What kind of paradoxes might result from applying the maximin principle in a scheme of distributive justice?

Question

I quote from John Harsanyi "Can the Maximin Principle Serve as a Basis for Morality? A Critique of John Rawls's Theory" (1975) (My italics).

But then came a growing realization that the maximin principle and all its relatives lead to serious paradoxes because they often suggest wholly unacceptable practical decisions. The other-Bayesian-school of thought, which is now dominant, proposes expected-utility maxi-mization as decision rule under uncertainty.

Additional Note: Harsanyi then adumbrates a scenario in which a man could fly from NY to Chicago to take up a better paying job than that which he already has. The worst-case scenario is that he dies flying to Chicago (forgive me if I am wording this inaccurately) and so the maximin principle demands this course of action be avoided, resulting in seemingly irrational action. Is this a fair analogous example of the maximin principle with which to critique Rawls?

Answer 1

I'm not sure if the example is conclusive, but it sounds like Harasanyi is highlighting a general problem with maximization/minimization theories. These problems generally apply to utilitarian and consequentialist theories which suffer from two initial weaknesses:

(1) The problem of perceived benefits versus realized benefits in the calculation. This seems to be the highlighted issues in your description of the quote. This type of problem is one where we need to know whether we evaluate based on outcomes or good-faith predictions. Thus, the sample case -- do we need to recalculate based on the agent dying while traveling to that mutatis mutandis better job or not? Or more softly, do we need to factor that into the calculation? Or to put it another way, is the morality of our actions contingent on the future? The more general problem this creates is that the very fact we ask the question makes it problematic to view maximization or minimization as the core of morality (this last piece is Bernard Williams' objection).

(2) The second problem for calculation-based strategies is that they seem to leave no action unprohibited if it is a net benefit to the goal. Most compensate by having a harm principle, but then we return to Williams' objection -- if we need a second non-maximization / minimization principle to check the calculation, then surely the calculation does not capture all of morality.

Having not read the text in question, I'm a little limited in what I'm willing to say on it specifically but I would guess he's suggesting that insofar as Rawls is about constructing this sort of ideal society behind the veil of ignorance that Rawls' theory is sufficiently calculative to succumb to the first problem.