Linda Hirshman and Jacob T. Levy have been going back and forth at this TNR blog about the role of John Rawls and of "theory" in general in contemporary politics and in politics in general. I’m generally sympathetic to Hirshman’s position, i.e., that ideas matter, especially to elites in politics, though I also agree with Levy’s--in some respects, anti-Rawlsian--caveat, which is that social and cultural conditions aren’t simply susceptible to our Promethean refashioning.
What I found unhealthy about Rawls’s position--a position that probably had more influence in law schools and hence with law professors and hence with judges--is its presumption that, ultimately, our "natures" (both in general and as expressed in our particular cases) don’t matter. The only constraints to which we should pay attention that those that come from our rationality and reasonableness. I regard this as a kind of hyper-Kantianism, which paves the way for a kind of liberal idealism that morphs into "progressivism." Levy says he wants no part of this, and I think that’s a good instinct, for what it requires is a confidence that we can, in effect, be causes of nature as a whole, that we can entirely master our circumstances. This, as I’ve argued before, is at the core of much of contemporary liberalism (or should we call it by its new name?). It’s also connected with what all too often passes for a certain kind of liberal electoral strategery, as if all that’s required is the correct language or frames.
When I welcome the return of thoughtful liberals to "the great conversation," what I mean is a return to talk about nature, human nature, and the imperatives and constraints connected with them.