I love Rawls more than anyone else on this blog; perhaps a lot more. But sometimes you have to take out the trash and save the baby from the bathwater. So in this post, I’m going to list the Rawlsian positions and arguments I think Rawlsians (and libertarians) should jettison. My goal is to take the next step in explaining why libertarians should take Rawls seriously, following my explanation of Rawls’s project. By separating the bad from the good, I hope to show you a Rawls worth learning from.

Unlike many of my posts, I’m not going to make one sustained argument. I’m just going to give a laundry list of criticisms of quickly summarized Rawlsian arguments. We can talk about the details in the comments.

(1) The Veil Is Too Thick: the veil of ignorance deprives parties to the original position of too much information. I much prefer David Gauthier’s veil, where we know everything about people in our society, but we don’t know who we are. Rawls wanted a thick veil to ensure impartiality, but impartiality doesn’t require selecting principles of justice apart from facts about, say, the talents individuals have. This criticism is standard and much explored by many.

(2) Limited Agreement Behind the Veil/Indeterminacy: Unless we assume, as Rawls does, that parties to the original position are identical in every respect, there are considerations that will lead even these parties to disagree. This generates a lot of indeterminacy early on in the Rawlsian decision procedure. On this, see this.

(3) Principles Aren’t the Object of Justification: I think general principles are too vague to be useful action-guiding norms. How could we hope to determine, say, whether a society complies with the Difference Principle? As I argued in my POD series, it’s rather difficult to do. So it’s better to try to justify rules at a finer grain. It’s easier to determine whether rules are justified because it is easier to measure deviations from them and figure out how to bring social practices into alignment with them.

(4) Constitutional Essentials Aren’t the Object of Public Justification: In the same vein, where Rawls only thinks we have to publicly justify constitutional essentials, I think our first goal should be to publicly justified laws and judge constitutional essentials in terms of how reliably they promote public justification. I draw on related arguments from Jerry Gaus and Jon Quong on this point in their recent books.

(5) Natural Talents Aren’t Collective Assets: individuals can make fair distributive claims based on how they exercise their natural talents. It’s hard to see how such collectivism could be justified in the first place. It seems much more natural to assume that natural talents belong to individuals, and that principles of justice that allow coercing people to use their natural talents to serve others are prima facie suspect. That’s the grain of truth in the self-ownership thesis.

(6) The Political Liberties Are Less Important: Rawls argues that a regime can only be legitimate if the political liberties are guaranteed their fair value. This means that Rawls thinks political liberties are among our most important. But that’s not true. Political liberties aren’t that useful for familiar political science reasons and they’re certainly not more important the civic liberties that allow people to pursue their projects, be loyal to their religious principles, and so on. See J Brennan on this here.

(7) Economic Liberties Are More Important: Rawls only makes freedom of occupation and right to own personal property basic status, along with the civil and political liberties. But there’s just no good justification for excluding other economic liberties, like the right to own private capital. If we’re going to do the whole Rawls ’71 thing, we should learn from John Tomasi (see his posts at BHL and his book here). (Note: giving more economic liberties basic status does not mean they have maximal reach and weight, as they do on some libertarian views.)

(8) The Difference Principle is Wrong Because Maximin Sucks: maximin as the principle of choice in cases of uncertainty is unsound. In the end, it’s too risk averse. So the argument for the DP fails insofar as it is based on maximin. I still like Harsanyi’s early criticism here.

(9) The DP Is Wrong Because Reciprocity Arguments Assume (5): the implausibility of maximin has led many Rawlsians (including Rawls) to offer a reciprocity-based argument for the DP. Basically, the DP is supposed to express the idea the people don’t deserve their natural endowments, and so can’t fairly insist to benefit from them while denying others the right to do the same from them. But (5).

(10) The DP Is Wrong Because It Is Reasonably Rejectable: as Rawls admitted in Political Liberalism and The Law of Peoples, reasonable people can reject the Difference Principle (as I discussed in my last post). Rawls doesn’t think reasonable rejectability means that the DP is not part of the most reasonable conception of justice, but surely there are less problematic conceptions of justice that are more reasonable, such as ones with a sufficientarian distributive principle, which are far more widely endorsed (see David Miller on this).

(11) Full Compliance Is Too Demanding: Rawls is a proponent of ideal theory. Principles of justice can count as valid if they can self-stabilize under conditions of full compliance. In other words, citizens of a well ordered society are presumed not to deviate from complying with coercive institutions based on their self-interest or even their own ideology or principles when they require deviating justice as fairness. Citizens and elected officials, for instance, will follow the law even at serious cost to themselves. But this is not a plausible psychological model on which to base selection of principles of justice. Rawls’s non-ideal critics are right (Sen and Schmidtz are good on this; and yes, the Sen book sucks in general, but it has a few insights).

(12) No Ideal Regime Types: Once we give up on justifying principles, full compliance and determinacy behind a thick veil, the idea that we can formulate ideal regime types as models of justice is implausible. The best we can do is make a series of much more precise arguments about which laws and policies to support. The case for a system of free institutions will arise from the conjunction of these restricted arguments.

(13) No Property-Owning Democracy: given the importance of the economic liberties, the rejection of ideal regime types, and the rejection of the difference principle, there are no good Rawlsian arguments for POD. Plus, even if the case for the two principles is sound, POD fails to satisfy them. See my POD series.

(14) Public Reason Is Misunderstood: public reason should not be restricted to shared reasoning or prioritizing shared reasoning. There’s no good argument for restricting public discourse in this way (as I have argued in print here and here). I’ve argued with a friend that Rawlsian public reason is also a very poor assurance mechanism.