Abstract
In his discussion of pleasure, Aristotle assumes the thesis that a perfect activity always and necessarily yields pleasure. The occurrence of pleasure is even presented as a sign that the activity is perfect. But this assumption seems to be too easy. It is possible that we do feel pleasure in activities which are not perfectly performed, and on the other hand, it is not certain at all that I will enjoy a perfect activity. Pleasure falls into the category of what J. Elster has called 'states that are essentially by-products'. Up to a point, Aristotle acknowledges this, but he does not follow this analysis to its final consequences. If one agrees, as Aristotle does, that there is a difference between the perfect activity and pleasure, it should be possible that an activity is perfect without yielding pleasure, or that pleasure will accompany even an activity which is not perfect