A few commenters looked at the post below and said, “Where’s the Rawls?” I was just making what I took to be a number of largely conceptual points about the economic patterns that emerge from social interaction — points mostly from Hayek and Nozick. That point is that the principles of social interaction are the primary subject of moral evaluation, not the patterns themselves.
This is Rawls’ view, too. Nozick’s criticisms notwithstanding, Rawls really isn’t concerned with patterns either. He’s worried about the principles of interaction, the terms of association. A just society is a scheme of cooperative mutuality. The basic rules of the game should benefit everyone. A good way to ensure a set of governing principles does benefit everyone is to pay special attention to how the least well-off fare under them. Rawls says: when choosing a set of principles, we should pick ones that leave the least advantaged class as well off as possible. I agree with this. This takes a comparative property of a pattern (the poor do better in this pattern than in alternative patterns) as a constraint on acceptable principles, but says nothing whatsoever about inequality. Indeed, Rawls offers what amounts to a powerful argument against fixating on inequality. Rawls says: fixate on the welfare of the least well off.
It’s also worth emphasizing that Rawls isn’t talking about the income distribution. He’s talking about primary goods. (“Distribution” in his sense is opposed to “allocation” and “distribution” is not about income.) Given a solid Hayekian understanding of the function of sound market institutions over time, it becomes easy to see that rising income inequality can and does often accompany innovation that leads to increasing equality in primary goods. But, again, Rawls isn’t even much concerned with equality in primary goods. He’s interested in maximizing the minimum.
It’s true that the difference principle is stated in terms of “conditions” social and economic inequalities must meet. I think this was a big, terribly confusing mistake, since it does no actual work. You could just ask: Are the poor doing better in this scheme than in alternatives? Now, this won’t satisfy pious Rawls purists, but Rawlsekianism does not try to hide the fact that it is a mongrel creed. For me, the main Rawlsian takeaway is that the mutuality at the heart of justice should lead us to put the welfare of the least-advantaged at the forefront of our deliberation over basic principles of association.