And Gabor Rona of Human Rights First, in an email yesterday, told me that he “must take issue with your endorsement of your reader’s ambivalence about suing Rumsfeld as an attempt to ‘relitigate.’ Fact is, you can’t ‘relitigate’ that which has not yet been litigated.” Both Rona and Serwer make good points. Yet, I confess, my mixed feelings remain–and my sympathy with all three points in my original reader’s email remains as well. The doctrine here is fuzzy, and Rumseld has some strong defenses that may well prevail before the Supreme Court. At the same time, these defenses would, if they do prevail, lead to an absurdity. Yet, on the third hand, what will come from extending Bivens to these cases will, in the long-run, not prove salutary either. Rather, the litigation–relitigation or not–of these Bush-era cases will create real costs for war-fighting that are hard to envision prospectively but that will prove no less real for that fact.
I'm certain that I'm less concerned about the implications for "war-fighting" that allowing the torture civil suits to proceed will have than Wittes. But the fact remains that, even if these cases are allowed to go forward, the courts have shown an extraordinary amount of deference to the executive branch in these areas, and in allowing the cases to proceed the courts would most likely draw some pretty bright lines allowing only cases similar in their "absurdity" to come forward in the future.
On the other hand, if the cases are blocked one way or another, there will be no deterrent -- criminal or civil -- to government officials allowing violations of the constitutional rights of American citizens in the future as long as those violations are committed in the name of national security. The Obama administration has already made the decision to look forward in all but two cases of detainee abuse since 9/11; it's past time for some manner of accountability, if only because the potential for moral hazard here is enormous, and the possibility of meaningful restrictions on executive power rather less so.