Michael Sandel has recently argued that at the heart of contemporary pop-

ulist uprisings lies the belief in meritocracy, i.e., the idea that high positions

in business and society should be allocated according to talent and effort.

Meritocratic thinking, he claims, leads to hubris among the winners (as well

as humiliation among the losers). To curb the hubris, he suggests that we

should allocate high positions in society on the basis of lotteries. If no one

can claim that their position in life is their own doing, this will make people

less arrogant and more open to considerations of the common good.

Such measures, however, would quite radically impact both the way

in which we allocate positions and jobs as well as the way in which we relate

to our positions and jobs. Before we opt for them, we should therefore

think about possible trade-offs and side effects. In this paper, I aim to out-

line four dimensions that should be taken into account: efficiency, (moral)

psychology, justification and symbolism. With regards to all four dimensions,

or so I will argue, there are possible trade-offs to be considered. It is therefore

less than clear that we should opt for Sandel's solution.

Suppose there are irreducibly collective interests. Suppose, that is, that at least some groups have interests that are irreducible to the interests of individuals. What might these collective interests consist in? In this paper, I examine three accounts of group interests, applying the Parfitian trichotomy of hedonism, subjectivism, and objective list theory to groups. To start, I briefly argue that group agents do indeed have irreducible interests. Then I defend the following three claims. First, hedonism fails as an account of collective interests. Second, subjectivism-for-group-agents (though not for non-agent groups) is a reasonable option but faces the same problems as subjectivism-for-individuals. Hence, one’s evaluation of the subjectivist account of collective interests should depend on what one thinks about the familiar debates about subjectivism-for-individuals. Third, objective list theory initially appears difficult to apply to groups, but a type of perfectionism-for-group-agents warrants further examination. I conclude with an indirect argument for irreducibly collective interests.