ABSTRACT
In order to maximize expected utility you need to compute that utility and compare it with that of alternative options. These problems are avoided artificially in the example of the charitable trust because, for the sake of simplicity, we are assuming that more money relieves more human misery. But in many cases such a computation is impossible to perform, particularly as different sorts of utility are incommensurable: how do you compare the utility of reading a good short story with the pleasure of a backrub, for example? By satisficing in the manner described, you avoid these problems about computation and incommensurability. So it is not only in the unusual case of the charitable trust that satisficing has
its merits. It is arguably not just an ad hoc solution to the present paradox.
