One quick take on the discussion: I see no problem with the ‘circular’ defense of epistemic principles Lynch quickly dismisses. For any principle, if it is true and general, then it better be applicable to itself. The fact that someone else doesn’t recognize something as a reason, makes it no less of a reason. Suppose the following is true:
PC: If it seems to S that p, then S is justified in believing p.
If that’s true, then I better be able to be justified in believing PC by way of PC seeming true to me. That might not be a reason for PC that will be convincing to my opponent, but if PC is true, the it’s a justifying reason. Further, this is nothing special about PC, plug in any fundamental epistemic principle and the same will be true of it. So, like G+ I like circles. They aren’t all bad.
This isn’t to say that disagreement isn’t significant, I think it is. Further, none of this is sheds much light on the overarching democratic problem Lynch is addressing. Just a quick early morning 2 cents.