John Rawls proposed a method, which he called the veil of ignorance, for determining which social customs were just and which were unjust. The veil of ignorance criterion is as follows: a rule is just if everyone would agree to it given that they were made ignorant of their position in society. That is, the just society would be chosen by people who had set aside considerations of their own gender, wealth, race, parentage, ect. Ideally this rule eliminates personal bias from the choice and thus guarantees the fairness of rules.
However, even behind the veil of ignorance there will not be consensus as to which rules are best, throwing into question the assumption that the veil of ignorance would reveal the unique best set of rules, which is one of the reasons Rawls seem to favor it. For example consider two possible societies, one in which the men are rich and the woman are poor, and another in which everyone is moderately wealthy. Even without knowing their gender some people would prefer the unequal society. This is because individuals have different tolerances to risk. Some would consider the chance to be rich worth the risk of being poor, while others would prefer the security of the society that makes everyone equally wealthy. Likewise preferences such as the health of the environment under such a society could interfere with the consensus.
One might suppose that we could fix this problem by asking that people behind the veil be ignorant of their preferences, which would include their tolerance for risk, as well as their social position. However once preferences are discarded it is meaningless to ask people to make a choice. For example consider two possible rules. Rule one is that people must walk to destinations within a mile, and rule two is that people must drive to all destinations. Rule one is safer and friendlier to the environment, while rule two is more efficient. Without preferences however those behind the veil have no reason to pick either of these rules, and so the idea of them being able to decide on any society seems impossible, after all we have even asked them to give up preferences such as “I want to be happy”.
There is a way to achieve the same basic result of the veil of ignorance without making people put aside their preferences. I call it the cut-first-choose-last rule, because it is based on the method for forcing children to divide dessert among each other fairly; the person who cuts the dessert into pieces gets to choose their piece last. When we apply this idea to picking societies we would say that a just society is one which a rational person would create even if they got to pick the position they were born into last. This implies that societies in which men are wealthy and women are poor are unjust, because no one would rationally choose such a society under the proposed rule, for if they did the first people to pick their roles in society would choose to be men, and by the time the designer was to pick their role they would be forced to be improvised.
From the veil of ignorance argument John Rawls concluded that just societies are those in which everyone has equal opportunities, except where unequal opportunities benefit everyone. This principle still holds under the cut-first-choose-last rule. For example consider a society of 4 people where the wealth is distributed in this fashion: $10, $10, $10, $10, and one in which the wealth is distributed in this fashion: $20, $11, $11, $11. The rational person would still choose the second society, which Rawls would consider just, because even though they know that they won’t receive the $20 role they would rather have $11 than $10.
There as still some oddities even with the cut-first-choose-last rule that arise because of different preferences, not towards risk, but towards what is valued. For example some might pick a society in which wealth is distributed as: $9, $9, $9, $9 if the environment was healthier, while others care less about the environment and would simply pick the society that made them richer. John Rawls conclusions are also impacted by this fact, because he argued that from the veil of ignorance that a just society is one in which there was maximal liberty. Because of the variation of preferences however the cut-first-choose-last rule does not yield this result, for the preference for liberty is not universal, some may very well prefer a society in which they are more secure.
I do not believe that this lack of a consensus is failure of the cut-first-choose-last rule; it simply shows that there are a wide variety of societies which are maximally just. One would have to choose among these societies using other criteria, not those of justice or fairness.