Should Rawls be Banished?

John Rawls seems to be showing up in discussions quite a bit lately:

Liberals' misplaced love of John Rawls, by Linda Hirshman, TNR: The year 2006 marked the thirty-fifth anniversary of the publication of the Bible of twentieth-century liberalism, John Rawls's A Theory of Justice. ... In over 500 densely argued pages, Rawls claimed that politics had to be conceived in fundamentally moral terms." ...

Rawls's appeal is that he created a justification for the liberal state that did not require a lot of apparatus. No appeals to history, no metaphysics about how people differ from animals, no lists of virtuous political values to constrain the process. Just close your eyes, Rawls said, and think of what kind of political society you would make if you didn't know who you were. Black, white, male, female, smart, dumb--you might be anyone who would then have to live in the society you imagined. Rawls said if you did this, you'd produce unlimited free speech and moderately redistributive capitalism. ...

Perversely, Rawlsian liberalism also produced a slippery slope into its opposite, complete selfishness. After all, unless you could achieve the degree of selflessness he required, there was no other place to stop. John Gray, the wandering Brit of contemporary philosophy, correctly called Rawls's hegemony "the legal disestablishment of morality." The game that Rawls set in motion, designed to eliminate common preexisting political values, could also produce the result that everybody simply advocated for himself.

It is not a coincidence that the only successful two-term Democratic presidency of the Age of Rawls was engineered in part for Bill Clinton by Bill Galston, a political theorist with a background in classical thought. Although Galston pays due homage to Rawls, his crucial work is ends-driven, not justified on the blindness of the procedure... Rawls's work--the best effort to take a tradition grounded in the bourgeois revolutions of the seventeenth and eighteenth centuries--and make it relevant to a modern, industrial state simply left the country to the conservatives.

As intellectuals have struggled to get the basic Rawlsian framework to work in the real world, none have made the argument for the metaphysical assumptions that must be a part of their political prescriptions. Even now, liberal thinkers like Paul Starr and Michael Tomasky, who are trying to generate richer visions for liberalism, cannot completely free themselves from Rawls's legacy.

This failure is but the sorry hangover of the years of Rawls. Make a public, political argument for classical political virtues of courage, philanthropy, and temperance, and armies of philosophy professors and various amateurs emerge from the blogosphere to remind you that Aristotle's metaphysics supported slavery. What is then left? But such insistence on purity is a victory only for those for those who would rather the Right be President. It is time for the thinkers of the Democratic revival to leave the senior common room.

I don't know as much about Rawls as I should - it's not something I rely upon when I think about government intervention - so for a bit more on Rawls I'll turn it over to Brad DeLong. He wrote this in response to a recent op-ed by Greg Mankiw that mentions Rawls as a basis for redistribution, though I've cut the part that responds directly to Greg:

Let's Not Tell Hilzoy!, by Brad DeLong: ...Rawls does not think that the primary goal of public policy should be to redistribute resources to help those at the very bottom. Rawls thinks that the first goal of public policy is to maximize liberty for all. He thinks that the second goal of public policy is to make everybody better off. Redistribution plays third fiddle in Rawls's orchestra: it is a constraint on social wealth maximization--things that make people better off must be shared: choose that set of social and economic arrangements that makes everybody better off, but don't choose a set of social and economic arrangement that makes some people better off at the price of making the worst-off even worse off.

Here is what I think is the best way to think of this third point, of Rawls's "Difference Principle":

A group of people are sitting around the campfire, after a hard day's worth of work and pay in which what jobs people did and how hard they worked and how they were rewarded was determined by some complicated and not very transparent process.

Looking around, the person who is worst off says: "Hey! Wait a minute! This isn't fair. Everybody else is better off than I am."

And one of the others replies: "I'm sorry. You do get less than everybody else. But we set things up in the best way we could. Given the constraints imposed by human psychology and the natural world, we couldn't have set things up in any way so that you would have been better off."

"Oh. That's OK then."

According to Rawls, an arrangement that passes this test is "just" and "fair."

Now I don't think that Rawls has it correct: I don't think that socioeconomic arrangements that pass Rawls's test are necessarily just, and I don't think socioeconomic arrangements that are just necessarily pass Rawls's test. There are too many lexicographic orderings wandering around Rawls's setup for any economist to be happy with it...