Psychological Egoism

Psychological egoism is the view that people are always selfish. When was the last time you did a good deed? Did you do it for its own sake, or for your own? The egoist says that all of us are necessarily self-regarding. I shall argue that this view is incorrect.
First we should ask, what kind of claim is this? Is it an a priori claim, or a generalization from experience? If it were the latter, we could never conclusively prove it: we could never show that necessarily all actions are selfish. So it must be a priori. But no a priori claim could be substantive: a priori truths are all analytic (that is, the predicate is contained in the subject). So if this claim were analytic, it would become trivial. (It is worth noting that Kripke's claim that there are a posteriori necessary truths does not show that a priori truths are not analytic.)
The situation is paralleled by pseudo-sciences such as Freudian psychoanalysis. As Karl Popper has argued, any theory can be maintained so long as it is drained of empirical content. Like psychoanalysis, psychological egoism makes no genuine claims and can never be refuted. But it purchases certainty at the price of becoming vacuous. I shall have more to say on this below.
The simplest way to see the egoist's mistake is to distinguish between the side-effects of an action and the reason for which it was done. Suppose we grant that in doing a good deed, we usually get a pleasant feeling (though I suspect this is false). Even so, this would not show that that feeling was indeed the motivating factor in our behavior; it could simply be a side effect of doing the good deed. Compare the case in which someone types for hours on a philosophy paper and gets a cramp in his hand. Did I type in order to get the cramp? Of course not. To be sure, it was a foreseeable result of my typing, but it was not the motive for my heroic efforts.
Perhaps we can go further in refuting the egoist's claims. Phil Washburn presents the opposing view in his book Philosophical Dilemmas. This is called psychological altruism. The altruist points to phenomena such as love that seem to show that people can be genuinely other-regarding.
What will the egoist say? The egoist has to admit that there seem to be acts of selflessness, such as the soldier who jumps on a grenade to save his comrades. Here Washburn's egoist appeals to re-interpretation: there is always a competing story the egoist can give that makes the act turn out to be selfish. Perhaps the soldier wanted to avoid the pain of living on as a coward, or wanted to become a hero and bring glory to himself and his family.
Yes, perhaps. But it's not enough simply to present such a story: one must also give some reason to suppose that it is true in the case at hand. Since egoism is an a priori claim, as we have seen, it supposes that in every case such a story will be not just available but justified by the evidence. But this is totally implausible. The egoist's `just-so' stories are just so much hot air.
We need not leave matters like this, however. For we can grant the egoist his ability to reinterpret all acts as selfish. What, then, becomes of the claim that we are always selfish? It has thereby become immune to empirical refutation or empirical tests of any kind. But as Popper has shown, this only drains egoism of content.