Response from Thomas Pogge on February 4, 2006

Jeremy Bentham comes to mind as someone who believed that people always seek their own happiness (pleasure minus pain). Given the great diversity of human conduct across epochs and cultures, it is easy to find plausible counter-examples. But such counter-examples can always be rebuffed by a remark such as the one you have in parentheses: The act appears to be selfless, but was really motivated by the satisfaction the agent expected to derive from assuaging his guilt or from the approval by others. Even when a person throws himself on a hand granade to save his comrades, one can say that the expected satisfaction from the anticipated approval of his comrades must have outweighed his dread of the anticipated pain and death.

The problem with such rebuffs is that they lead to circularity: The fact that a person acted in a certain way is taken as sufficient to show that he must have had some selfish motive for doing so. And the proposition in question (that all human actions are born of self-interest) then becomes entirely immune to refutation. You repeat it no matter what actions are performed. And it no longer expresses then a (disturbing) discovery about the particular world we live in.

To demonstrate this, take another proposition: People always do what they prefer. This proposition is harmless and trivial when preference and choice are conceptually linked so that an agent’s preferred conduct option is, by definition, the one she chooses. Or consider the proposition: Every person is directly motivated only by desires that are her own. This again is not a psychological discovery but a harmless conceptual truth: Another’s desire can motivate me only indirectly, when I know about it and desire that it be fulfilled. To explore your own understanding of the proposition in question (that all human actions are born of self-interest), ask yourself what you would count as good evidence for a genuinely altruistic act. If you find, on reflection, that you would count no conceivable conduct as such good evidence, then your proposition is independent of what human beings do and are like, and hence not informative about them. If, on the other hand, you find that you recognize the possibility of conduct that you would recognize as good evidence for non-selfish motives, then you are likely to find actual such conduct in the world — though you may also conclude, as Immanuel Kant did, that it is never certain that a person’s conduct is not selfishly motivated.

For some interesting empirical work on the extent to which human conduct may be altruistically motivated, see C. Daniel Batson: The Altruism Question: Toward a Social-Psychological Answer .