I’m writing a paper for a conference in December on aggression. My paper is provisionally titled, “When May We Kill Government Agents?: Against the Special Immunity Thesis”. This will eventually be part of a larger project on disobedience, sabotage, and resistance. (For some past thought experiments on this topic, see here.)

The basic idea is this: Ordinary people can become liable to be killed under certain conditions. A person is liable to be killed when he is doing something deeply wrong, unjust, or harmful to others, and when killing him would serve some kind of defensive purpose, such as self-defense, the defense of others, or to prevent him from causing greater injustice. The permissibility of killing is also restricted by a doctrine of necessity: at the very minimum, when a non-lethal alternative is equally effective at stopping someone from committing injustice, it is not permissible to kill him.

We might dispute how to fill in all the details, but most people (except extreme pacifists) accept this broad outline.

My thesis in this paper is that whatever conditions render it permissible to kill a fellow civilian in self-defense or defense of others also render it permissible to kill a democratic government agent, even an agent acting ex officio. So, for instance, if it is permissible to kill a private civilian who tries to lock you in his basement because he doesn’t want you to smoke pot, it will be permissible to kill a cop who tries to lock you up for smoking pot. If it is permissible to kill Cobra Commander when he tries to launch a nuclear weapon against Russia, it will be permissible to kill a general or president of a democratic country doing the same.

In contrast, many people seem to subscribe to what I call the “Special Immunity Thesis”. This thesis holds that there is a set of (possibly disjunctive) conditions P under which it is permissible for one civilian to kill another civilian in self-defense or in defense of others. However, P is not sufficient to render it permissible to kill the agents of democratic governments acting ex officio.

So, for instance, someone who accepts the Special Immunity Thesis might hold that you may kill a civilian who starts shooting at kids in a park, but you can’t kill a police officer who is shooting at kids in a minivan. Or you might be able to kill Cobra Commander to stop him from ordering the invasion of a South Asian country, but you can’t kill generals or congresspeople to stop them from ordering the exact same invasion.

The person who accepts the Special Immunity Thesis might believe that it, in self-defense or defense of others, it can be permissible to kill democratic government agents acting ex officio, but holds that the conditions under which it is permissible to do so are different and more stringent than the conditions under which one may kill civilians in self-defense or defense of others.

Note that I am limiting myself here to discussions of killing in self-defense or defense of others against rather immediate threats. I am not discussing whether anyone might deserve to die as an end in itself, nor am I here discussing whether non-violent versus violent resistance is a morally superior or more effective method of social change or of overturning unjust laws. So, the question here is not “Should we go around killing police officers to try to get marijuana decriminalized?” but rather, “If a cop tries to arrest me for pot possession, and I know that allowing myself to be arrested will almost certainly mean I spend 2 decades in jail, can I fight back?”

So, here’s the question that makes this post a bleg. What, if anything, might justify the Special Immunity Thesis? Is there an in-principle difference between government and civilian wrongdoers, such the the former are protected from defensive killing even when the latter would be liable to be killed?

In the current draft, I start by consider the question of whether democratic states have legitimacy and authority. I explain how “legitimacy” (the permission right to coerce) is irrelevant to the question at hand, and how “authority” (the duty to obey and defer) probably doesn’t exist. (I add that even if it did, even if there were a general duty to obey the law and to defer to state officials, there’s little reason to think this duty extends so far as to forbid killing wrongdoing government agents in cases where killing would be justified, if only the wrongdoer were a civilian.) I then consider questions about peaceful alternatives to violence, principles that forbid vigilantism, questions about government agents following orders, issues about government agents acting in good faith, and issues about whether governments  might retaliate or use extortion in response to violent self-defense or defense of others. None of these purported reasons to accept the Special Immunity Thesis succeed, I argue. Instead, if they are reasons to forbid killing in self-defense or defense of others, they apply equally well to civilian cases. These arguments allow that the conditions under which one can kill government agents and civilians are the same, but at most show that they are statistically less likely to obtain when government agents are involved than when civilians are involved. So, for example, if one is more likely to have peaceful ways of resolving governmental threats, then it is less likely that it will be necessary to kill government agents. But the necessity proviso for killing governmental agents and killing civilians is the same.

I don’t really commit myself to saying that killing is justified in any particular case. I’m not trying to settle all the hard cases or provide a complete theory of defensive killing. Rather, I just argue that the ethics of killing civilians and killing government agents in the same. The legality or legal status of government agents’ actions don’t make any difference.