Rawls’s Difference Principle

Prof. Green led things off by explaining what Rawls was doing. Officially, he was clarifying the meaning of the two principles of justice that he proposed to submit to the parties in the original position. Unofficially, he was arguing both that libertarianism is morally defective and also that only what he called “democratic equality” adequately addressed the moral problems with libertarianism.

Prof. Brown explained how the concepts and methods Rawls uses are close to but not identical with those employed in economics. The most important difference is that economists use concepts such as Pareto efficiency to talk about a distribution of goods while Rawls uses it to talk about a distribution of expectations generated by social rules. That isn’t to say Rawls was wrong; just that his usage is a little different from what you might be familiar with.

Background on Rawls

We are sort of jumping in to the middle of Rawls’s theory. This can be a problem when he uses some terms that you might not find familiar, such as “original position.”

If you want to get up to speed on Rawls’s theory, I think the notes I posted for my political philosophy class should help. Specifically, have a look at:

  1. The Original Position
  2. Arguments for Rawls’s Principles

About those graphs

The graphs are supposed to represent the shares of goods and opportunities held by representative members of social classes under different rules. The idea is that different rules with have different distributive effects. For example, rules restricting the supply of doctors will make most people worse off while making some much better off. Rules allowing private property make everyone much better off and some better off than others.

The axes are labelled x1 and x2; x1 and x2 refer to representative members of social classes. But what the axes really represent are greater or lesser shares of goods and opportunities for the people who belong to those social classes. (It would be comical if x1 and x2 got bigger or smaller as you move along the axes.)

Figure 5 represents the maximin rule. The only improvement from the point of view of justice comes from moving from a lower curve to a higher one. As far as justice is concerned, there is no difference between, x2 having a lot (being at a point very high on the y axis) or having an equal share (being at a point on the 45 degree line). Suppose the share held by x2 is near the top of the y axis. This would mean that the share going to x1 would be on the 45 degree line. If it is possible to move to a higher indifference curve, giving x1 more, that would be an improvement from the point of view of justice even if it meant that the share going to x2 would go down to a lower point on the y axis. In other words, the graph means that all that matters is raising the minimum as high as it can go.

Figure 6 represents the difference principle. It supposes that more goods will be produced if inequality is allowed. That is what the line OP represents: how much would be produced under different rules that result in x1 having more than x2. (Since they are below the 45 degree line, you know that x1 is getting more than x2.) Given the way the indifference curves are drawn, x1 is permitted to gain so long as x2 does as well. Point a is where this stops: any further gains going to x1 would not benefit x2 as well, so they are not permitted by the difference principle. The difference principle, then, allows inequalities only if doing so improves the position of the worst off class which, in this case, is x2.

Incidentally, as Professor Brown said in this class in 2014, this means that Rawls abandoned the principle of efficiency. Suppose the OP curve in figure 6 were flat to the right of point a. That would mean that x1 could gain more than the share at point a without making x2 worse off. The principle of efficiency would count gains for x1 that do not make x2 worse off as an improvement. Rawls’s difference principle, however, counts it as making society worse. It views point a as superior to any point to the right on even a flattened OP curve. Therefore, a society that satisfies the difference principle would not necessarily satisfy the principle of Pareto efficiency.

Finally, figure 8 represents utilitarianism. The indifference curves have a slope of -1 because that is the point where a gain for one person means an equal loss for someone else. At that point, there is no way of making the aggregate of utility greater and so that is where utility is maximized. I had not understood that until today’s class.

Why not natural aristocracy?

I promised to explain why I think Rawls should stick with natural aristocracy, the north east box.

In a nutshell, the only difference between natural aristocracy and democratic equality is that the system of democratic equality has the educational system that will seek to remove the social influences on the course of your life. If it works perfectly, everyone will have exactly the same opportunities to get ahead as everyone else who has the same natural talents. The naturally talented will all do equally well, regardless of whether they were born in rich families or poor ones. And the naturally talentless will all do equally badly, again, regardless of whether they were born rich or poor.

Needless to say, such an educational system will be phenomenally expensive. There will be fewer resources going to the poor in democratic equality than there would be in a natural aristocracy. That is because while both systems are committed to making the worst off class as well off as it can possibly be, democratic equality is committed to treating this goal as a subordinate to achieving equal opportunity. So democratic equality has to have fewer resources to transfer to the worst off class.

Maybe that would be OK if achieving equal opportunity is worth it. But I don’t see how, on Rawls’s view, it could be. Your natural abilities are just as morally arbitrary as the ones you get by virtue of your social class. So why should it matter that some people get more than their talents merit or that others never develop their talents at all? The quotation from Milton Friedman on the handout expresses the idea rather nicely, I think.

I should add that I myself think that equal opportunity is a good thing. (Don’t ask me to explain why, though!) I am just saying that I do not see that Rawls is in a very good position to defend giving as much weight to equal opportunity as he does.

That’s what I was going to say.


Rawls, John. 1999. A Theory of Justice. Revised edition. Cambridge: Harvard University Press.


There was a handout for this class: 08.Rawls.handout.pdf