David Friedman once said “I read Rawls’ A Theory of Justice early on and never was able to figure out why anyone took it seriously, beyond the fact that it provided arguments for conclusions they wanted to reach.” Politics can very often lead people into reverse arguments – starting with a conclusion and then crafting arguments which seem to justify that conclusion. John Rawls seemed to admit this as well, remarking in A Theory of Justice that “We want to define the original position so that we get the desired solution.”   

David Schmidtz argues at many points throughout his book Living Together that the way Rawls frames the original position fails to do justice to the question of justice. In order to reach the “desired solution”, Rawls stipulates that everyone will voluntarily act in perfect accordance with the demands of justice – because this incredibly unrealistic assumption is necessary in order to reach the solution Rawls desires. Schmidtz writes:

Assuming perfect compliance leads down one road, realistic assumptions down another. We assume perfect compliance not because realistic assumptions go nowhere but because they don’t go where Rawls wants his theorizing to go.

In another section, Schmidtz explains why Rawls’ approach of handwaving away the issue of compliance from his armchair is a mistake:

Yet what makes x merely a “distracting detail” is that it makes no difference to the question at hand. Nothing changes when we set it aside. Therefore, whether x is a mere distraction is a matter for discovery, not a stipulation. 

For example, suppose we aim to determine water’s boiling point. To keep it simple, suppose we classify altitude as a distracting detail and set it aside. That idealization may sound reasonable, but it would, as a matter of contingent fact, be incompetent. Why? Because altitude is no mere distraction when determining water’s boiling point. As it happens, boiling point is a function of atmospheric pressure, and atmospheric pressure is a function of altitude…It takes experience to know whether altitude is a mere detail.

Simplifying is risky. It is fine to set aside details to reveal an underlying logic operating across worlds. But if we set aside the fact that incentive structures affect behavior in law-like, robustly predictable ways, then we aren’t setting aside details to reveal a system’s underlying logic. We are setting aside the underlying logic. Rather than setting aside what makes no difference, we are setting aside what changes everything. 

It is easy to slide form ignoring to ignoring with prejudice: setting details aside not because they don’t affect the argument but precisely because they do. 

There’s another issue Rawls hand-waves away in his Theory of Justice that beyond what Schmidtz points out.

Rawls’ theory is a form of hypothetical social contract theory. This theory argues that under certain stipulated conditions people would have hypothetically agreed to a given social arrangement, and this stipulated hypothetical agreement therefore generates real world, enforceable obligations to support that arrangement. Rawls is not lacking in ambition with his hypothetical agreement either. Not settling for a mere majority rule, Rawls says consent must (and would be!) unanimous. Ever the idealist, Rawls says “the requirement of unanimity is not out of place and the fact that it can be satisfied is of great importance. It enables us to say of the preferred conception of justice that it represents a genuine reconciliation of interests.”

Why is Rawls so sure there would be unanimous agreement in the original position? This is his entire justification on that point:

To begin with, it is clear that since the differences among the parties are unknown to them, and everyone is equally rational and similarly situated, each is convinced by the same arguments. Therefore, we can view the agreement in the original position from the standpoint of one person selected at random. If anyone after due reflection prefers a conception of justice to another, then they all do, and a unanimous agreement can be reached.

Rawls simply stipulates that the only possible sources of disagreement are ignorance, irrationality, and personal bias. Therefore, in the original position where everyone is equally knowledgeable, rational, and has no knowledge of their personal position, there cannot be any genuine disagreement about justice. In Rawls’ vision, there are no genuine differences in thinking among different minds. Remarkably, it also turns out that the solution to the issue of justice that would be most persuasive to these perfectly informed, perfectly rational, perfectly unbiased hypothetical people is…the system John Rawls himself prefers without needing the benefit of being in such a superhumanly privileged position. It’s fair to say Rawls succeeded in tailoring the original position with all the assumptions necessary to reach his desired result, but I must admit I find this only serves to undercut his case rather than enhance it.