Rawls Made Easy As Easy As I Can Make Him

by hilzoy

As both a background to Sebastian's post on Rawls and a justification of my own views, here's a short version of Rawls' arguments on justice. In what follows, I'll indicate, at some points, which texts I'm drawing on; I'll refer to A Theory of Justice as TJ, and Political Liberalism as PL.

The Basics: The basic question Rawls sets out to answer is (not coincidentally) one I raised in a previous post, namely: given several different ways of structuring the basic institutions of society (or, as I said in my posts, the rules of the game), how do we decide which is just? Note a few points: (1) He is talking about the rules of the game, not ad hoc interventions. (2) He limits his inquiry to justice within a society, and assumes that the society in question exists in conditions of "moderate scarcity", meaning that "natural and other resources are not so abundant that schemes of cooperation become superfluous, nor are conditions so harsh that fruitful ventures must inevitably break down" (TJ 127.)

His answer is, basically: a just system is one that would be chosen by people in what he calls the 'Original Position'. (Trust me: all the off-color jokes involving this phrase have already been made.) In the Original Position, people know all sorts of general facts about the world, including not just science but also economics. However, they do not know any particular facts about themselves: their race, their gender, their specific views about what kind of life they want to lead, their religion and values, and so forth. They do know, however, that they have some view about what sort of life they want to lead; they just don't know which it is.

They are also assumed to be fully rational, and mutually disinterested. 'Mutual disinterest' means that while, for all they know, they may care about all sorts of other people in real life, in the Original Position they are concerned to maximize their own good. Note that the better off I am in real life, the more I can do to help others, so making this assumption in the Original Position enhances my ability to be generous in the real world, if I want.

So the basic idea is: we decide what principles we should use to assess the rules of our society as if we didn't know who we would be, or even which kind of person we would be -- born rich or born poor; black, white, Asian, Native American, or whatever; Catholic, Muslim, atheist, Jain, or an adherent of some other religion; smart or not so smart; male or female; and so on. What we would choose under those conditions determines what rules are just.

According to Rawls, under these conditions people would choose the following principles as criteria by which to assess social institutions:

I. Each person has an equal claim to a fully adequate scheme of basic rights and liberties, which scheme is compatible with the same scheme for all.

II. Social and economic inequalities are to satisfy two conditions: first they are to be attached to positions and offices open to all under conditions of fair equality of opportunity; and second, they are to be to the greatest benefit of the least advantaged members of society.

(These are generally referred to as 'the two principles'.)

So: why does Rawls say these things? I'll consider first his argument for the Original Position, and then his argument for the two principles.

The Original Position: Rawls' argument is part of what's known as the 'Social Contract' tradition, which has two main branches: those involving actual contracts and those involving hypothetical contracts. The first group (basically) says that government was set up by some contract or agreement, which we are bound to follow. It faces questions like: how do you know about that agreement? and: why on earth are we bound by it?

Rawls does not make an 'actual contract' argument. This means that he does not claim that anything like the Original Position ever existed, and is not vulnerable to the objection that it didn't or couldn't exist. But, like all hypothetical contract theorists, he does have to answer one very obvious question, namely: if this contract is hypothetical, then why on earth should we care about it?

There are, after all, lots of imaginary contracts. For instance, I can imagine everyone agreeing to give all power, wealth, resources, and money to me, and pledging to spend the rest of their lives in my service. But so what? The fact that I can imagine this doesn't have any obvious implications for social justice. In particular, the fact that I can imagine a world like this doesn't even begin to imply that the rest of you are all obligated to live by this imaginary contract of mine. Rawls' contract is just as imaginary as my 'I am the cosmic dictator of everything' contract; so why should we be bound by it?

One answer (which is wrong, but has gotten a lot of play) is: because the people in the Original Position are our very own selves, stripped of such unnecessary encumbrances as gender, race, religion, and so forth. The contract made in the Original Position is binding on us because it was made by our Real True Deep-Down Selves. Rawls doesn't think anything of the kind.

What Rawls actually says is quite different. He refers to the Original Position as an 'expository device' (TJ 21), whose point is:

"simply to make vivid to ourselves the conditions that it seems reasonable to impose on arguments for principles of justice, and therefore on those principles themselves. (...) The aim is to rule out those principles that it would be reasonable to propose for acceptance, however little the chance of success, only if one knew certain things that are irrelevant from the standpoint of justice."

So: if I want to convince someone that some principle is just or fair, I have not (we usually think) so much as made an argument if I just say: I am blonde, and this principle would be good for blondes, and that's why you should think that it's a fair principle. That no more supports the claim that that principle is fair than the claim that I filed my tax returns early this year supports the claim that the moon is made of green cheese.

However, while people are not normally inclined to mistake claims about my tax returns for good arguments about the composition of the moon, they do sometimes convince themselves that things that will favor them are actually just. The point of saying that the people in the Original Position do not know their own gender, class, race, religion, and so on is just to prevent them from making arguments that they would make only if they knew that those arguments would favor them in particular. (E.g., to prevent me from making arguments that would favor blondes, Seb from making arguments that would favor male lawyers, etc.) And the reason to rule these arguments out is: that we don't think that they are good arguments.

Similarly, the people in the Original Position are perfectly rational. This isn't because Rawls thinks that people in real life are perfectly rational; it's because we wouldn't accept an argument that said: there they are, thinking about what sort of principles to be guided by, when all of a sudden they make this incredibly stupid mistake about the nature of economics and human psychology, and decide to become (for instance) Maoists; and since their choices are binding on us, we should be Maoists too. Of course we should conceive of them as perfectly rational, since we don't want them to make mistakes, and wouldn't accept their conclusions if we thought that they had.

The people in the Original Position are concerned only to advance their own ability to live the lives they want. They are not concerned with others' ability to do so. They are not altruistic, and thus they will not sacrifice their own interests to advance those of others, nor are they envious, and thus they will not sacrifice their own interests to harm others. They are concerned only with themselves. (Note: since the parties do not know what they value in real life, they do not know whether or not they will be concerned with the interests of others. However, having more freedom or opportunities or income will presumably help them to advance the interests of others more effectively in the actual world, should they wish to do so; therefore, the assumption that they are concerned to advance their own interests In the Original Position does not prejudge this question.)

The reason for this assumption is that justice is concerned to adjudicate conflicting claims, and principles designed to specify what is just should not build in, at the outset, the assumption that some people are willing to give up their claims out of kindness or altruism (or, for that matter, spite), but should rather be concerned to explore how far those claims are legitimate. The entire structure of the Original Position ensures that no one can insist on his or her own interests unfairly; this particular aspect of it ensures that no one gives up his or her claims without some good reason to do so.

As I said to Bernard in a comment, one way to think about this assumption is by analogy with dividing a cake by the familiar method of letting one person cut the cake, and then everyone else choose their slice first. When you do this, the whole point is that the person who cuts the cake should think: if I cut a bigger piece, someone else will take it; so I should cut all the pieces equally. That is: you assume that everyone is selfish, and thus will pick a larger slice if the person who cuts the cake produces one. If the person who carves the cake gets to think: "oh, but X is altruistic, so I can make one slice very small and X will take it, and as a result my slice will be bigger!", then the whole thing falls apart. It's the combination of (a) the assumption that all the recipients of cake are selfish, plus (b) a set-up designed to ensure a fair result given this assumption, that produces a fair result. Same thing with Rawls.

So the basic point is: the people in the Original Position are imaginary beings, not "ourselves as we truly are", or something. We care about what they decide because we have set up the conditions under which they choose to embody, or make vivid, conditions that we (in the actual world) do accept as constraints on arguments about justice. ("Or", Rawls says, "if we do not, then perhaps we can be persuaded to do so by philosophical reflection." (TJ, p. 21.)) This is why he says:

"No society can, of course, be a scheme of cooperation which men enter voluntarily in a literal sense: each person finds himself placed at birth in some particular position in some particular society, and the nature of this position materially affects his life prospects. Yet a society satisfying the principles of justice as fairness comes as close as a society can to being a voluntary scheme, for it meets the principles which free and equal persons would assent to under conditions that are fair. In this sense its members are autonomous and the obligations they recognize self-imposed." (TJ 13; emphasis added.)

Motivation: Thus far, I've been talking about the things we don't know and/or don't care about in the Original Position. But that doesn't get us to a choice, since thus far we have not provided any motivation for the people in the Original Position to choose anything at all. (I mean: if I said to you: suppose you don't know this, that, and the other thing: what would you choose? you'd probably answer: beats me; what am I trying to accomplish? What do I care about? Same here.)

Rawls' answer is: the people in the Original Position do not know what, in particular, they want to do with their lives. But they do know that they have some idea or other about what they want to do, or the kinds of lives they want to live, and that it matters to them that they be able to live these lives. Moreover, they know that they might change their minds about which sorts of lives are worth living over time (say, by undergoing a religious conversion.) And they want to secure their ability to live as they think best. (Or, as Rawls puts it in PL: their "capacity to form, to revise, and rationally to pursue a determinate conception of the good." (PL 312, among other places.))

By itself, this doesn't advance us much: different conceptions of how you want to live your life lead you to value different things, so just knowing that you have some conception or other doesn't necessarily give you any idea of how to choose. So Rawls adds one more thing: we know that there are certain all-purpose goods -- things that are useful (or at least not harmful) given almost any conception of how you want to live your life.

Rawls calls these all-purpose means 'primary goods'. Some of them are 'natural': their distribution does not directly depend on society (though the rewards they command might). Some of them, on the other hand, are 'social': their distribution does depend on which set of rules we adopt. The social primary goods that Rawls identifies are: rights and liberties, opportunities and powers, income and wealth, and the social bases of self-respect. Rawls says: since these primary goods are useful given (almost) any views about the kind of life you want to lead, we can assume that you would prefer more rather than less of them. And that's what you'll try to achieve in choosing among principles.

Note one thing about this view: it is all about securing one's freedom to live whatever kind of life one thinks best. The entire justification of Rawls' two principles is his claim that they would be chosen under fair conditions by people whose only motivation is to secure that freedom.

Results:

As noted above, Rawls thinks that people in the Original Position would choose these principles:

I. Each person has an equal claim to a fully adequate scheme of basic rights and liberties, which scheme is compatible with the same scheme for all.

II. Social and economic inequalities are to satisfy two conditions: first they are to be attached to positions and offices open to all under conditions of fair equality of opportunity; and second, they are to be to the greatest benefit of the least advantaged members of society.

These principles are what's called 'lexically ordered': in a dictionary (=lexicon), you have to go all the way through through the As before you get to any of the Bs; here you have to fully satisfy the first principle before you get to the second. There are no tradeoffs between liberty and prosperity, any more than, in a dictionary, a word like 'Azzzzzzzzzzzz' can come after a word like 'Baaaaaaaa', on the grounds that the latter somehow makes up for beginning with B by having so many As later on. Moreover, there are no tradeoffs between the first part of the second principle ('fair equality of opportunity') and the second part ('to the greatest benefit of the least advantaged.')

Why? Well: let's start with liberty. The liberties mentioned in the first principle, according to Rawls, are: "freedom of thought and liberty of conscience; the political liberties and freedom of association, as well as the freedoms specified by the liberty and integrity of the person; and finally, the rights and liberties covered by the rule of law." (PL 291) The reason these liberties take priority is clear: they are so closely connected to our capacity to live as we wish that we should not trade them off for anything other than: more such liberties. (E.g., securing the rule of law may require restrictions on who can speak in a courtroom, and when; it does not require a prohibition on advocating a given religion under any circumstances.)

For instance: if you know that you might hold some religious view, and that if you do, you will probably take it seriously, then you cannot afford to gamble with religious freedom. You cannot, for instance, think: well, if a minority religious view is oppressed, odds are I won't be in it -- that's why it's a minority, after all -- so why not? If what matters to you is your ability to live by your conception of the good, you won't take this gamble and jeopardize it, especially in the absence of any reason to think that the benefits of this oppression, whatever they are, will matter to you. Nor, for similar reasons, will you be inclined to accept the possibility that you might be a slave, or forced into servitude, or mutilated by others at will. Nor will you want it to be possible for you to be thrown in jail for no reason, and without the possibility of pleading your case. These liberties are just too important.

Next, fair equality of opportunity. That there should be some sort of equality of opportunity is presumably clear: if you don't know who you are, you will not want the best jobs to be reserved for Catholics, or people with an even number of letters in their name, or whoever. By 'fair equality of opportunity', Rawls means something more than the absence of overt discrimination. He means, specifically, that insofar as possible, people with equal talents, willingness to work, etc., should be able to compete equally.

'Insofar as possible' means that he is not trying, for instance, to somehow remove all the differences that could possibly affect people's success from their childhood, upbringing, and so forth. That would be impossible (and also crazy.) However, while fair equality of opportunity does not require some sort of total homogenizing, it does require that when we can set things up in such a way as to level the playing field so that people can compete fairly, we should.

The most obvious example here is education. The mere absence of overt discrimination is consistent with inequalities in the kinds of education available to kids from different backgrounds that are extreme enough to bar some from effective competition for lots of jobs. Clearly, such inequalities exist now: kids in some neighborhoods do not have access to any schools that either teach them what they need to know in order to compete for skilled jobs, be informed citizens, etc., or inspire in them any desire to learn those things on their own.

If you were in the Original Position, Rawls says, you would regard this sort of stunting of what might end up being your potential as unacceptable; and you would not be greatly consoled by the thought that if only you had somehow or other acquired for yourself an education that most people have to go to good schools and good colleges to get, you would then find that you faced no further discrimination. (You would be a little consoled: for one thing, it's not impossible to educate yourself as a child; it's just very difficult; but 'difficult' is better than 'impossible'. For another, having 'No Xs need apply' written into the social fabric harms people in more ways than one: it deprives them of jobs, but it also deprives them of some of the social bases of self-respect, which (as you may recall) is another of the primary goods.)

Still, if you were concerned to advance your ability to live the kind of life you think best, and you didn't know who you would turn out to be, then while you'd prefer a system under which most poor kids (or black kids, or whatever) can't get jobs because they don't get a decent education to a system in which they were barred from those jobs outright, you'd prefer to either of those options a system that gave everyone (and thus you) the education that would enable them to compete fairly for the jobs they want. You'd think: of course I can't be guaranteed success; but I can be given the opportunity to try my best, without, for instance, being doomed from the outset because no one could be bothered to teach me to read. And if I am trying to secure my ability to live the life I think best, I will make sure that I have that opportunity, as far as it's possible to do so without sacrificing my liberty.

Finally, income and wealth. Here, Rawls says: given two systems, each of which satisfies the principles already set out, you should choose the one in which the least well off (in terms of income and wealth) are as well off as possible. Any inequalities that benefit everyone are fine; given a choice between two otherwise similar systems, one of which includes those inequalities while the other does not, the least well off will be better off, and so we should choose that system. Inequalities that do not benefit the least well off, however, are not fine; and given a choice between two otherwise similar systems, one that includes them and one that does not, we should choose the latter.

Here again, the reasoning is pretty clear. The people in the Original Position do not know who they will be. For this reason, they cannot think: well, I am pretty sure that I will be reasonably well off, or: that I will be poor. They do know that money will generally help them to live the sort of life they think is best. Moreover, in general, a given amount of money makes a bigger difference to your capacity to live the kind of life you want the less money you have. ($100 extra dollars a year may make a big difference to you if you make $500/month; it's likely to make a lot less of a difference if you are Bill Gates, in terms of your ability to realize your conception of the life you want to lead.) If you're concerned to secure your ability to live the kind of life you want to lead, I think, contra Sebastian, that you'll prefer a system in which your worst case scenario is less bad.

This is especially true given that when we're choosing among two different sets of rules, we normally do not know the kinds of probability distributions that Seb used in his post. Normally, the comparisons are much rougher than that, as regards the actual amounts that people end up with and the proportion of people who will end up in a given position.

Be that as it may, I suspect that if we were in the Original Position, even if it might make sense to be willing to trade off some benefit to the least well off for a large average benefit to most other people if the least well off will not fall below a fairly high level under either scenario, it would not make sense to make this tradeoff if the least well off ended up in, say, the condition of today's minimum wage workers. I can see being willing to gamble if you're ensured a standard of living equal to to US middle class; I think it's a lot less likely if the alternatives you might end up with are markedly worse than that. If you have a US middle class standard of living, you can realize a pretty broad range of conceptions of the good life. If you have the standard of living of a minimum wage worker, your options are a lot more limited; and someone who is concerned to secure her ability to live the kind of life she thinks is best will not want to run that risk.

***

In any case, that's Rawls made as simple as I can make it. I don't expect everyone to be convinced; but I hope that the basic idea behind the view will seem at least somewhat plausible, and that the idea that it yields the results Rawls says it does will not seem downright insane.

One reason this seemed worth explaining is just that Rawls really doesn't fit a lot of preconceptions about liberals, despite being one of the most prominent liberal political theorists. As I said, his view is not about not about "equality of results", or "leveling down", or imposing some sort of homogeneity on everyone, or punishing people for being successful, or anything like that. It's about trying to protect our freedom to live the kinds of lives we think we should live. At bottom, Rawls is all about freedom -- along with the fairness that comes from designing a system in such a way that you can't tailor it to fit your own interests. And, in our present political climate, that seemed to me to be worth noting.

***

(Note: Rawls does not talk a lot about health care. I differ from him, and follow Norman Daniels, in thinking of access to health care as part of fair equality of opportunity. But that's another story.)

Comments

[liberal japonicus]

what he calls the 'Original Position'. (Trust me: all the off-color jokes involving this phrase have already been made.)

but isn't that the point of the Veil of Ignorance, that we don't know that all these jokes have been made?

[Sebastian Holsclaw]

"Here again, the reasoning is pretty clear. The people in the Original Position do not know who they will be. For this reason, they cannot think: well, I am pretty sure that I will be reasonably well off, or: that I will be poor. They do know that money will generally help them to live the sort of life they think is best. Moreover, in general, a given amount of money makes a bigger difference to your capacity to live the kind of life you want the less money you have. ($100 extra dollars a year may make a big difference to you if you make $500/month; it's likely to make a lot less of a difference if you are Bill Gates, in terms of your ability to realize your conception of the life you want to lead.) If you're concerned to secure your ability to live the kind of life you want to lead, I think, contra Sebastian, that you'll prefer a system in which your worst case scenario is less bad."

Right, this is the crux of our disagreement. I think it is very possible that this decision is sensitive to the absolute level of the worst-off, not just the relative level.

I'm also glad you focused on the 'lexically ordered' concept. I suspect much of my frustration with latter political philosophers and those who invoke Rawls for their political ends involves a lack of attention to the lexical order of the principles.

[dutchmarbel]

I never heard of him before (I feel I should blush now :) ) but if I get it right shorter Rawl is "what would a just society be if everybody believed in reincarnation"?

he social primary goods that Rawls identifies are: rights and liberties, opportunities and powers, income and wealth, and the social bases of self-respect.

Are those supposed to be a lexically ordered too?

[Step2]

The problem a lot of people have with the Original Position is that it requires a level of neutrality many people reject at some fundamental level.

Alternatively, they may take the Adam Smith opt-out and say that the invisible hand of the market will penalize those who unfairly discriminate. This places the market in direct competition with the Original Position, since both are assuming self-interested decisions made by rational agents. Fortunately, almost nobody makes the assumption these days that the market is rational.

One criticism I have read about Rawls that seems valid, is that he believed the motivations of people would not change from the noble concept of securing their own freedom to the corruption of taking advantage of the rules. That position seems a bit naive.

[hilzoy]

Step2: the market is not in competition with the Original Position. It's hard to see how it could be: after all, the market works in the actual world, while the OP is a hypothetical construct designed to serve a purpose the market does not pretend to serve, namely providing a justification for criteria by which the justice of rules can be assessed.

It could turn out that the OP conflicted with the market by, say, leading to the conclusion that market economics is somehow illegitimate. But it doesn't: the first principle (liberty) includes "the right to hold and to have the exclusive use of personal property" (PL 298), which presumably entails the right to buy and sell that property. What constraints, if any, should govern what we can buy and sell (children? organs? the sun?), and what rules should govern the market (e.g., accounting rules for publicly traded companies, zoning restrictions on the use of real property, etc.), are not settled by this right; but those questions concern the form a market system should take, not "the market" per se.

Moreover, if it turned out that markets would penalize those who discriminate unfairly, then anti-discrimination laws would be unnecessary, but e.g. support for education would not. So I'm not sure I see the point here.

Rawls does not think that the motives of people in the actual world will mirror the motives of people in the OP. He has a lot to say about human motivation, but not along the lines of the 'criticism' you mention.

Dutchmarbel: it's only the principles that are lexically ordered.

[rikchik]

Dutchmarbel: perhaps, but only if they believed in random reincarnation. Historical societies that believe(d) in reincarnation also believed that you deserved the circumstances of your new incarnation based on your behavior in previous lives, so it was acceptable to discriminate against the lower castes.

[SomeOtherDude]

Josh Travino wrote:

The march of history seemed to be on the side of the statists, and the great questions of the day revolved about how, rather than whether, the state would manage the lives of the people. Against this, the conservative was ill-equipped; and so he often enough took his refuge in anger, in symbolism, and in isolation from the public square.

http://www.redstate.com/story/2006/6/27/203014/811

How can you even begin a dialogue with a group of political activist who believe their right-wing statism is some magical anti-statism?

This is a political class who has embraced the state to engineer democracy through war and occupation and believe the regulation of church discipline should be the federal government’s responsibility.

Until right-wing statists begin to except the fact that they are NOT anti-state but right-wing state, you’ll be debating pass each other.

[SomeOtherDude]

That should be “accept” and not “except.”

I sure would hate someone to miss my point because of my grammar

[Sebastian Holsclaw]

I hate to threadjack, but I think you miss his point if you believe it has something to do with the current Republican party being anti-statist.

It seems to me that the point "and the great questions of the day revolved about how, rather than whether, the state would manage the lives of the people." is reinforced by the modern incarnation of the Republican Party.

[Donald Johnson]

I didn't read Step2 as saying that the market necessarily is in conflict with the OP. I thought he was saying that market fundamentalists seem to believe this--they define market transactions as inherently just, so arguments based on the OP are at best superfluous. They might reinforce what market fundamentalists already believe, but if they don't, then so much the worse for the OP.

[bob mcmanus]

"Rawls does not think that the motives of people in the actual world will mirror the motives of people in the OP"

Am I taking too broad or simple interpretation of the OP to see its common application in real life?

For instance, were arguments for preserving (or abolishing, on the part of Yglesias) the 60-vote filibuster an example of an OP argument? "Well, we might want it when we are in the minority, so we should preserv the rule now."

[hilzoy]

bob: it's the same basic idea, though the OP is a lot more detailed, since it's designed for a much more specific task. But yes, the idea is the same: that you should favor rules not based on whether they advantage your side, but on whether they're fair.

(The big difference, though, is that in the filibuster argument, we can all expect to be on the wrong side eventually. Not so with the structure of society, absent reincarnation. -- I mean, even if I lose my shirt tomorrow, there's no undoing my upbringing, education, etc.)

[Jay S]

Sebastian: I'm also glad you focused on the 'lexically ordered' concept. I suspect much of my frustration with latter political philosophers and those who invoke Rawls for their political ends involves a lack of attention to the lexical order of the principles.

Who are these people? I don't remember ever hearing of Rawls. I do hear Warren Buffet describe the birth lottery thought experiment, which is apparently related to Original Position. Perhaps if you name names, we can see what you are talking about.

[Sebastian Holsclaw]

"Who are these people? I don't remember ever hearing of Rawls."

Umm. I don't even know how to respond to "I don't remember ever hearing of Rawls" other than to say that you aren't the target audience of this discussion if you haven't. (I don't mean it in a snotty way. If we were talking about Jazz and you said you had never heard of Coletrain I would have difficulty knowing how to respond in much the same way).

But in an attempt to explain, Rawls was a hot topic among people interested in politics during my college life, and has been such for the more philosophical looks at politics for my entire lifetime. He is regularly invoked (though I'm beginning to suspect improperly invoked) in discussions of egalitarianism. Discussions of the original position and max-min societal ordering show up regularly at say CrookedTimber.org.

[jayann]

JayS, here's an appreciation by Nussbaum, which will give you an idea (I didn't read it, I'm taking it on trust)

http://chronicle.com/free/v47/i45/45b00701.htm

Sebastian, "Coletrain"?! (do you use voice software?)

[Sebastian holsclaw]

Lol, Coltrane.

No, I'm just a horrific speller. Sheesh, smack me please. :)

[Sebastian Holsclaw]

It was instant karma punishing me for being snarky.

[jayann]

It was instant karma

right :)

[bob mcmanus]

The trackback at the top leads to an appreciation of Rawls and hilzoy-on-Rawls.

I liked this:"Rawls coherently argued that you have to extract your specific life situation before you can apply the concept of fairness." ...punk**a

Personal experience & history may not be the worst teacher, but it is certainly not the best.

[Bernard Yomtov]

hilzoy,

Thank you very much for posting this. I recall a time when some friends were hard at work on Rawls-based dissertations, but that was long ago, and I didn't pay that much attention.

One initial reaction I have is that the logic is a bit facile in some respects, and rests on assumptions not made explicit, not least a high degree of risk aversion in economic matters. As I noted in response to Sebastian, this is, as an empirical matter, accurate, so maybe that's OK.

Other things I find puzzling. Consider your example that Catholics should not get priority for good jobs. As a matter of my individual expectation, I don't see why I would worry about this in the OP. After all, I might end up a Catholic, so the probabilities balance out. You say that Rawls concludes that

insofar as possible, people with equal talents, willingness to work, etc., should be able to compete equally.

But if I know nothing of what my talents will be why should I not be as willing to gamble on being Catholic as on having whatever other attributes will help me compete? Why would I not be indifferent between a system that favored Catholics and one that favored smart people? It makes no difference to my chances as seen from the OP.

Is the idea that we would agree to reduce or eliminate the weight of personal attributes, and reward only behavior, to eliminate randomness as much as possible?

The description of the economic conditions as those of "moderate scarcity" strikes me as more evasive than helpful. This lack of precision is what makes Sebastian's objections reasonable. Rawls could be describing the US, or Mexico, or China, or Saudi Arabia. I think the conclusions about permitted inequalities would vary across these countries.

Anyway, these are things I will try to reflect on.

Thanks again for the post.