He starts the paper saying that “selfish individuals cohabit with other-regarding ones”. He is talking about two kinds of people.

He demonstrates a difference between men and women which is, I think, well known in game theory. Now this difference is a different way of thinking and we might expect that, like saying men are taller than women, there will be plenty of exceptions. So we are dealing with two (or more) ways of thinking and not with sex. Tallness is easy to identify but what is the character of this difference?

A clue is on page 14 (I have only skim read it) which reverts back to the opening remark. He says:

The higher the degree of agreement with the statement “If a person is poor, this is often the result of a lack of effort on their part” the lower the percentage of the sum that subjects leave to the receiver in the dictator game.

So there are two kinds of people. The terms selfish and other-regarding are scientifically unsatisfactory as one is pejorative. We could call them hard and soft: those who agree with that statement about the poor, and those who disagree. Back on page 4 he identifies two kinds of altruism, the unconditional and the conditional which I would posit characterise the soft and hard people respectively.

Because I do not have time to read it in detail I don’t know to what extent he follows through with these two kinds but what I do think is game experimenters should be testing people with statements like the one about whose fault it is if you are poor and showing that it is not really a division between men and women; it is a division between hard and soft (and maybe more categories).

Yes, altruism and selfishness will be affected by the particular institutional arrangement but they will also be affected by the nature of the person, a nature that, like tallness, it would be possible to measure. I’d suggest, though, that unlike tallness, it does not conform to a normal curve but is a double bulge, ie genuinely two kinds of mentality.