tion concerning right and wrong in a way, which is, indeed, comparatively unimportant, but which yet deserves to be noticed.
It may have been observed that our theory does not assert that a voluntary action is right only where it causes more pleasure than any action which the agent could have done instead. It confines itself to asserting that, in order to be right, such an action must cause at least as much pleasure as any which the agent could have done instead. And it confines itself in this way for the following reason. It is obviously possible, theoretically at least, that, among the alternatives open to an agent at a given moment, there may be two or more which would produce precisely equal amounts of pleasure, while all of them produced more than any of the other possible alternatives; and in such cases, our theory would say, any one of these actions would be perfectly right. It recognises, therefore, that there may be cases in which no single one of the actions open to the agent can be distinguished as the right one to do: that in many cases, on the contrary, several different actions may all be equally right; or, in other
