Systems of Justified Belief

Here's more from Auburn University philosopher Roderick T. Long (Reason and Value: Aristotle versus Rand) on coherentism and reflective equilibration. The point of all this is to show that we can rationally construct reality-based systems of belief that do not require infallibility.

Aristotle thus appears to be counseling two steps. First, pursue a coherentist strategy for discovering first principles, by provisionally accepting all the beliefs that initially seem true or plausible to us (including, but not limited to, our perceptual beliefs) as defeasibly justified, and then try to see which candidates for first principles best cohere with our total set of beliefs in wide reflective equilibrium. Second, once the first principles have been thus identified, pursue a foundationalist strategy for demonstrating and thus explaining the truth of our initial beliefs (or some of them) by deriving them from these first principles.

The things well-known to us are genuinely known but not fully understood. By starting from them, however, we can reach a point where we understand first principles. At that point, we will be able to turn back and demonstrate the beliefs from which we started, turning them from mere knowledge into understanding.

Aristotle’s epistemology is primarily concerned not with knowledge but with explanation: our beliefs already count as knowledge, but in order to understand their objects we need to get them into these structures, and at the foundation of the structure will be beliefs whose objects need not be explained in terms of anything else. (We are not enchained by our present beliefs, however. As the Socratic Method indicates, the process of getting our beliefs into such structures, and eliminating internal inconsistencies, will undoubtedly involve some revision—sometimes minimal, sometimes drastic—in those beliefs themselves.)