"Is it easier to know what our culture’s epistemic norms are in terms of which it is to be determined whether or not giraffes are taller than ants than it is to know that giraffes are taller than ants? Is our knowledge concerning the identity of those norms to be construed on a realist model, or must we rather consult our culture’s epistemic norms in order to use them to figure out the identity of those norms? If the former, why is it that we can have objective knowledge concerning the identity of our culture’s epistemic norms when we can’t have it with regard to the relative height of giraffes and ants? And if the latter, how is an infinite regress to be avoided?"

“Surely at least some of our logical principles … are far more evident and obvious to us than are any of the empirical generalizations we find in psychology, or in any other of the empirical sciences. Thus, to attempt to explain logic in terms of psychology is to explain the more certain by the less certain. It is, in short, to commit the fallacy of obscurum per obscurius.”

“There is an obscurum per obscurius problem as well. In order to know whether or not giraffes are taller than ants we must first know whether or not there is a consensus that giraffes are taller than ants, and if there is, whether or not the communication that produces that consensus was free, open, and undistorted. But isn’t it obvious that it is easier to determine whether or not giraffes are taller than ants than it is to determine either or ? Or, to put it another way, wouldn’t any skeptical doubts about our ability to determine even something so obvious as that giraffes are taller than ants also be more than sufficient to wipe out any hope of being able to know about the outcome, and degree of openness, of any process of public communication?”

“In order to judge that something is not working, wouldn’t we have to know what it’s effects really are? … And surely some of our social practices cannot be judged to be “working” except by invoking the goal of objective truth. For example, to know whether or not the criminal justice system is working well, wouldn’t we have to know something about the accuracy of its results?”

“If our aim is the objective truth, we can have a motivation and a reason for revising our epistemic norms whenever we have reason to think they aren’t helping us optimally to reach that goal, and we can have a sounds basis for criticizing others’ claims that we should adhere to the norms. But how can we have any of these things if the truth just is whatever follows from our epistemic norms?”