Expanding Congressional Oversight of Kill/Capture Ops Conducted by the Military: Section 1036 of the NDAA

By Robert Chesney
Thursday, December 8, 2016, 6:25 PM

The National Defense Authorization Act for Fiscal Year 2017 ("NDAA") is likely to complete its journey into law this week, and so the time has come to spotlight some of the nuggets in it that might matter to Lawfare readers. One of the most important, in my view, involves the little-known, emergent oversight architecture associated with kill/capture operations conducted by the military in locations other than areas of "theaters of major hostilities."

I have written previously about the creation of a SASC/HASC-based oversight system designed to ensure greater Congressional awareness of kill/capture missions conducted by the military outside a theater of major hostilities. The statute calls such activities "sensitive military operations," and I'll use the acronym SMO).

Here is the current statute (10 U.S.C. 130f) resulting from these efforts. One way to understand this is to see it as analogous to the more-familiar oversight system we already have for Title 50 covert action (except that there is no requirement of a Presidential Finding and the oversight runs to SASC and HASC rather than SSCI and HPSCI). Kill/capture missions conducted as covert action will trigger oversight under that existing system, but such missions when conducted under Title 10 authority (note the potential relevance here of the "traditional military activities" exception to the statutory definition of covert action) would not trigger that particular system (despite presenting many of the same underlying concerns). In short, the new SMO oversight architecture helps minimize oversight dropoff when it is JSOC rather than CIA that is conducting a kill/capture mission outside the "hot battlefield" areas.

Well, the new NDAA will tweak the SMO oversight system a bit. According to the Conference Report, Section 1036 of the NDAA will make the following changes:

TIMING OF NOTIFICATIONS: Section 1036 requires notifications to be delivered to SASC/HASC within 48 hours of the SMO, in contrast to the current approach which simply says that notification must be timely.

UPDATES ON NOTIFICATION PROCEDURES: Section 1036 requires SecDef to give notice to SASC/HASC within fourteen days if and when DOD changes its procedures for ensuring compliance with the SMO notification requirement.

QUICKER NOTICE IN THE EVENT OF A LEAK: Section 1036 also requires SecDef to give immediate notice of an SMO, if possible, in the event there is an unauthorized disclosure of that SMO. (It would seem this would only be useful when the leak occurs within 48 hours of the SMO and when the regular notification has not already been given. I suppose the idea is that there could be an immediate leak right after an SMO, and SASC and HASC don't want to sit around for up to two days waiting to hear about the SMO through formal channels in such circumstances).

CHANGING THE SCOPE OF WHAT TRIGGERS THIS OVERSIGHT SYSTEM: Section 1036 also changes the definition of what counts as an SMO in the first place. The current version of section 130f defines SMOs as follows:

(d) Sensitive Military Operation Defined.—The term “sensitive military operation” means a lethal operation or capture operation conducted by the armed forces outside the United States and outside a theater of major hostilities pursuant to—

(1) the Authorization for Use of Military Force (Public Law 107–40; 50 U.S.C. 1541 note); or

(2)any other authority except—

(A) a declaration of war; or

(B) a specific statutory authorization for the use of force other than the authorization referred to in paragraph (1).

(e) Exception.—

(1) The notification requirement under subsection (a) shall not apply with respect to a sensitive military operation executed within the territory of Afghanistan pursuant to the Authorization for Use of Military Force (Public Law 107–40; 50 U.S.C. 1541 note).

(2) The exception in paragraph (1) shall cease to be in effect at the close of December 31, 2017.

Put in plainer terms, a kill/capture operation currently counts as an SMO only if it is: (i) conducted by the military; (ii) occurs neither in the US nor in a theater of major hostilities; and (iii) occurs under color of either the 2001 AUMF or stand-alone Article II authority (but not any other AUMF, such as the 2002 Iraq AUMF); (iv) but operations in Afghanistan under the 2001 AUMF are categorically excluded until the end of next year (i.e., they don't count as SMOs even if one thinks Afghanistan no longer is a theater of major hostilities).

So how does section 1036 change this? It ditches the old definition (including the Afghanistan exclusion) altogether, in favor of something that looks designed to be a bit broader (and that more clearly encompasses the use of lethal force in defense of others). Under the new version, 10 USC 130f(d) will define an SMO to encompass four scenarios, without respect to the domestic law basis for the operation:

(1) A lethal operation or capture operation--
(A) conducted by the armed forces outside a declared theater of active armed conflict; or
(B) conducted by a foreign partner in coordination with the armed forces that targets a specific individual or individuals.
(2) An operation conducted by the armed forces outside a declared theater of active armed conflict in self-defense or in defense of foreign partners, including during a cooperative operation.

Let's unpack that a bit, for it is formulated a bit unclearly. The gist of this definition is to define several distinct scenarios that require SMO oversight:

SMO scenario 1--US kill/capture away from hot battlefield: Part (1)(A) is similar to the status quo. It focuses on the US military conducting a kill/capture operation outside of places like Iraq and Syria (i.e., locations of publicly-acknowledged active combat operations).

SMO scenario 2--Foreign kill/capture with US military "coordination": Part (1)(B) is novel and quite interesting. Here, the idea is that the lead entity is a foreign entity (e.g., a foreign government's military or other security services, or for that matter a non-state armed group), but with some unspecified degree of US military support. Note that location this time is irrelevant. If a Pershmerga unit takes the lead in trying to kill or capture an Islamic State target in Syria, with JSOC coordination, it would count as a SMO subject to the notification system.

SMO scenario 3--Self-defense of US forces, self-defense of "foreign partners": Part (2) is a bit of a mystery to me. Part (1)(A) already covers US military kill/capture operations outside the hot battlefield, without reference to the underlying purpose or legal theory. Part (2) (which is specific to scenarios in those same locations but only where the aim is self-defense of either US forces or "foreign partner" forces) thus seems superfluous at first blush. So what might explain it? One possibility is that this is meant to encompass a type of activity other than a kill/capture operation. Go back and re-read (2), and you'll see that it just refers to "an operation," not a "lethal operation or capture operation" as in (1). Perhaps it's just an oversight, since the self-defense scenario certainly does imply force being used (and it's not clear to me why SASC/HASC would want to bring non-lethal, non-capture activities into the SMO framework anyway). At any rate, it seems more likely to me that inclusion of (2) was an excess-of-caution specifically intended to ensure that when force is used in self-defense of allies and the scenario is in the nature of a force-protection strike on an enemy force rather than a carefully-planned attempt to kill or capture a particular person--as seems to have happened some in Somalia--it is crystal clear that it still counts as a SMO.

Bottom line: Just in time for a potential expansion of the range of circumstances in which the U.S. military (particularly JSOC) engages in kill/capture operations, we are getting at least a slight expansion of oversight by the Senate and House Armed Services Committees. Seems like a good thing.