The theory of A Theory of Justice is that the way to answer questions about justice is to ask what the parties in the original position would choose.
Specifically, Rawls argues that his two principles of justice would be chosen by the parties over utilitarianism.
The features of the original position are listed in the handout. These are the most crucial points.
The parties only care about the people they represent. (Strictly speaking, they care about the next two generations as well; see the discussion below for more on this.)
The parties are behind the veil of ignorance. They do not know anything that would enable them to estimate the probability of being helped or hurt by any decision they have to make.
The only thing the parties know about what makes life good for the people they represent is that they need primary social goods. Primary social goods are things that are necessary for any rational plan of life. Rawls thinks that liberty, opportunities, wealth, and a sense of self-worth are all primary social goods (§11, §15).
The parties know the people they represent have psychological limits. They have a sense of justice, meaning they are capable of complying with social rules, but they are also subject to the strains of commitment, meaning that they will not comply with rules that prevent them from satisfying their most important interests.
The parties know that the people they represent are in the circumstances of justice. This is something that Rawls gets from Hume. The idea is that they face moderate material scarcity. They do not live in the Garden of Eden, where rules are unnecessary. Nor do they live in conditions of extreme scarcity, where there is not enough to go around no matter what social rules are adopted. Rather, they live in a world where everyone can benefit from the adoption of social rules, much as Hume imagined everyone would benefit from adopting conventional rules for property rights.1
The very most important feature of the original position is the veil of ignorance. The decision is supposed to be fair because the parties do not know who they represent. In fact, they know almost nothing about their society.
I illustrated this by saying they could not know the probability of being right handed. If they knew that there was a ninety percent chance of being right handed, they would gang up on the lefties. That is not because they have any animus towards the left handers. It is just that this is what the rules tell them to do. The rules tell them to think exclusively about the interests of the people they represent. So if there is a gamble that makes sense with nine to one odds, they would take it. Of course, rules that favored right handers over left handers would be unfair. So Rawls prevents the parties from making rules like that by depriving them of any knowledge of how likely it is that they would represent right handers.
We talked briefly about the assumption that the representatives care two generations into the future. I said that this was an awkward patch. The problem is that if the original position only represents one generation, it won’t say there is anything wrong with that generation’s using up all the available resources and saving nothing for future generations. So he said they care about the next two generations.
Why not just stick future generations into the original position as well? One problem is that the different generations are not what Rawls calls the circumstances of justice with one another. There are no mutually beneficial rules since people in the future can’t benefit people in the past or present. Why is it so important that they be in the circumstances of justice? I’m not sure. Another problem is that it would be tricky to identify exactly who is represented. That’s because the choices we make about our society now determine who will exist in the future. We won’t know who counts as a future person until we know what we’re going to do. But then the future persons can’t help to determine what we’re supposed to do. (And if we represent all possible persons, we’ll be in deadlock, as any choice we make will cause some of them never to exist.)
Gabriel also had a clever idea. Rawls tries to block the parties from forming coalitions, such as the imagined coalition of the righties against the lefties. He does this by prohibiting them from knowing what the odds of being right handed or left handed are. If you don’t know the odds of being left handed, you can’t tell whether a policy of adding a tax on lefties is a good bet for you or not. But, Gabriel asked, why can’t they make conditional rules like “if the ratio of right to left handers is 10 to 1 or greater, then impose a special tax on left handed people”? That is certainly contrary to the spirit of Rawls’s project, but I can’t say if he has a rule on the original position that explicitly forbids this kind of thing. So Gabriel may have found a loophole.
Rawls, John. 1999. A Theory of Justice. Revised edition. Cambridge: Harvard University Press.
The difference between Rawls and Hume is that Rawls worries that rules adopted through conventions may not be as fair as alternative rules decided on by parties in the original position.↩︎
There was a handout for this class: 23.RawlsOriginalPosition.handout.pdf