Two-Level Utilitarianism
RM Hare’s two-level utilitarianism is an attempt to accommodate deontological intuitions within the framework of utilitarianism.
We all feel instictively that there are some acts that are just wrong, even if they do maximise utility, that there are some means that no end could ever justify. This is difficult to account for on utilitarianism, which holds that morality is all about achieving good ends, and so that no act is intrinsically right or wrong.
Hare’s two-level utilitarianism attempts to account for this feeling from a utilitarian perspective. Two-level utilitarianism holds that act utilitarianism is true, but that we ought to operate practically as if rule utilitarianism were true, because that is the approach that maximises utility.
In an ideal situation, Hare holds, we ought to act as act utilitarians. Given sufficient time to calculate the consequences of each of the various courses of action open to us, and sufficient foresight to do so accurately, we ought to perform whatever act it is that maximises utility. We do not, however, have either that much time or that much foresight. Utilitarianism must therefore be adapted to the circumstances of the real world.
The approach to ethical decisions that will serve us best in practice is not act utilitarianism, but rule utilitarianism. Attempting to perform utilitarian calculations that we just don’t have time to do won’t maximise utility. What will maximise utility is having a set of rules that generally tell us what to do without too much fuss. We therefore ought, because of our limitations, to act in accordance with rules.
Further, it’s good for us to have a strong attachment to these rules. The rules are generally reliable, and it is therefore good for us to be very reluctant to break them. This is where our deontological intuitions come in: they help us to maximise utility.
In this way, then, Hare combines act utilitarianism with rule utilitarianism, and explains our attachment to the idea that certain acts are intrinsically wrong.
|
