What should Americans think about violating the sovereign space of other countries and killing a U.S. citizen who is a "senior operational leader" of al Qaeda or an associated group? Doing so may or may not be a bad idea, but it is surely a big one.
The Justice Department memorandum on the subject that was leaked earlier this month, however, is of limited use in helping think through these matters. The memo sets forth three criteria that must be met before a lethal strike against a U.S. citizen located outside an area of active hostilities is to be legal. The individual must pose an "imminent threat of violent attack against the United States"; his capture must be judged to be "infeasible"; and the operation itself must conform with "applicable law of war principles"—meaning more than anything else that there is little chance that innocent life will be lost when launching an armed drone.
The imminent-threat requirement sets an impossibly high bar, as the detailed intelligence to make such a judgment is rare—terrorists, unlike armies, tend not to reveal their hands before they act. The inclusion of this condition in the memo almost certainly stems from traditional international law, which has long judged that actions launched to pre-empt an imminent threat are in reality a form of self-defense.
An armed response to an imminent threat is considered to be qualitatively distinct from preventive actions, which are aimed at a threat that is gathering but not imminent. Such actions have little standing in international law, as a world of frequent preventive actions aimed to interrupt threats perceived as gathering would be chaotic.
But terrorism doesn't fit this traditional distinction. Terrorist threats exist in a gray area—they are more than gathering but not necessarily imminent.
Instead of the Justice Department criterion, what is needed is a standard that would establish as legitimate targets those individuals who are part of al Qaeda or like groups dedicated to and capable of attacking the U.S., even in the absence of detailed data about an imminent attack. It must be kept in mind that the U.S. is engaged in a continuing state of war with al Qaeda and its offshoots.
The second criterion, regarding the unfeasibility of capture, should be kept. But capture must also be feasible in the near term and of low risk to American or host-country forces that would be asked to undertake the mission before a drone strike would be ruled out.
The third criterion in the memo—consistency with the laws of war—fails to go far enough. U.S. drone policy must also be consonant with smart foreign policy. This means a strike must be undertaken only when it includes the near-certainty that the target is a highly dangerous terrorist, that the strike is likely to succeed, that collateral damage will be minimal and there is no viable alternative, and that the estimated gains from killing the terrorist would be greater than the possibility the strike would alienate the local government and population, thereby generating more terrorism and less cooperation.
Such considerations would rule out "signature" strikes, which target people who are behaving in ways that resemble how terrorists tend to behave—such as groups of men carrying weapons in areas frequented by terrorists. Instead, specific evidence of an individual's involvement in terrorism would be required. Such considerations would also tend to rule out strikes against low-value targets, strikes where innocents would be killed, and strikes not all-but-certain to succeed.
The standards I am arguing for here would lead to fewer drone strikes. There is a danger that policy can be too restrictive, making impossible what should be difficult. But the process that currently exists for authorizing drone attacks lacks sufficient controls, especially when the targets are U.S. citizens.
It can and has been argued that a citizen who joins a terrorist organization abroad at war with this country forfeits his right to due process. Still, the U.S. government needs to proceed with great caution.
Drones are a relatively new but fast-developing technology. Dozens of governments as well as militias, gangs, cartels and terrorist groups will come to possess them, especially those configured for reconnaissance as opposed to strikes. The U.S., by what it does and how it does it, ought to set and reinforce a norm that the use of armed drones should be exceptional.
This argues for reform of U.S. policy and procedures that is based on public discussion, not in classified memorandums. In his State of the Union, President Obama promised a "durable legal and policy framework" that would be forged with Congress and be "more transparent." He should be held to his pledge.
Foreign governments and their citizens should understand America's policy on drones and how we arrived at it. This will not influence those most opposed to us, but it will help with the majority of governments and people around the world who appreciate the need to balance legitimacy and self-preservation.
Mr. Haass is president of the Council on Foreign Relations and the author of "Foreign Policy Begins at Home," to be published by Basic Books this spring.
This article appears in full on CFR.org by permission of its original publisher. It was originally available here (Subscription required).