First, whether the agent be an individual or a corporate person, it may seem unclear in what way the original position qua decision procedure ensures that all agents come to affirm the same principles of justice in the end. This question proves all the more worthwhile in that Rawls allows that conceptions of justice are subject to change over time, as evidenced by both philosophical and constitutional history. Certainly, the author explains time and again that the positions up for consideration in the original position are drawn from a predetermined list of conceptions from moral philosophy and, moreover, that the precise political conception issuing from the four-stage sequence is likely to vary within a family of similarly reasonable conceptions. Indeed, justice as fairness allows for just possibilities.
That said, it remains to be seen how the agent, having completed the original position qua representational device, would then know whether others had arrived at the same result. In other words, for the resulting political conception to carry political value with a plurality of people, an element of deliberation between people must come into play at some point, be this during the four-stage sequence or following the elaboration of its political conception. Otherwise, the original position leads to neither dialogue nor action.
If we suppose that the agent takes care to simulate difference of opinions within the original position, perhaps by imagining herself from a variety of positions, that deliberative, discursive element might come into play. Yet it bears mentioning that, for Rawls, there can be no variety of positions therein, for the relevant standpoint proves identical for each party to deliberation in the original position qua representational device. Nor, on his view, would agent qua rational person come to favor any conception of justice other than justice as fairness and its two principles of justice, for the reasons exposed in A Theory of Justice. This would effectively undercut the worry that different people might arrive at different conclusions from the constraints.
Rawls’ efforts to clarify above have the further advantage of showing in what way the original position qua representational device is more dialogical than other authors have allowed in the past. The author draws special attention to this when he remarks:
[J]ustice as fairness apparently supposes that citizens’ conception of justice can be fixed once and for all.. This overlooks the crucial point that we are in civil society and that the political conception of justice, like any other conception, is always subject to being checked by our reflective considered judgments. Using the idea of perpetuity here is a way of saying that when we imagine rational (not reasonable) parties to select principles, it is a reasonable condition to require them to do so assuming their selection is to hold in perpetuity (PL, 399).
In truth, this would seem to undercut the frequent criticism that the original position requires one to determine a political conception once and for all, such that the basic conditions are in place for the machine to continue functioning well into the future. Against such a reading, Rawls recalls that the agents running through the original position nonetheless remain within history upon exiting the representational device.
Accordingly, the political conception resulting from the four-stage sequence would then serve as a measure against which inherited institutions might be checked for overall justice. On the other hand, those same critics might, much like Habermas, still worry that the fact of imposing constraints on discourse ahead of time prevents the dialogue from being fully dialogical. To this, the author would only need respond that the constraints are, again, reasonable: each would want for herself the protections offered by liberal rights and democratic institutions, were other facts of her future social existence withheld. More simply, those same critics could themselves be charged with presupposing such rights and institutions all the while being unable to show how they intend to secure those given conditions of unconstrained dialogue.
This does not preclude critics from voicing other concerns. Supposing that the individual is engaged in the original position, one may ask whether it remains reasonable to exclude her real reasons from deliberation if these are nonpublic in the relevant sense. Rawls would likely maintain that it remains fair insofar as, were she to allow her own nonpublic reasons into the deliberation, she would then be required to represent others’ nonpublic opinions in order to ensure the proceedings’ fairness. Yet careful recreation of the latter might require conceptual resources or argumentative sophistication of which she is ordinarily incapable due to limits on human cognitive and rational capacities.
In other words, the individual cannot be everything to everyone nor even to herself. This accounts for a practical motivation underlying the setup of the original position qua representational device in which a single agent, be it individual or corporate person, works through the proceedings. Namely, the reasonable constraints imposed reduce in one way the cognitive burden placed on the agent throughout the decision procedure. There is only so much which she both needs to know and to take into account in order to reach a fair result.