UPDATE, June 26, 2009: Anyone interested in how I ended up developing this argument should check out the paper just published in PPR.
One of the more difficult issues for Kantian moral theorists is how, if at all, our moral obligations should be sensitive to others' wrongdoing. It seems fairly obvious that what we are morally required to do can change in response to others' immoral conduct. A clear example is promise keeping: If A and B agree to a mutually beneficial promise, but A doesn't fulfill the terms of their promise, B is presumably not obligated to fulfill them either. So A's wrongdoing influences B's moral obligations. Another example is punishment: Since punishment is the infliction of harm, suffering, or deprivation (which is typically wrong), it must be the case that the wrongdoer's wrongdoing justifies inflicting otherwise wrongful harm, suffering, or deprivation on her. This issue is acute for Kantians because Kantianism has long been seen as somehow more "principled" than consequentialism. The challenge for Kantians is to offer an explanation of how our moral obligations should be sensitive to others' wrongdoing that invokes key Kantian values or principles (rational autonomy, the categorical imperative, etc.) without becoming so sensitive to others' wrongdoing that Kantianism becomes indistinguishable from outright consequentialism.
The example that has of course stimulated much of the discussion surrounding this problem is Kant's treatment of 'the murderer at the door' in his essay "On a Supposed Right to Lie." There Kant seems to say that lying to a would-be murderer about the whereabouts of the innocent victim he intends to kill would be morally wrong. Most Kantians (and most reasonable people in general) find this conclusion troubling if not absurd: If ever there were a situation in which lying is not only morally permitted, but even morally required, that would be it!
What follows is my own (admittedly long-ish) attempt to answer the 'murderer at the door' problem in Kantian terms. Whether my attempt is of value in addressing the larger theoretical problem of how our moral obligations should be sensitive to others' wrongdoing, I'm not sure, but here goes:
My argument is actually quite simple:
1. It is morally permissible (from a Kantian perspective) for an innocent person (i.e., a person who does
not deserve to die for some other reason) to lie to a murderer in self-defense.
2. If it is morally permissible (from a Kantian perspective) for an innocent person to lie to a murderer in
self-defense, then it is also morally permissible (from a Kantian perspective) for any person to lie to a
murderer in the defense of another innocent person.
It is morally permissible for any person to lie to a murderer in the defense of another innocent person.
Obviously, my task is to defend the truth of the premises.
P1. One of the great strengths of Kantian moral theory is its powerful account of what's wrong with manipulative or coercive behavior such as torture, promise breaking, and lying. Such acts in effect treat others as mere means to our ends, inasmuch as we manipulate the other person's behavior or expectations in order to advance our own happiness. In the particular case of lying, the manipulation or coercion involved is doxastic, as Sissela Bok has observed: A lie attempts to cause another person to act on a belief that the liar believes to be false so as to benefit the liar.
But I would suggest that in the 'murderer at the door' example, two conditions necessary for lying to be wrongful (in Kantian terms) are not present: First, the lie is not meant to advance the happiness either of the liar or of the potential murder victim, but to thwart the abuse of the victim's autonomy that her murder would represent. Hence, if lying to the murderer is manipulation at all, it is manipulation in the service of the would-be victim's autonomy, a central Kantian value. Second, while Kantian ethics prioritizes the value of autonomous rational agency over happiness, it does not follow from this that we are obligated to honor another agent's autonomous choices no matter the ends that a given exercise of autonomy is meant to serve. In the 'murderer at the door', the murderer intends to exercise his autonomy in the service of a morally impermissible end, and Kant claims that we are obligated to promote others' ends (i.e., others' happiness) only if those ends are themselves morally permissible. I take that to be an indirect statement of the notion that whether we must honor another's autonomy on a given occasion may depend on whether that autonomy is being exercised in morally permissible ways. Autonomy may be the highest moral value, but it is not therefore an unconditioned value. To manipulate by lying is to deceive in order to thwart another's otherwise permissible ends. In the murderer at the door example, this is not what we are considering doing.
It would seem to follow that if I am innocent (i.e., I do not deserve to die for other reasons), my lying in self-defense is not an impermissible violation of the liar's rational autonomy.
2. But would it be permissible for me to lie to the murderer in order to save another's life? To see why it would be, I'm going to introduce what I will call the Kantian Symmetry Thesis. An 'agent' is the person who actually performs the act, and a 'beneficiary' is the person whose autonomy is protected or advanced by the agent's act.
Kantian Symmetry Thesis: Any morally permissible act performed by agent A in which A is also the act's beneficiary is also permissible if another agent B (relevantly similar to A) is the beneficiary of A's act instead.
We see throughout Kant's casuistry implicit appeals to this thesis: Suicide, Kant thought, is wrong for just the same reasons (and in just the same circumstances) that homicide is wrong. Similarly, in Kant's sexual ethics (not that we should accept much of it of course!) acts that treat another's sexuality as a mere tool of one's happiness (rape) are wrong in the same way that acts that treat one's own sexuality as a tool of one's own happiness (masturbation) are wrong. In other words, Kant did not think that there exists a special moral relationship to oneself such that the obligations one bears vis-a-vis oneself are importantly distinct from those one bears toward other agents. (This is a way in which Kant was not a 'liberal,' since one important feature of liberalism has been the idea of a domain of self-regarding behavior that is governed by different norms from the domain of other-regarding behavior).
So, if it is permissible for me to lie to prevent a murderer from killing innocent me, than by the KST, it would follow that it would also be permissible for me to lie to prevent that same murderer from killing another innocent.
Again, I don't know if my thoughts about how to address the murderer at the door hold any general lessons about how to develop a non-ideal Kantian theory (one that takes account of others' wrongdoing), but any ideas or suggestions are certainly welcome.