin Heller at Opinio Juris objects to this standard (as he has before) as unsupportable under international law.  According to Heller, there simply isn’t sufficient evidence of state practice backed up by opinio juris to demonstrate that the law of neutrality has been used within non-international armed conflicts with NSAs.

I want to step back for a second and address this question at a deeper level of abstraction.  The question is how one understands the violation of sovereignty.  To my mind, there are two possible views here.  In the first, any finding of a violation of sovereignty is per se an indication that the defensive force is illegal.  In the second, though, a finding of a violation of sovereignty is just the beginning of the analysis, but not necessarily determinative that the action was illegal. 

How can a finding that a violation of sovereignty has occurred not trigger a finding that the action was illegal?  Well, consider first the doctrine of necessity in international law (see this article on the topic by Robert Sloane).  I don’t think the doctrine could apply in this case, because its preconditions are quire narrow, but the structure of its analysis is instructive.  In a case where necessity were to apply, one would fine that a state’s sovereignty had been violated but also find that the other state’s actions were nonetheless lawful because they were necessary to avert a profound national interest, such as saving a large number of their citizens from a natural disaster.  Incidentally, this is the same type of analysis one finds in domestic necessity cases, particularly in private law.  The ship owner docks his ship to another person’s dock during a storm, damaging the dock in the process.  The dock owner’s property interest is violated, but the action is lawful because it was necessary for the ship owner to do.  (And, on an efficiency level, the ship-owner can compensate the dock owner for the smaller damage to the dock, as in the famous Ploof v. Putnam).

A similar structure exists in how civil law jurisdictions hand cases of individual rights. Whereas common law jurisdictions will simply ask whether an individual has a particular right, civil law jurisdictions will sometimes ask first whether there is a right, and then only second ask whether it was an abuse of rights (abus de droit) to assert or exercise that right or whether that right should give way under the circumstances.  As George Fletcher famously described it in The Right and the Reasonable, this is a two-step or textured structure of analysis, as opposed to a flat or one-step structure that one finds in American-style rights analysis.

My point here is to insist that the so-called "extraterritorial" use of force against NSAs can be analyzed in this fashion.  One can concede, for example, that the territory and sovereignty of the host state is violated during a defensive attack against the NSA, but this does not necessarily entail that the use of force against the NSA was, all-things-considered, illegal. 

Of course, one needs some doctrine to explain why force is appropriate against the NSA.  On this point, I think it is essential to understand the natural law origins of Article 51 of the UN Charter.  As I have argued before, Article 51 carves out the inherent right of self-defense from the Article 2 prohibition on the use of force.  The reference to inherent is, in my view, a recognition that the right stems from natural law, a point made even more clear in the equally authoritative French language version (droit naturel) of the Charter.  The natural law origins of the right suggest that the right is not limited to the state-centric confines of positivist international law; attacks by any group, including NSAs, trigger the right of self-defense within natural law, and the positivist international law embodied by Article 51 preserves (or carves out) that pre-positivist right.  As a final note, the fact that Article 51 simply says “if an armed attack occurs” rather than “if an armed attacks occurs by a state” further emphasizes that attacks by NSAs trigger the natural right of self-defense.

Ah, but what of the ICJ’s opinion in The Wall case, the most recent and authoritative decision by the World Court to address this issue?  (Heller also criticizes the Deeks essay for failing to adequately consider ICJ jurisprudence).  That’s a good point, but I would ask any reader to look at that Wall opinion and see what sources of law the ICJ opinion cites for the proposition that attacks by NSAs do not trigger the right of self-defense…

Saturday, December 17, 2011

Shades of Essen Lynching

A recent trial court decision in Pennsylvania has caught my attention, in particular because the facts of the case bear a striking resemblance to the originally obscure, but now semi-famous, Essen Lynching case during World War II.  At issue in both cases is the liability of individuals for group conduct – probably the central riddle of substantive international criminal law. 

First, here’s what happened in the Pennsylvania case, called Commonwealth v. Poland, 26 A.3d 518; 2011 Pa. Super. LEXIS 1784 (2011).  A young woman was walking in a subway station when she was attacked by a group of individuals.  She was beaten by about half of them, while the other half watched and provided encouragement.  After the attack, the entire group fled.  After summoning the police and locating the group on a sequestered subway train, she was able to identify which members of the group physically assaulted her and which members were the onlookers. 

The defendant in the case, Poland, was one of the group members who apparently watched and allegedly encouraged the assault.  He was convicted of conspiracy to commit aggravated assault.  On appeal, the Superior Court of Pennsylvania upheld the conviction, and concluded that the defendant’s individual conduct was sufficient to establish liability for conspiracy.  In particular, the court noted that the key element to a conspiracy charge is an agreement between the defendant and the attackers, though the agreement need not be explicit; it can be inferred from the circumstances. 

In Poland, the court said that the inference was legitimate because:
Poland was part of a group when [the victim] was unfortunate enough to walk by them; Poland was part of that group when individual members were either participating in the beating or cheering the assailants on; Poland and the others fled as a group after the assault. As in French, “the actors’ relationships and their conduct before, during and after the criminal episode established a unity of criminal purpose sufficient for the jury to find conspiracy beyond a reasonable doubt.”
First, the case is interesting because it stands in some tension with the received wisdom about American criminal law, i.e. that defendants cannot be convicted for watching a crime occur.  Of course, some U.S. cases (which are taught in all criminal law casebooks) have allowed convictions where the observers were found to have encouraged or incited the assault, though these cases are not exactly common, and usually the encouragement is something more than just watching the assault.  

In this case, Poland’s liability extended from his membership in a group whose members were either committing the assault or encouraging it.  Notice that the court’s reasoning does not appear to hinge on whether Poland personally egged on the assault.  Rather, the court’s reasoning seems to be that the group as a whole either participated, or encouraged, the assault, and that Poland was undoubtedly part of the group.

Second, the case is remarkably similar to Essen Lynching, an obscure and hardly noticed case until Cassese cited it in the Tadic opinion establishing JCE as a mode of liability at the ICTY.  Essen Lynching involved the death of two captured British airmen.  A Nazi officer ordered an underling to march the airmen to another location; the Nazi officer also said, within earshot of a gathering crowd, that the underling should not interfere with the crowd’s treatment of the airmen.  Predictably, the POWs were beaten by the crowd and killed.

Cassese cited Essen Lynching for the proposition that each member of the crowd was equally responsible for the death, regardless of whether they had delivered the fatal blow or not (or any blow, for that matter), because they all participated in a common criminal endeavor.  This was certainly the military prosecutor’s position, although from the skeletal opinion in the case it is impossible to know where the judges actually agreed with this legal theory.  I’ve written about the case here and here, and several other scholars have now extensively analyzed it, as well as the similar Borkum Island.  My general position is that vicarious responsibility is appropriate only if the defendant shared a bona fide joint intention with the physical perpetrator, even if the actual execution of the assault was delegated to the physical perpetrator.

Of course, there are differences between the cases.  Poland is ostensibly about conspiracy as a separate offense, while Essen lynching deals with vicarious liability and a mode of liability.  But both cases deal with the central issue of group criminality.  More specifically, it seems to me that both cases hinge on a rather controversial concept of membership in a criminal organization.
Incidentally, the ICC has just issued a lengthy decision on this very issue -- group complicity and Article 25(3)(d) of the Rome Statute.  I will discuss that opinion in my next post.

Thursday, December 15, 2011


As has been widely reported in the press, the confessed killer in the Norway mass killings, Anders Behring Breivik, has been declared insane by court-appointed psychiatrists.  After extensive interviews, the psychiatrists concluded that he suffered from paranoid schizophrenia and lived in his own delusional universe. Although this finding could, in theory, be overturned, it is worth discussing how the psychiatric conclusion will affect how the case proceeds. It turns out that the Norwegian system is quite different from how the U.S. or other common law jurisdictions would handle the issue.

According to press reports, Breivik’s trial will be limited to determining whether he is, in fact, the killer, and will not address his culpability for the killings.  Assuming it is determined that he committed the killings (which he freely admits), he will be sent to a state psychiatric facility until he is no longer a danger, with both regular psychiatric and judicial review of this finding of dangerousness (at least every three years).  If both psychiatrists and judges agree that he is no longer dangerous, he could be released back into society.

A similar process is currently ongoing for John Hinckley, who has been in psychiatric care for decades after attempting to assassinate President Reagan.  His doctors believe that he can be released with supervision, and a hearing will be held next month to determine whether he can be released from his hospital.  Prosecutors and the Secret Service are both objecting to the plan.

But much else from the Breivik case is quite different from how a similar case would be handled in the United States.

Under typical common law criminal procedure, the defendant himself has to elect whether to plead not guilty by reason of insanity.  Of course, some defendants are so insane that they cannot rationally make this determination, and many defendants dislike the stigma of mental illness, and therefore decline to plead insanity even if it is in their best interest.  In Breivik’s case, for example, he was apparently displeased when he first learned of the psychiatric assessment.  But the interesting fact about Norway's procedure is that the issue isn’t for Breivik to decide.

Under the common law procedural system, how do we protect defendants who are so crazy that they refuse to plead insanity?  A judge can declare a defendant mentally unfit to stand trial, meaning that the individual is so mentally incapacitated that he or she is unable to assist in his own defense, so as to make the trial a meaningful judicial exercise.  This standard is much higher than the standard used to determine criminal insanity. If the judge makes this determination, the defendant is sent into psychiatric care until the defendant is cured (enough to understand the trial process), at which point the defendant is sent back to the court for the trial to be held, even if that takes years.

The key difference in the Norwegian system appears to be that if Breivik is cured he will be released back into society, not sent back to the court to stand trial for his crimes.  In that sense, it appears that the psychiatrists are not making a determination about his capacity to understand the trial, but rather are making a determination about his mental situation as it pertains to his culpability for the massacre.  In the common law system, the idea that this determination (criminal insanity) can be made outside of the trial process is utterly unfathomable.  It is a party-driven process.  The defendant elects to plead insanity, the prosecution most likely objects, and then the judge explains the standards for criminal insanity to the jury.  The jury has the final responsibility to determine, after hearing all of the facts, whether the defendant understood the nature and quality of his or her actions (depending on which standard is used).

The real conclusion here is that the Norwegian system is far less party-driven than the typical common law system.  The court takes a much more active role in structuring these decisions – even something as crucial as criminal insanity. 

Some Norwegians reacted with disgust when they learned that Breivik had been declared insane.  Having not examined the defendant, I don’t want to pass judgment on the merits of that decision.  But suffice it to say that in the United States this decision would be much less likely.  Unlike in Norway, where the decision largely rests with court-appointed psychiatric experts, American courts would leave the issue of criminal insanity to the jury, after hearing expert witnesses hired by each side. And jurors are notoriously stingy about granting an excuse based on insanity.

Friday, December 9, 2011


Gareth Evans announces in the latest Foreign Policy that we have won the war against genocide.  By that he means that humanitarian intervention has become widely accepted as an appropriate response to atrocities in general and genocide in particular. 

His best – and most obvious -- piece of evidence is Libya.  After the world community was shocked by the Gaddafi regime’s use of force against peace protesters during the early days of the Libyan uprising, when it was a mere protest and not yet a civil war, NATO finally decided that it was time to act.  Although everyone knew that Gaddafi’s regime had engaged in brutality during its four-decade reign, the killings of political protesters in broad daylight was apparently too much.  Gaddafi was indicted by the ICC, NATO intervened with bombers, and even the Obama Administration got involved with drones and technical advisers – even allegedly flouting the War Powers Resolution in the process.

It’s certainly true that humanitarian intervention is a much more politically salient option today than it was during NATO's intervention in Kosovo.  However, it is important not to generalize too much from the Libyan example.  Calls for an intervention in Syria have largely gone nowhere, despite the fact that many of the government attacks against protestors equal, or even exceed, what happened to the protesters in Libya.

Furthermore, political acceptance of humanitarian intervention is one thing; legal acceptance is another.  One of the most popular legal arguments for intervention is the Responsibility to Protect, and I am often asked by my students what I think of the concept.  Although I support the legality of humanitarian intervention, I remain skeptical of RTP as a convincing doctrine.  Here’s why:

First, it’s important to remember that RTP comes in many variations, as Carsten Stahn admirably pointed out in a great early essay in the American Journal of International Law.   In its thinnest variation, RTP simply means that governments should insulate innocent civilians from the horrors of war.  This is rather uncontroversial, and to my mind is a conclusion that is already implicit (or even explicit) in international humanitarian law and international human rights law.

In its thickest variation, however, RTP attempts to provide an argument for unilateral humanitarian intervention in the absence of Security Council authorization. That's the holy grail for advocates of humanitarian intervention -- an argument that somehow gets around the Security Council's exclusive authority to regulate the use of force under the U.N. Charter (with the exception of Article 51 -- more on that below).  

Unfortunately, though, the argument for this thicker version of RTP is rather thin.  Here are the arguments: the concept was developed by an international commission (ICISS), then adopted by the General Assembly, and also enjoys wide support among the international community (thus evidence of an emerging opinio juris).  All three of these arguments constitute a slight-of-hand move, though.  First, the commission that came up with RTP was hardly international.  Yes, it had members from all around the world, but it was set up by the Canadian government, not the United Nations.  Hardly international.

Second, the General Assembly’s 2009 resolution on RTP (A/RES/63/308) was hardly an endorsement of any norm of unilateral intervention, which as I said above, is a chameleon of a concept that can stand for anything as mild as the need to protect civilians from the devastation of war.  In fact, the General Assembly resolution says absolutely nothing about justifying the use of force in the absence of Security Council authorization.  That's the important part, and the General Assembly has never endorsed this idea.  (The 2009 resolution says that the Assembly “Decides to continue its consideration of the responsibility to protect”).

Third, when states lend support to the idea of RTP, it is impossible to know whether they are endorsing the thin or thick version of the concepts.  And that, I think, is the beauty of RTP.  It can mean anything to anybody, so many states have reason to support it.  It's only when you start clarifying which version you are talking about that the disagreement really starts.

Responsibility to Protect has had a remarkable career in a very short time period.  Activists and even some scholars accept it like it is gospel, even though the evidence in support of it is questionable.  The concept has been well marketed, and the fact that it has a good website, and a great acronym (RTP or R2P) doesn't hurt.  Doctrines without acronyms always have a problem taking off.  At a certain point, so many people will be talking about RTP like it is law that it might actually become law.  It's the greatest example of bootstrapping that I've ever seen in public international law.  Credit where credit is due, I suppose.

Although I disagree with RTP, that's not to suggest that I'm an opponent of unilateral humanitarian intervention.  On the contrary, I believe it is legal in some circumstances.  In particular, my view is based on a reading of Article 51 of the UN Charter, which refers to the "droit naturel de légitime défense".  According to my position, outlined with George Fletcher in Defending Humanity, the concept of "Legitimate Defense" encompasses both self-defense and defense of others, thus suggesting that the use of force is legal to fend off an aggressor and to come to the aid of a victim of aggression, regardless of whether they request the assistance or not.  As a final note, the reference to "droit naturel" (natural law) indicates that the right belongs to nations and peoples because it predates positive law.  It is a right that extends all the way back to the international State of Nature.  It is a right that is recognized, but not created, by Article 51 of the Charter.

The virtue of recognizing this argument for humanitarian intervention under the Doctrine of Legitimate Defense (should I call it DLD?) is that it does so within the confines of the UN Charter and the carve-out announced in Article 51.  The thick version of RTP, by contrast, serves only to de-legitimize the UN Charter and its regulation of the use of force.  And that's dangerous fire to play with.  Even human rights activists have reason to worry once the UN Charter is no longer seen as authoritative on the issue of the use of force.  The consequences of that development could be disastrous.

Tuesday, December 6, 2011

Surrendering to the Drones

Last week I was at University of Pennsylvania Law School for a panel on the Legal and Ethical Implications of Targeted Killings.  The impetus for the panel is the forthcoming publication of a book that I co-edited with Claire Finkelstein and Andrew Altman called Targeted Killings: Law and Morality in an Asymmetrical World.  The book is scheduled to be released by Oxford University Press at the beginning of March.

Several of the questions during the panel centered on issues of surrender and capture.  Although technically the United States has a “capture or kill” program, the drones are often used for targeted killing operations.  True, they are also used for surveillance, which might precede an operation to capture a suspected terrorist, but the drone strikes we are concerned about involve the deliberate killing of suspected terrorists.

One of the comments during the panel discussion was quite elegantly phrased: You can’t surrender to a drone.  This is, in fact, a familiar point made in the media and the public discussion about targeted killings.  Drones are operated by soldiers (or CIA employees) who sit in front of computer screens located miles away from the actual attack, and the soldiers never see the victims with their naked eyes.  The drone operators scrutinize high-resolution video feeds captured by the cameras of the drones that will also launch a missile if the killing is approved.  Should this bother us, either from an ethical (e.g. philosophical) standpoint or a legal one?

I do not believe that it should.  The idea that a suspected terrorist cannot surrender to a drone suggests, somehow, that the appropriate course of action is to send in troops who could potentially accept the surrender of the targets and capture them rather than kill them. This argument deserves a little more scrutiny.

It is important to distinguish between the duty to accept surrender and the duty to afford the enemy the opportunity to surrender.  These are two different normative requirements, under both the laws of war and the morality of warfare.  Everyone agrees that a soldier cannot kill an enemy combatant who effectively communicates his desire to surrender to the opposing soldier.  In fact, declaring that “no quarter will be given” (i.e. refusing to accept surrenders) is both a violation of international humanitarian law and a war crime generating individual criminal responsibility. 

However, that’s entirely different from the duty to offer the enemy the opportunity to surrender.  If either the law or morality of warfare required such an opportunity, then it is hard to see how any form of aerial bombardment (whether by drone, manned fighter jet, or bomber) could be lawful under the proposed standard.  True, you can’t surrender to a drone, but you also can’t surrender to an F-16 or a stealth bomber or any other form of aircraft that fires weapons from a distance.  That doesn’t necessarily suggest that the proposed norm is incorrect, but it does suggest that it is highly revisionary and would require wholesale revision to current practices of modern warfare.  Although such a view would not necessarily be fatal to a moral requirement for warfare, I do believe it is fatal to a legal requirement that is necessarily based on customary international law (and requisite state practice). 

Another individual at the event asked about the Doctrine of Double Effect and argued that it would be ethically wrong to use a predator drone to attack a truck with a terrorist and an innocent civilian in it.  Under the Doctrine of Double Effect, one might argue that it is morally permissible to attack the truck because one’s purpose is to kill the terrorist and the death of the civilian is a mere side effect (though a knowing one).  The questioner rejected this application of the DDE, however, because he said that the attack should be carried out by soldiers, thus making the killing of the civilian not truly necessarily.

I take it, however, that his objection was not really about the DDE.  It was really more about the demands of the principle of distinction, i.e. distinguishing between combatants and civilians, and suggesting that this mandated one choice of weapon over the other.

Again, without fully answering the question on its own terms, let me simply note that the question isn’t really about drones, but more about air power.  And under the view described by the questioner, I take it that any form of air power in lieu of ground combat would be per se immoral because it would produce collateral damage.

What unites both of these subjects is the relevant baseline for the discussion.  Many of the asserted objections to targeted killings turn out not to be objections to drones per se but actually to the strategic use of air power generally. 

As it happens, though, the recent technological advancements associated with air power (and smart bombs) make these weapons more discriminating than they have ever been in the past.  A missile fired by a drone or F-16 today might ironically be more discriminating than an artillery shell fired a decade ago (or a missile fired today by a less technologically advanced country).  Furthermore, within the category of air power, a drone operator has more real-time high resolution images of his target than a fighter pilot, who in many cases is dropping a bomb on a set of defined coordinates regardless of who happens to be in the building at that time.