What is Rawls’s veil of ignorance actually supposed to show?

Question

I teach economics students, and when I introduce Rawls they treat the original position as a risk-aversion exercise. Behind the veil you don't know whether you'll be rich or poor, so you pick the society where the worst-off do best: maximin.
Then someone always cites John Harsanyi, who argued that rational people behind a veil would maximise average utility instead, since they would have an equal chance of being anyone.
So is Rawls's argument basically a claim about risk aversion? If so, isn't Harsanyi simply right on decision-theoretic grounds? Or is the veil meant to do something else?

Deepa Nair · Answer

The veil isn't primarily a model of individual choice under risk. It's a device for representing fairness: it removes the information that would let people tailor principles to their own advantage. A Theory of Justice (1971) gives several reasons for maximin in that special situation: no basis for probabilities, outcomes below the minimum would be unacceptable, and the "strains of commitment" — you must be able to live with the result whatever position you end up in.

Tomasz Wójcik · Answer

From the decision-theory side, Harsanyi's argument assumes you should assign equal probability to being anyone. Rawls deliberately denies the parties that kind of information. So they disagree about the setup, not just the rule.

Javier Morales · Answer

That's helpful, Deepa. So my students are right that maximin looks like extreme risk aversion, but wrong that Rawls is relying on it. The reasons come from the situation, not the parties' psychology.

Hamza Qureshi · Answer

In court I find the strains of commitment argument the most practical one. A constitution that leaves some group with nothing to lose won't be stable, however good the average figures look. Rawls is partly making an argument about legitimacy.

4 answers