Are Interpersonal Comparisons of Utility Indeterminate?

On the orthodox view in economics, interpersonal comparisons of utilityare not empirically meaningful, and ``hence'' impossible. To reassess this view, this paper draws onthe parallels between the problem of interpersonal comparisons of utility and the problem of translation of linguisticmeaning, as explored by Quine. I discuss several cases of what the empirical evidence for interpersonal comparisonsof utility might be and show that, even on the strongest of these, interpersonal comparisons are empiricallyunderdetermined and, if we also deny any appropriate truth of the matter, indeterminate. However, the underdeterminationcan be broken non-arbitrarily (though not purely empirically) if (i) we assign normative significance to certainstates of affairs or (ii) we posit a fixed connection between certain empirically observable proxies and utility.I conclude that, even if interpersonal comparisons are not empirically meaningful, they are not in principle impossible.