Via Andrew Sullivan, Joshua Cohen and Thomas Nagel in the TLS:
The recognition that political life presents a distinct moral problem is therefore a decisive break in Rawls’s development, and it would be interesting to know when and how it occurred. Rawls says in the Introduction to Political Liberalism (1993) that A Theory of Justice (1971) did not draw a distinction “between moral and political philosophy”. He was, however, concerned from early on with the special moral problems of political society. As early as his doctoral dissertation (submitted in 1950, and addressed to the role of reason in ethical argument), his philosophical work was animated by a sense of the political that is not evident in the undergraduate thesis. Thus he says in the dissertation that a “democratic conception of government . . . views the law as the outcome of public discussions as to what rules can be voluntarily consented to as binding upon the government and the citizens”. For this reason, he continues, “rational discussion . . . constitutes an essential precondition of reasonable law”, and an investigation of the “rational foundation of ethical principles” serves as “an addition to democratic theory, as well as to ethical philosophy”. The close association here of issues of philosophical ethics with concerns about public argument in a democracy marks a sharp departure from the view in the thesis, and anticipates the idea that justice as fairness is “the most appropriate moral basis for a democratic society”. Yet while Rawls came to see that a just society could not be a community “integrated in faith under God”, his personal knowledge and experience of religion were important for the formation of his later view, in particular his views about the kind of public reasoning that we could reasonably expect in a democratic society. Rawls’s liberalism, unlike that of many liberals who know very little about religion, is founded on a vivid sense of the importance of religious faith and an understanding of the difference between genuine and merely conventional religion. He knows what he is talking about when he says in Political Liberalism that the Reformation “introduces into people’s conceptions of their good a transcendent element not admitting of compromise”, that “this element forces either mortal conflict moderated only by circumstance and exhaustion, or equal liberty of conscience and freedom of thought”; and that political thought needs to understand “the absolute depth of that irreconcilable latent conflict”. By insisting on the importance of a terrain of political justification that is consistent with such ultimate commitments but does not depend on them, Rawls was not devaluing religion. On the contrary. The importance of liberty and of separating the state from religion is that they make possible the commitment of all members of a pluralistic society to common political institutions and a shared enterprise of public justification, despite their ultimate disagreements about the nature of the world, the ends of life, and the path to salvation. Such disagreement, he emphasizes, is not a disaster, but the natural consequence of reason’s exercise under free conditions.