My first post on drones and targeted killing posed a set of ethical issues that the practice of targeted killings raises. My second post on the subject further analyzed the dynamics of drone strikes, arguing that neither law enforcement nor warfighting constituted the right ethical or legal framework for governing the practice. In that post, I concluded:
Drone strikes seem to me, when conducted apart from formally declared war, to fall between warfighting and law enforcement. Neither set of rules fully applies. But relying upon drone strikes apart from any moral framework invites abuse (power always tempts those who have power to abuse it), chaos (decisions based upon varying, even contradictory criteria), and undermines the rule of law nationally and internationally (as happened with the detention of those the Bush administration called illegal combatants following 9/11
This post offers an ethical framework, drawn from law enforcement and the just war tradition, for deciding when and how to use drone strikes for targeted killing in a morally justifiable manner.
First, this framework does not apply to targeted killings by drones (or any other means) against military personnel in the course of a formally declared war. In such instances, the jus in bello ethical paradigm from the just war tradition applies.
Second, this framework never applies when conventional law enforcement methods are applicable. The rule of law necessitates that police attempt apprehension; police killing a suspect is justifiable only when the exigencies of attempting apprehension necessitate it, e.g., in self-defense against a suspect presently attempting to kill police, hostages, or others.
Third, a nation's definition of the "rule of law" should be transparent, i.e., public. Nations can profitably collaborate on this definition, establish it in international law, and then publicize a list of those nations in which the rule of law reasonably exists/does not exist. At a minimum, the "rule of law" requires written laws, an independent and fair judiciary, law enforcement by police held accountable for conforming to the law, and safeguards for the accused (right to competent counsel, to avoid self-incrimination, to see/hear the evidence, to a jury trial, and to due process). Addressing this through international forums will pressure nations to adhere to the rule of law and discourage nations from prioritizing perceived short-term national interests over long-term healthy relationships, e.g., the U.S. has often supported friendly dictators for short-term gain to the long-term detriment of its interests and those of the citizens ruled by the dictator.
Fourth, targeted killings are morally permissible only in nations in which the rule of law does not exist and in which the nation planning the killing has made its assessment that the rule of law does not exist public. Making a public assessment affords the other nation an opportunity for reform as well as inviting international discussion of whether the assessment about the rule of law is reasonable and widely held.
Fifth, apprehension of the suspect by the police in the country in which the suspect is (unlikely since the rule of law does not hold in that nation) or by agents (military or intelligence) of the nation seeking to apprehend/kill the suspect is highly unlikely or would entail major military action. This principle, for example, precludes needing to conduct a major invasion, recognizes the implausibility of putting an uncompromised agent inside many groups (e.g., al Qaeda or a drug cartel), and yet incorporates the possibility of raids by Special Forces or intelligence personnel (e.g., the strike against Osama bin Laden should have attempted apprehension rather than killing).
Sixth, a court that is part of an independent judiciary must authorize the targeted killing. Requiring authorization by a court greatly reduces the possibility for an official within the executive or military to abuse the power to authorize killing. This requirement also means that more than one person, even if the court consists of a single judge, approves the killing because the person seeking the approval has obviously reached an affirmative conclusion about it. Similarly, requiring a court order makes moot the issue of whether the person is a U.S. citizen, since, regardless of nationality, the person benefits from a legal process designed to balance individual rights with collective security.
a. Establishing this court requires legal counsel that will act on behalf of the person the government proposes to target for killing. This legal representative's duties include challenging the completeness or accuracy of evidence presented, advocating any reasonable prospect for apprehension instead of targeted killing, etc. Although imperfect, requiring a representative incorporates the advantages of an adversarial process in avoiding abuse of power by the other parties. Legislation creating the court should also create or designate a State Department office to provide these legal representatives. Assigning this responsibility to the State Department precludes potential conflicts if the Department of Defense, Homeland Security, or Justice (all involved in prosecuting/attacking those who would destroy the nation) had the responsibility.
b. The court should rely upon a reasonable standard of doubt test in its decisions. A lower standard of proof is inappropriate because the court's decisions are literally matters of life and death. Utilizing this standard of proof attempts to balance risk to the individual with risks to the nation's collective well-being.
b. Obtaining a court's authorization is unlikely to impede the timeliness of action, especially if the court's procedural guidelines emphasize timely adjudication of cases. Even if this were untrue, without some system of checks and balances on the authority to kill, the potential for abuse outweighs any occasional advantages of rapid action. Furthermore, developing a case against an individual requires careful analysis of information generally gathered from several sources over time, i.e., the process is inherently deliberative rather than rapid.
c. Of necessity, these court proceedings might have to occur in secrecy. However, the court should preserve a full transcript of its proceedings, released to the public in ten years, or sooner if the court deems adverse consequences national security improbable. The accountability this transparency provides constitutes another important barrier against potential abuse.
Seventh, once the court issues an order authorizing a target killing, the President (or designated senior representative) will oversee implementation and approve any attempted execution. If apprehension of the target appears possible, then apprehension rather than killing shall become the primary goal (e.g., the suspect travels to a nation in which the rule of law prevails). Any approved targeted killing mission shall employ the minimum force consistent with reasonable odds of success and carefully weigh prospective collateral damage against the value of the target's death. In every case, uncertainty should favor restraint.
Eighth, a post-mission evaluation, with a report to the court that authorized the killing, should follow every targeted killing mission, whether or not the mission succeeded. In evaluating missions, the court and the executive should ensure that officials acted with reasonable due diligence given information available to the decision maker. The evaluation is intended to keep decision makers honest, not to indulge in unhelpful second-guessing in light of information known after an attack. The law should stipulate appropriate penalties for decision makers within the executive branch who wrongly execute their duties in reviewing and approving proposed targeted killing missions.
These eight protocols represent an initial attempt to articulate an ethical framework for targeted killing in cases that fall between traditional law enforcement and warfighting. With drones having killed an estimated 2700-4600 people, the need to establish clear ethical and legal frameworks is urgent. Although probably improvable through discussion and trial, the framework seeks to balance individual rights and collective security, preserving an option for targeted killing when apprehension is impossible and the likely cost of allowing the target to remain alive outweigh the international ramifications transgressing national boundaries and abrogating normal justice procedures.