The second part of the book then moves on to discussing "forms of reasoning" and presents an argument for taking account of plural types of reasoning as advanced from many sources of concern. This argument for plurality oddly echoes the concerns of John Rawls' second major work Political Liberalism which alleged, in view of the ineluctable pluralism of contemporary society the need to abandon the search for "comprehensive" views of social justice in which moral and political considerations were considered together in favour of a "purely political" theory of public reason. Interestingly, the latter work also concludes with a "reply to Habermas" despite the general argument of Between Facts and Norms also having voiced a criticism of a type similar to that of Political Liberalism of political theories that are guided too much by concerns derived from morality.
The oddest thing about the degree of convergence between these three books is the way in which very similar concerns are expressed despite the authors having quite different aims in view in many respects. Sen deliberately sets his theory up in opposition to that of Rawls, though it is true that it is mainly the early and not the late Rawls that he has in his sights. Habermas' concentration on providing a theory of law that can be disentangled from overly normative theories is again a reply to the early Rawls whilst Rawls' own late work is a considerable revision of the theory he presented earlier on.
However, and returning to the puzzle about the lack of direct engagement on Sen's part with the question of justice, what seems most at issue for each of these works, albeit in different ways, is the exploration of what might be called the theory of public reason. Each of the works marks, thus, not a theory of justice but, at best, a prolegomena to one.
In Rawls' early work A Theory of Justice the subject matter of justice was described as "the basic structure of society, or more exactly, the way in which the major social institutions distribute fundamental rights and duties and determine the division of advantages from social cooperation" (p. 6, both editions). Such a concern was narrowed in one crucial respect that Sen complains about in his discussion of impartiality by being understood in a contractarian way as "a closed system in isolation from other societies" and, in another way, also complained about by Sen, in terms of describing a "well-ordered society" in which everyone is presumed to act justly and to do his part in upholding just institutions.
These two ways in which the subject of justice were narrowed by Rawls concerned focus on institutions as the primary distributive agent of justice within one society on the one hand and a conception of the maximal way such institutions could be conceived of as delivering justice on the other hand. Sen complains both about the focus on institutions per se and on the "closed" character of such focus as concerned only with the operation of one society but his more fundamental complaint is the pursuit of perfection in the theory. It is worth looking at the contrast between Rawls' reason for advancing the theory in terms of such a search for perfection, however, and the reasons why such a search is condemned by Sen. Rawls presented the theory of justice offered in the work of this title most comprehensively as a reply to utilitarianism. Indeed he wrote quite clearly: "My aim is to work out a theory of justice that represents an alternative to utilitarian thought generally" and a characteristic of such thought, commented upon by Rawls, is precisely the comprehensive character of its response to problems of morals and politics. It is to provide a theory that covers the same degree of ground as utilitarianism that Rawls writes A Theory of Justice and yet it is precisely this ambition of it which his later work abandons and which is also the subject of criticism of the early work by both Sen and, in a different way, Habermas.
Sen's basic reason for attacking the kind of comprehensive view presented by A Theory of Justice is his argument that arrival at a set of principles on which all could agree is neither necessary nor sufficient for an account of justice. However the question that emerges at this point concerns the structure and nature of a reason that can guide a given plurality of approaches, a concern that utilitarianism addressed in its own way and which justice as fairness is intended to respond to in a different way. Both theories begin with the understanding that there exist a plurality of persons with separate systems of ends and then determine how to respond to this given situation with procedures for demarcating a public reason. The shift from A Theory of Justice to Political Liberalism is not one that shifts this concern since the theory of public reason is central to the latter work. Utilitarians also work with a conception of public reason. For Rawls (both early and late) and the utilitarians such a theory of public reason is at the heart of their conception of how to provide an account of legislation (though utilitarians, typically, have no specific account of justice).
Rawls (both early and late) and the utilitarians conceive of the notion of such public reason in terms of an account of impartiality. So also do Sen and Habermas. At this point we can begin to see a key set of structural parallels between the accounts that emerge from each of the theories. In a certain light, the divergence between Rawls' early and late theories can thus seem to have less importance than he gave it. It can also appear that the distance between the early theory and the concerns of Sen and Habermas have, to a degree, been overplayed.
However, what enables a return to understanding the difference between the early Rawls and the works of Sen and Habermas is the manner in which the early work arrives at the conception of public reason offered. Here the conception of public reason is derived from a contractual procedure and driven towards an account of some over-riding principles that are taken to be central to the pursuit of justice, considered as the "first virtue" of social institutions. The problem suggested in Political Liberalism is that the early view presupposes a conception of morality and, in this work, an "impartial" approach is taken towards comprehensive views of it in favour of a view of "reasonable pluralism". It is this that is supposed to mark the understanding of a "reasonable" view of the political but such a view does itself make striking presuppositions including ones favourable to "democracy". It is this that marks the theory in question as "liberal".
In making these presuppositions the later Rawls effectively conceals under the heading of "politics" some deep moral choices that are meant to guide the interaction between persons and shape the basis of public reason. In a sense, although the topic of public reason is much more to the fore in Political Liberalism than in A Theory of Justice the truth is that the former work presupposes an account of practical reason and subordinates it within its description of public reason.
Exactly the same thing occurs within the work of Sen as in Political Liberalism and it is precisely this way in which an assumed account of practical reason is effectively subsumed within the description of public reason that produces the effect in both works of a kind of pious appeal. Both works really take the essentials for granted and this follows from the abandonment, I would suggest, of the aspiration towards a theory that has a clear reference to the ideal. Take out the ideal and you are only left with a supposition about the real, a supposition that was well supplied by utilitarian calculations. In a central sense, the abandonment of the comprehensive view ceded the ground of such views back to the utilitarian picture that the early Rawls was so careful at providing an alternative to.
It would be a different matter to reply to Habermas and would require a more detailed posting but suffice it to say that the disappearance of "discourse ethics" from Between Facts and Norms is of a piece with the retreat from comprehensive views in Political Liberalism and the endless pursuit of pluralism in The Idea of Justice. All three of these works move in a non-ideal space and, whilst the need for non-ideal theory is real enough, the shape and fate of each of these works suggests strongly that the abandonment of comprehensive views was part of a structural neglect of the theory of practical reason.
Essentially, in arguing for a view of public reason that emerges from a more generic account of practical reason, I am aiming not merely at a rescue of the perspective of A Theory of Justice but also for the need to think about the manner of approach Rawls developed in this early work in relation to the classic manner in which Kant's Doctrine of Right relates to his general account of practical reason. The latter is itself a controversial topic but the need to see Kant's own conception as "comprehensive" is part of the requirement for a revival of such an approach if a theory of justice is to be defended, a theory, that is, that enables distributions to be seen from the perspective of a public reason that has been guided by an approach to practical reason.