Columbia law professor Trevor Morrison sent the following email over the weekend concerning one of my posts on the Senate NDAA language. In essence, Trevor suggests that I am over-reading the provision, and he homes in on the effective date language to argue that the language, if passed, might not actually encumber FBI operations in the second of two scenarios I outline. He writes:
I share your first concern with the bill's requirement of military custody in all cases -- it looks like it would mandate cutting short potentially very productive FBI interrogation and the like. I wonder, though, if the text of the bill provides an answer to the second concern you raise.Your second concern is that section 1032(a)(1)'s "shall hold" language would require that the government cut short productive surveillance of un-arressted persons by proceeding with arrest and military detention. But look at section 1032(c), the "effective date" provision. It says that section 1032 applies to persons "who are taken into the custody or brought under the control of the United States on or after th[e] date" the statute is enacted.So can't one say the requirement of military detention doesn't apply until a person is "taken into the custody or brought under the control of the United States"? I don't think the text is anything like a model of clarity on this point, and arguably it would be better for this limitation to be made differently, in a section not also devoted to defining the general trigger point for section 1032. But still, doesn't this language suggest that section 1032 doesn't apply until the person is taken into custody, and thus that mere surveillance of a person would not trigger it?The alternative would be really weird, right? If section 1032's applicability depended on a person's being taken into custody on or after the date the statute is passed yet it also required the military to take certain persons into custody who are not yet in its custody, then the applicability of the "shall hold" provision would be contingent on the future event of the person being taken into custody. That would be bizarre, no? As soon as you arrested someone covered by this section, the very fact of their arrest would mean that you had been violating section 1032(a)(1) up until the moment of their arrest.Admittedly, section 1032(c) speaks of individuals who are taken into the custody "or brought under the control of the United States," which presumably means something less than being in U.S. custody. I'm not entirely certain what this would cover. Perhaps it means to cover persons who are detained by other countries, at the behest of the United States? But whatever it means, I would be surprised if it covered people who are simply being surveilled by the government.
Trevor's reading of the proposal has undeniable appeal as a way of mitigating its impact, and I would find it entirely convincing were it not for that phrase he flags at the end: "or brought under the control of the United States." As it is, that language is a bit of a wild card. But even with it, Trevor makes a good argument that the effective date language would give the government some latitude to address the nightmare the larger provision would see to create. That still leaves us with the first scenario, on which, alas, Trevor has not found a loophole.