Nov 10, 2010

Double Effect for Consequentialists?

The principle of double effect is typically described as part of deontological, rather than consequentialist, ethical theories. Consider the classic trolley case, for example, where one encounters a trolley hurtling down a track toward a group of five people whom it will certainly kill; diverting the trolley down a side track would result in the trolley striking just one person who is standing on the tracks. For consequentialists, it seems the case looks like this: one can do nothing, in which case five people die, or one can divert the trolley, in which case one person dies. It is natural to think that this case has the same profile of consequences as a case in which one can kill one person (and, for instance, harvest his organs) to save five: in each of these cases, one can do nothing, in which case five die, or intervene, in which case one dies.

Continuing this line of thought, it looks as though consequentialists have no reason to care about double effect, or at least no reason that derives from the content of their moral theory. The pairs of cases to which double effect has been thought to apply, like terror bombing and strategic bombing, seem identical with respect to consequences by stipulation. This means that there is no distinction between them on consequentialist grounds, and the consequentialist can therefore ignore any differences between them.

This reasoning is suspect, however. We are asked to move from (1) to (2):
(1) two cases are identical with respect to their ultimate consequences
(2) the cases are identical with respect to all their consequences
And if (2) is true, then consequentialists can ignore the difference between the cases.  I think we should resist the move from (1) to (2).

When you choose to harvest the organs of the one to save the five, you have not just actualized a state of affairs in which one person dies and five live. In addition, you actualize a state of affairs in which you kill the one in order to save the five. Prima facie, these are not identical states of affairs, and it would seem open to a consequentialist to distinguish this pair on the ground that there is disvalue in a state of affairs in which an agent kills in order to save (in the organ harvest or bridge case) relative to a state of affairs in which an agent kills as a proportional side effect of saving (in the classic trolley case). If consequentialists think that states of affairs have value, and that this value determines what it is morally right to do, then perhaps they should take some interest in states of affairs that include the relation of moral agents to the outcomes they bring about, or at least give an account of why they do not. And, once these sorts of states of affairs are taken into account, the trolley case can be distinguished from the organ transplant case on purely consequentialist grounds.

The consequentialist will have to account for the value disparity due to the way agents relate to the outcomes they bring about, but it is at least open to the consequentialist, I have argued, to suppose that something like the principle of double effect figures in the correct moral theory. Peter Singer has argued that proponents of double effect may need to rely on some form of consequentialism to cash out the proportionality constraint [1]; it turns out that consequentialists may need to rely on some form of double effect if they want to take seriously the value of states of affairs that include agents’ relations to outcomes.

There are some important subtleties here that I'm overlooking for now, and will discuss in a follow-up post. See here for my follow-up.

[1]“Saving a life is a proportionate reason for bringing about the death of a fetus; saving a trim figure is not. Hence the doctrine of double effect has, after all, a consequentialist element…”
Peter Singer, “Do Consequences Count? Rethinking the Doctrine of Double Effect,” The Hastings
Center Report, Vol. 10, No. 2 (Apr., 1980): 29.

No comments:

Post a Comment