Does a Fair Procedure Guarantee a Fair Outcome?

“The same rule applies to everyone” is often treated as enough for fairness. In A Theory of Justice, however, John Rawls distinguishes three forms of procedural justice. Perfect procedural justice has both an independent standard for the right outcome and a method that guarantees it: if the person who cuts a cake receives the last piece, they have an incentive to divide it evenly. Imperfect procedural justice also has an independent standard but cannot guarantee the result; a trial aims to convict only the guilty, yet mistakes remain possible. In pure procedural justice, there is no single correct allocation outside the procedure. When the starting conditions and rules are fair, a lottery's outcome can be accepted because of the process itself.

These distinctions show that procedures do different kinds of work. Suppose a community course allocates places on a first-come, first-served basis. The rule may be public and consistently applied, yet systematically disadvantage people without flexible work hours or fast internet. The absence of secret dealing does not make the conditions of participation fair. By contrast, where eligible applicants have comparable claims and the places cannot be divided, a genuinely random draw need not become unjust merely because someone loses.

My view is that fair procedures constrain power, reduce favouritism and can sometimes make an outcome legitimate. They should not become a command to stop asking questions. We must still ask who can participate, whether the starting point is acceptable, what problem the procedure is suited to, and whether errors can be reviewed. Public and consistent rules matter greatly; they do not morally cleanse every result.

https://plato.stanford.edu/entries/justice/
https://plato.stanford.edu/entries/rawls/