Originalism and Drone Strikes
With drones strikes on U.S. citizens in the news (and see here for Jonathan Adler's roundup of commentary), below is the latest version of my originalist assessment of the matter. (An earlier version, focused on the al-Awlaki episode of 2011, is here, via the Harvard Journal of Law and Public Policy (HJLPP)).
(1) The core constitutional provision at stake is the due process clause of the Fifth Amendment: "No person shall be ... deprived of life, liberty, or property, without due process of law." As Nathan Chapman and Michael McConnell explain in their comprehensive 2012 article Due Process as Separation of Powers,
By the time the Fifth Amendment was enacted, everyone agreed that due process applied to executive officials and courts. It meant that the executive could not deprive anyone of a right except as authorized by law, and that to be legitimate a deprivation of rights must be preceded by certain procedural protections characteristic of judicial process: generally presentment, indictment, and trial by jury. . . . [D]ue process has from the beginning been bound up with the division of the authority to deprive subjects of life, liberty, or property between independent political institutions. In modern parlance, due process has always been the insistence that the executive – the branch of government that wields force against the people – deprive persons of rights only in accordance with settled rules independent of executive will, in accordance with a judgment by an independent magistrate.
Or, as Chapman and McConnell say elsewhere in the article, the English understanding of due process was a “guarantee of judgment by an independent institution according to procedures designed to take the case out of the hands of the king.”
If this understanding of the due process clause applies to drones strikes, they are evidently unconstitutional; in ordinary practice, of course, the President cannot simply kill people he thinks are enemies of the regime (just as the historical English practice imposed this limit -- at the time, a vitally important one -- on English monarchs).
(2) Although not free from doubt, the best reading of eighteenth-century ideas of jurisdiction and allegiance appears to be that constitutional protections such as the due process clause applied within U.S. territory and to U.S. citizens abroad (but not otherwise). (See especially Philip Hamburger, Beyond Protection, 109 Columbia Law Review 1823 (2009)). As I put it in the HJLPP article:
Extending constitutional rights to citizens abroad makes sense in an eighteenth century context. U.S. citizens, especially political and economic elites, commonly traveled overseas at the time. And executive agents could surely reach overseas as well. It would be odd if the founding generation thought that citizens would be exposed to the whims of executive discretion respecting their life and liberty simply by crossing the border. Further, the Constitution’s text itself does not have a territorial limitation, and though a limitation might well be implied for noncitizens – who were understood to be outside of both the allegiance and protection of the United States … the reasons for this implication would not apply to citizens. Relatedly, the understanding was that noncitizens were protected by international law rather than constitutional law; U.S. citizens in contrast would not have been protected by eighteenth century ideas of international law against actions by their own government.
That due process rights would not extend to noncitizens, incidentally, shows that the constitutional limit would not pose a material interference with the President's conduct of foreign affairs, since in most cases the targets would not be citizens.
(3) Despite the clause's unqualified text, there are obvious exceptions based in history. The executive could use deadly force domestically to stop imminent harm. And the rule did not apply to activities on the battlefield – it was never thought in the eighteenth century that battlefield combatants had any sort of protection against being killed, even if they were citizens fighting against their own country, nor that combatants had to check the citizenship of their opponents before launching attacks. But, as the President's recent memo on targeting makes clear, strikes are not limited to battlefield situations (unless one takes the whole-world-as-a-battlefield view), and strikes are not limited to exigent circumstances of imminent harm, unless one defines "imminent" to mean something is usually does not (see this discussion by Kevin Heller).
(4) It's possible that there is historical precedent for unilateral executive killing in the founding era that's somewhat analogous to the present situation, but I'm not aware of it. The closest that's been invoked is from the Civil War (as in this post by originalist scholar Saikrishna Prakash in the New York Times' Room for Debate). But not only is this far removed from the founding (and in any event most of the killing took place on the battlefield), the Civil War is a murky precedent. In a number of legal contexts the Lincoln administration took the position that ordinary constitutional law did not apply to the south because it was more in the nature of a foreign enemy (while at the same time denying that its secession was legal or effective). Thus, after initial hesitation, Lincoln did not treat southern soldiers as traitors, and -- as in The Prize Cases in 1863 -- he claimed authority to seize southern property (even noncombatants' property) as enemy property under the laws of war, without limitation by the takings clause, simply by virtue of the owners' residency in seceded states. Whether or not this approach was constitutional, it does not easily analogize to the situation of Americans traveling or residing abroad whom the executive suspects of aiding U.S. enemies. As in many respects, the civil war in this regard simply seems sui generis.
(5) Finally, assuming one accepts as a general matter that U.S. citizens have constitutional rights abroad, I don't see how, under the President's theory, the power of non-exigent, non-battlefield extrajudicial killing can be limited to overseas situations. If the President believes that someone within the U.S. is an enemy of the state, and he believes that capture is at the moment not feasible, what constitutional principle would prevent the President from using deadly force (apart from a principle that would also limit it overseas)? That prospect should be sufficient to cast great doubt on the whole argument. The most central concern of the due process limit in the eighteenth century and earlier was surely that the king not have power to kill domestic opponents by labeling them enemies of the regime.
In sum, the original meaning of the due process clause is that the President cannot unilaterally kill U.S. citizens he thinks are potentially dangerous. Perhaps there are examples of historical practice that suggest an exception applicable to the present case (as there are obvious historical exceptions on the battlefield and for prevention of imminent harm). But the burden should be on those who want an exception to the text, and that burden shouldn't be met merely by the claim that it would be more convenient to have such an exception.
UPDATE: Ilya Somin has a thoughtful post here: The Drone Targeting Dilemma. And according to this story, originalist-oriented Senator Mike Lee comments: “You would think that consistent with the principles of due process, the government shouldn’t be able to kill one of its own citizens without some kind of showing that they present an imminent threat.”
FURTHER UPDATE: Nicholas Rosenkranz points out this passage from his article The Objects of the Constitution, 63 Stan. L. Rev. 1005, 1042-43 (2011):
As a matter of grammar and structure, the Due Process Clause … is, at least at its core, a conditional check on executive power …. The central function of the clause is to create a check on such deprivations …. Here the check is generally judicial. Due process generally cannot be purely intra-executive …. All executive power is vested in a single person, and so an intra-executive check on executive power is not really any check at all.