The Difference Principle is not so much excessively risk-averse as excessively jerry-rigged. OK, we can’t aggregate as utilitarians but then we resort to some notion of primary goods with intersubjective validity. OK, the size of the worst-off group is itself endogenous to the contractarian process. But just how big is that group supposed to be? Can it be 99 percent of society? OK, people behind the veil don’t know their particular identities, but just how "thin" is their knowledge supposed to be? And must their choices be purely self-interested? All these criticisms are well-known. You might try to shore up Rawls on any one of these points but the entire apparatus is simply too wobbly.
The bottom line is that you can’t get lexical orderings out of a moral theory unless you build them in upfront. And without lexical orderings, well, Rawls, like many illustrious minds before him, does not succeed in
sidestepping the dirty mess of aggregation. The critical moral question is how we should compare the interests of some people to others in a real world setting; don’t expect to find an easy way out of that one.
Rawls’s Principle of Equal Liberty is if anything on weaker ground than the Difference Principle. Equal Liberty? Who says? At what margin? At what cost? Lexicality can’t plug all the leaks in this shaky boat, and no it can’t save Robert Nozick either.
The biggest problem is simply why the imaginary agreement behind the veil of ignorance should have moral force. Now I like preferences as much as the next guy, but imaginary preferences take me only so far. That is just one piece of information in a much broader comparison of plural values. I’m not even sure that imaginary preferences should override the very real preferences of very real people in very particular situations. Why should they? "Fairness" is just one value of many.
I read Rawls as a very very smart and intellectually honest guy, determined to resurrect Kant, avoid the aggregative problems of consequentialism, and move at least one step beyond Sidgwick. He knew how hard it was to even attempt such a success and he makes all the requisite moves to get us there, albeit without, in the final analysis, squaring the circle.
Matt Yglesias adds commentary; he notes, correctly, that for the current Left Rawls doesn’t offer such an inspiring vision. I’ll put it this way: if you have to work that hard to establish "Sweden is great," you should be spending more money on plane tickets.
Just to clarify, there are at least three Rawls doctrines: "Justice as Fairness," TJ, and Political Liberalism. I like the first one best, but won’t cast my lot with any of the three. At the end of the day I come away thinking that it is Sidgwick (and
maybe Kierkegaard?) who is the central moral theorist of the last two