Originally posted by FreakyKBH
Even observing requires faith.
Faith in one's ability to perceive accurately.
Faith in one's analysis of that perception.
Faith runs the gamut, really.
I think you're running together two different things: 1) S observing P and, 2) S being justified in believing that he is observing P.
There is no plausible construal of 'faith' such that it's required for (1). If P is the case and S is (appropriately) causally connected to the fact that P, then S observes P. The heavy lifting here is spelling out what constitutes an appropriate causal connection between S and the fact that P.
There is certainly a case to be made, however, that for S to be justified in believing he observes P, S must
also have the justified beliefs that his perceptual systems are reliable, that the world is at least roughly as it seems to him in perception, that he is not conceptually confused, etc. But note that these are
only requirements if Internalism about epistemic justification is correct. If Externalism is correct, then S doesn't need these further, supporting reasons. Rather, S simply needs to be a reliable observer in fact.
But suppose we grant your tacit assumption that Internalism about epistemic justification is correct. Even then, you don't get your conclusion that faith is required; all that's required is justified belief...
I think you might reply that we simply can't have justified beliefs of the sort mentioned above. After all, there is always a chance that our perceptual systems have gone awry, that we're in anomalous circumstances, confused or whatever. Because these supporting beliefs are always epistemically uncertain, we can't be justified in holding them. If, at bottom, that's your view, then you'll have to be prepared to admit that almost none of our beliefs are epistemically justified. It's logically possible you're wrong about your name, address, etc., so since there's a chance you're wrong, your beliefs about your name and address can't be justified. But that's absurd. You know your name and address. Knowing that P, like justifiably believing P, doesn't require epistemic certainty.