Argument for Rawls’s principles Notes for April 23

Main points

Today we discussed Rawls’s reasons for thinking that the parties in the original position would choose his principles of justice.

Next week we will see why he thinks the parties would prefer his principles of justice to utilitarianism.

The reasoning for the two principles

We have already seen Rawls argue for his two principles of justice when we discussed what I called his arguments against libertarianism.** See A Theory of Justice, §§ 11–13. But those arguments aren’t part of the official theory of A Theory of Justice. They are just Rawls’s autobiographical reasons for formulating his two principles of justice in a particular way for presentation to the parties in the original position.

The theory of A Theory of Justice is that the parties in the original position will choose Rawls’s principles over all of the other alternatives. This is the first part of that theory.

Specifically, Rawls’s argument turns on two points:

  1. It is rational for the parties to use the maximin rule for making decisions.
  2. If the parties do use the maximin rule, they will choose Rawls’s principles of justice.

Is it rational to use the maximin rule?

Generally speaking, it isn’t. But that isn’t relevant. The question is whether it is rational for the parties in the original position to use it. They are in a very strange situation, after all. Rawls maintains that three features of their strange situation make this strange rule appropriate (pp. 154f).

  1. They don’t know the probabilities of whether a particular choice would harm or help them (the real individuals they represent).
  2. They don’t care much about gains over the minimum.
  3. Going below the minimum would be unacceptable.

I put off quarreling about the first point, since it is what will be at issue between Rawls and the utilitarians on Monday. But I asked how it is that the parties would know the things mentioned in the second and third points. How do they know whether gains over the minimum are important to them or not? And how do they know what they would find unacceptable?

Really, there are two questions that I did not keep fully separate.

  1. How do they know when they find sacrifices in liberty unacceptable?
  2. How do they know that the relative measures employed by the difference principle reflect what they find acceptable or unacceptable

When I look back at the relevant page in Rawls’s text, I see why I didn’t keep these topics separate. He didn’t separate them either. See §26, p. 156.

Still, let’s see what happens when we pull them apart.


We know that Rawls’s big objection to utilitarianism concerns liberty. He’s worried that the interests of a large number of people might outweigh the interests of individuals in utilitarian calculations, with the result that individual liberty would be suppressed on some occasions.†† See 3.4 in the handout on Mill’s On Liberty.

We know that Rawls finds this unacceptable. But why do the parties in the original position find the loss of liberty unacceptable? Remember that they get something in return: the benefits of living in a society that maximizes happiness. Chances are good that they will be winners rather than losers. But Rawls thinks that the cost of being a loser is so high that the parties will reject utilitarianism to avoid the risk. We’ll come back to this on Monday.

On Wednesday, when we read Hart, we will pick up two other features of Rawls’s treatment of liberty.

The first of these is Rawls’s distinction between the general and special conceptions of justice. The general conception allows sacrificing liberty in order to improve the position of the worst off class but the special conception does not.

In drawing this distinction, Rawls maintains that the parties would allow for limits on liberty in some cases. When a society is too poor for its members to enjoy liberty, Rawls allows the government to limit liberty, but only for the sake of economic development aimed at making it possible to enjoy liberty. Once the threshold of economic development has been crossed, sacrifices of liberty are ruled out. The very poor society is governed by what he calls the general conception of justice while the wealthier one is governed by the special conception of justice.

When we read Hart, we will also talk about some slippage between these two ideas.

  1. Any sacrifice of liberty is intolerable.
  2. Sacrifices of the liberties on a fairly short list are intolerable. For example, freedom of speech, freedom of religion, freedom to join political parties, and so on.

Rawls can’t seriously mean the first of these because he’s imagining there will be a state that will impose regulations, collect taxes, and enforce laws. All of these activities limit liberty. But he didn’t defend any particular list either.

In sum, if you look at p. 156, you see that Rawls points to argument later in the book in order to show that losses in liberty would be unacceptable. It looks as though it’s all coming down to Section 82, which we’re reading for Wednesday’s class. So he gets an incomplete on liberty.

Economic goods and opportunities

We’re still talking about why it is rational for the parties in the original position to use the maximin rule. The question at hand is: how do they know that they don’t care much about gains above the minimum and that they find falling below the minimum unacceptable?

I find this curious because the minimum defined in Rawls’s principles of justice is relative to what others have. It is defined in terms of equal shares. Specifically, the minimum share of economic goods and opportunities is whatever people would have if these things were distributed equally plus whatever they would gain through inequality. We never go below the level of goods and opportunities that people would have in an equal distribution; we only go up.

So the parties have to think that anything less than they would have under an equal distribution, plus gains from inequality, would be unacceptable. But how do they know that? And is it even true?

To make these points, I asked you to consider an alternative way of discovering what level of material wealth and opportunity is unacceptable to people: social science. Go study people and find out how much wealth they need, such that any less would be unacceptable to them. This would not, presumably, be a relative measure. Instead, it would produce an absolute measure: people need housing, food, clothing, health care, indoor plumbing, heat in the winter, and so on, regardless of whether others have much more or much less.

The parties in the original position don’t consider any evidence like that. So why are they so sure that the relative measure of economic wealth that they come up with will truly reflect what the people they represent find minimally acceptable? Maybe it will be too low. More probably, it will be too high.

Looking at p. 156, I don’t see any argument suggesting that having fewer matieral resources and opportunities than the second principle provides would be unacceptable. All I see are references to the unacceptability of giving up liberty for material gains.

Rob’s interpretation

Rob, of course, hit me with a clever alternative to what I was saying that has taken me a few days to figure out. Here it is, in my words, not his.

“The parties find anything less than the minimum of gains over strict equality unacceptable because they are charged with getting as much for themselves as they can. That is the best that the parties in the original position can do for themselves and anything less than the best is unacceptable to them.”

In other words, we should treat the term “unacceptable” as applying to the parties rather than us. The question is what they would find unacceptable given the ways they think. It isn’t about what we would find unacceptable or identifying absolute standards of poverty and wealth in the real world.

That’s immensely clever and it does neatly dodge the line of argument I was making. But I don’t think it’s what Rawls had in mind. I will present two pieces of evidence.

First, here is Rawls’s description of how to meet the second condition for using the maximin rule. Using the maximin rule is rational, he says, for a person who “has a conception of the good such that he cares very little, if anything, for what he might gain above the minimum stipend that he can, in fact, be sure of by following the maximin rule” (§26, p. 154). That doesn’t sound like a description of how the parties in the original position think. It might be an accurate description of how the people that the parties represent think, namely, us.

Second, I don’t think Rob’s alternative fits Rawls’s treatment of liberty. I know he says that “the persons in the original position have no desire to try for greater gains at the expense of the equal liberties” (§26, p. 156). But if you look at the previous sentence, you’ll see that this is what follows if we can establish the priority of liberty.” That happens in §82 (title “The Priority of Liberty”) so it isn’t something he can take for granted here.

Question: aren’t the parties in the original position supposed to decide whether liberty takes priority over increasing wealth? Answer: yes. And, as long as we’re at it, yes it’s odd that the argument is put off until later. The idea is that the parties will look to the general facts about human psychology and the social sciences. That will give them the arguments that we’ll read in §82. Those arguments will convince them to insist on the priority of liberty.

This page was written by Michael Green for Social and Political Philosophy, Philosophy 33, Spring 2008. It was posted April 26, 2008.
Name of website