Reflection on Certainty: #5

Are you sure?

Knowledge is regressive.  Every conclusion we reach relies on some prior shared premise.

Is there any piece of knowledge we can point to that is not grounded on prior knowledge?  Is anything truly self-evident?  I'm honestly not sure.  The axioms we often call self-evident, such as the law of non-contradiction, do not in fact seem to be self-demonstrating. They are merely so far above (that is, prior to) other statements that standard deductive proof becomes a practical impossibility. Self-evident axioms are so called not because they are demonstrated, but because they are so foundational to our rationality that rejecting them would be rejecting the very idea of rationality.

In short, the basis for what we call certainty is not our positive knowledge of something, but rather our inability to conceive of its opposite.

As humans, we are rational creatures.  Our minds work in certain ways, and don't work in others.  It is physically impossible to operate outside of those constraints.

This is what I call methodological certainty. We are allowed to feel and express certainty, by this standard, for those ideas without which we would not be able to function.  Methodological certainty concerns the sine qua non of our rational humanity: statements without which we are not. 

For instance, we might be able to deny the law of non-contradiction by our words, but it would not have the force of rational belief.  We could not defy it in action, much less in actual processes of reasoning.  Our minds don't function in that way.  We can conceive of these limits on our mind, but we can no sooner transcend them than we can transcend the intrinsic limitations of our physical bodies.

One danger is that humans are eminently capable of self-deception. We might deny a given statement, and we may even live and reason based on our denial of that, without ever encountering the logical consequences of that idea that would throw other aspects of our worldview in jeopardy.

What principles can we categorize as methodologically certain?