I'll preface this by saying that, with one exception that I address in the column (regarding legal realism) so far I've found Sen. Graham to be the best of the questioners (R or D) by a fairly wide margin. From his opening statement ("Unless you have a complete meltdown, you are going to get confirmed") to his tough but fair questioning about the concerns about Judge Sotomayor's alleged bullying from the bench (not allowing her to attribute the complaints to her tough questioning, given that her 2d Circuit colleagues also ask tough questions but do not elicit the same reaction), Graham has struck me as honest, fair-minded, and astute. That's not to say, of course, that I agree with everything he has said or implied in his questions.
Now to the homework assignment. Near the end of a wide-ranging round of questioning, Sen. Graham asked a question about the legality of indefinite (or in the Obama argot, "prolonged") detention of enemy combatants. Here it is:
Under the law of armed conflict, do you agree with the following statement, that if a person is detained who is properly identified to accepted legal procedures under the law of armed conflict as a part of the enemy force, there is not requirement based on a length of time that they be returned to the battle or released? In other words, if you capture a member of the enemy force, is it your understanding of the law that you have to, at some period of time, let them go back to the fight?Judge Sotomayor pleaded that she was not a specialist in the law of war (fair enough), and so Sen. Graham asked her if she would think about the question a bit and have an answer for him in round 2. My guess is that she won't answer the question directly, because this really is the sort of thing that could come before the Court. But Sen. Graham does deserve an answer, so I'll take a crack at it.
I haven't fully researched the law of armed conflict on this point, but I'll accept for purposes of argument the claim (which Sen. Graham made a bit later) that in fact the Geneva Conventions do not set any time limits on detention of POWs or other combatant detainees. Nonetheless, there are a number of reasons why we might think that even if Nation X fighting Nation Y can hold Y's soldiers for as many years as the conflict ensues, the answer could be different for a conflict between Nation X and a non-state actor.
For one thing, in a case of armed conflict between sovereign nations, there is some reciprocity. POW exchanges can be negotiated, as can terms for parole of released captives. To some degree, this is also true even with non-state actors. For example, Israel has negotiated prisoner exchanges with non-state actors. It might even be possible for the U.S. to negotiate terms of release with some of the non-state entities with which we are fighting, such as the resurgent Taliban. However, for jihadists only loosely affiliated with groups that themselves have no clear territorial or other quasi-sovereign base, diplomacy seems likely to be unavailable to reduce (on humanitarian or other grounds) the duration of detention. In such circumstances, the law of armed conflict might depart from the POW model.
Likewise, there is an expectation that armed conflicts between sovereign states end. To be sure, we can speak of the Hundred Years War, but that was really a series of wars, each admittedly quite long but none longer than a quarter century. And most inter-sovereign wars are much shorter. Perhaps more importantly, inter-sovereign wars have discernible end points, even when, as with the Korean conflict, there is no formal peace treaty. Thus, while indefinite detention of POWs is a theoretical but unlikely possibility for inter-sovereign conflicts, for sovereign/non-state conflicts, the norm is flipped. We do not have an expectation that the conflict will end rather than peter out, and so just about everybody initially detained based on a combatant status determination will be detainable indefinitely.
These problems are exacerbated by the difficulty of identifying combatants for non-state enemies. Somebody can be, in Sen. Graham's words, "properly identified to accepted legal procedures under the law of armed conflict as a part of the enemy force," such that there are sufficient legal grounds to hold him initially, but over time the stakes will rise, perhaps at some point making the substantial risk of initial mistakes in the fog of war too great to justify further detention. To put it slightly differently, something like a preponderance of the evidence that Joe Blow is a battlefield terrorist may be enough to hold him for a day, a month, or even a few years, but we might require more to hold Blow (who claims that he was simply an errant tourist, aid worker, or journalist) for life.
Now I'll freely acknowledge that I have made no effort here to tie any of these considerations to the actual terms of the international law of armed conflict. The closest thing I can find to a suitable text would be Article 109 of the 1949 Geneva Convention on POWs, which gives contracting parties the discretion to make arrangements for transferring detainees to neutral countries if they have "undergone a long period of captivity." That--and the references to cessation of hostilities--strongly suggests that the relevant body of rules was written without any clear thought about what to do with people like some of those held at Gitmo. One might therefore think that according a right to something like eventual repatriation would be consistent with the spirit of the Geneva Conventions, even if one grants Sen. Graham that the letter of the law does not require such a right.
But to say that is only to begin the discussion. One might think that due process (to be litigated in a habeas court or an acceptable substitute) demands more for prolonged detention than for initial detention. And of course Congress could demand something of this sort by statute, regardless of whether it is required to do so by international law or the Constitution. So the ball is back in your court, Sen. Graham!
Posted by Mike Dorf