Thoughts on S.3707, Part IV: Problems and Suggestions

By Benjamin Wittes
Sunday, September 5, 2010, 12:15 PM

The Graham habeas bill is far from perfect. It would benefit a great deal from serious engagement from the administration (which has failed so far even to respond to it), habeas counsel, and human rights and civil liberties groups. The following, in no particular order, are some areas in which I believe it could stand improvement.

First, the bill is too harsh when it comes to the burden of proof placed on a detainee who argues that he quit an enemy group before being captured. Under the bill’s proposed “withdrawal” standard, the government still has the initial burden of proving by a preponderance of the evidence that a detainee was part of an enemy force. Once the government meets that burden, however, the bill provides for a rebuttable presumption that the detainee remained part of such forces until the time of his capture. The burden thus falls on the detainee to prove withdrawal, and appropriately so, in my opinion. The trouble is that instead of requiring the detainee to prove withdrawal by the preponderance of the evidence standard, the bill would require the detainee to offer proof rising to the markedly higher standard of “clear and convincing” evidence. This seems both analytically troubled and rather unfair. It gives rise to the untoward possibility that a detainee might prove to a court the preponderant likelihood that he left enemy forces before capture, but be ordered detained anyway because he could not prove that by clear and convincing evidence. In my view, if the government’s burden is to prove the fact of membership by a preponderance of the evidence, and a detainee proves by a preponderance of the evidence that he has quitted enemy forces, then the government has not carried its burden and the detainee should prevail.

Second, the bill contains detailed provisions relating to discovery and the handling of classified information. These are issues that Bobby and I have not studied carefully in either the bill itself or in the existing habeas litigations, and that we did not treat in our report. I do not pretend to know them all that well, in other words. I am informed, however, that the provisions with respect to discovery are markedly more restrictive than current practice in the habeas cases. If this is correct—and a quick inspection of the bill and the reining case management order suggests that it is—it needs to be fixed. I haven’t seen a lot of evidence that discovery is a major problem in these cases, so my inclination is that current practice is probably balancing the equities relatively well. Unless that’s very wrong, this should be an area in which legislation seeks to codify existing practice, not alter it.

Third, the bill should be amended to provide twice-annual post-habeas administrative review for those detainees who lose their habeas cases. Under the Detainee Treatment Act of 2005, the Defense Department arguably is obliged to provide an annual administrative review process for all military detainees held outside the United States. This is a good start, but does not go far enough—and this bill is the right place to clarify and refine this review system. The administrative review obligation should be stated more definitively, and should specify that review occurs twice-yearly (thus matching up with the standard provided by the Fourth Geneva Convention, which American forces happily used as a guide to their massive detention operations in Iraq for many years). Considerable thought should be given, moreover, to fleshing out the details of the detention review process in order to ensure a proper balance among the competing interests it seeks to address.

Fourth, the bill should contain language clearly establishing that the government does not need to prove that a detainee poses a future danger in order to prevail in a habeas case. The D.C. Circuit has recently said as much, but a district judge had held the opposite in an earlier case, and it remains to be seen just where the issue will come to rest. The proper forum to revisit the danger posed by a detainee is in the administrative proceedings described above, conducted if and only if the detainee loses his habeas petition. Just as it is worth explicitly requiring such a provision in this statute, it is worth eliminating doubt on this point as well: the government’s power to detain in the first instance is a function of the detainee’s affiliation with or support for an AUMF-covered group, not whether the person is personally likely to engage in combat or other personally-dangerous activity upon release. (For certain types of prospective detentions, I actually support a dangerousness requirement, as described here, but that’s another story.)

The nature of what the government needs to prove, already a significant advance in the bill over current law, could use still greater specificity. In particular, a number of questions have arisen in district court cases on which greater guidance would be useful. For example, some judges have suggested that taking orders from an enemy group is the hallmark of being “part of” that group—and thus the key thing judges should demand that the government prove.  The D.C. Circuit in a string of recent decisions has suggested otherwise, but how best to understand the meaning of “part of” in the wake of these decisions remains unclear. And there remains a substantial question about what sort of “support” should properly give rise to detainability. My sense is that the best way to handle these problems is not by amending the definition of “unlawful enemy belligerent” itself but by instructing the courts to consider certain matters in the context of looking at “the evidence as a whole” and the “totality of the circumstances” (which the bill would already require the courts to do). Specifically, Congress might consider adding language to that passage along the following lines:

In considering whether the government has carried its burden of proving that a petitioner meets the definition of an ‘unlawful enemy belligerent,’ the District Court shall consider: (a) evidence that the petitioner operated at the direction and control of Al Qaeda, the Taliban, or associated forces, (b) evidence that the petitioner engaged in pervasive association over time with Al Qaeda, the Taliban, or associated forces of a type not susceptible to innocent explanation, and (c) evidence that the petitioner engaged in substantial financial, recruiting, or logistical work on behalf of Al Qaeda, the Taliban, or associated forces of a type not susceptible to innocent explanation.

Finally, the provision permitting the government to give notice to the court concerning associated forces should be clarified to provide for greater transparency and congressional involvement. As written, the provision would do nothing to prevent a secret, sealed notice to the court and no notification of Congress at all, when the government effectively expands the legal war by interpreting the AUMF as covering a new group. The law should require that these notifications be filed in public and that the intelligence and armed services committees of both houses of Congress should receive notice as well.