Epistemology as a Two-Party System?

Daniel’s piece brings up a broad range of points, and I doubt I’ll be able to sustain a response that hits them in any exhaustive fashion. Part of the difficulty in doing so (apart from the length) is that I’m having trouble following some moves that Daniel makes. Perhaps it would be best to start with what I think are some problems with how he’s structured his ideas, so that people can at least follow where we differ in our basic orientation to all of this.

Framing the issue of self-referential authority, knowledge, and justified assent in religion in terms of “axioms” is probably a helpful way to go, but I think there can be some vagueness about what this means for religious claims that are axiomatic. Daniel states that “axiomatic claims are supposed to be simple, basic, tautological or definitional if possible.” This seems to be the case for “all bachelors are unmarried” or for the concept of something like a triangle. Perhaps it’s also the case for concepts of God that involve various perfections. These are definitional because they are not contingent. But what about the claim that “God exists”? Often this existence is posited as necessary existence and articulated in terms of a divine simplicity (equating essence and existence) in a way that makes this claim concerning God sound axiomatic in the way that Daniel describes axioms. But if it is this sort of axiom, it is a quite spectacular version of it insofar as the existence of this divine being is included within the axiom… the definitions of a triangle and a bachelor aren’t “useful knowledge” because they stand regardless of the actual existence of any bachelor or triangle in the world. The whole point of various proofs of the existence of God seems to be, however, to have one’s cake and eat it too… to establish an axiomatic theological claim that also has a foothold in contingent facts. The attempted proofs may be deemed a failure (although many people remain convinced by them), but they certainly strike me as something quite different than the other axioms that Daniel provides as comparisons, precisely because the proofs for God’s existence consciously attempt to present in one instance a claim that is both axiomatic and relevant knowledge for contingent facts.  Daniel thinks this is the basis of the problem, but I wonder why this is the case.  It doesn't seem to lack any properties of what Daniel is describing as axiomatic, it simply makes things more difficult by incorporating existence into that which is definitional and so usually taken to be independent of contingent facts about things.  One person's sui generis is another person's fast-and-loose-with-the-rules.  I'm not sure it will be easy to reach an agreement about what is the case here.

Here I’ve just moved pretty sloppily through metaphysical questions that have been debated with much more acuity for some time. As we move into scriptures and religious bodies, however, it seems that we move into territory that is more obviously different from the definitional tautologies with which Daniel attempts to identify them (although I could still see them being called “axiomatic” as long as the meaning of this is clarified appropriately). Surely, as Daniel says, “the authority of scripture is assumed”… but is it assumed in the same way that the definition of a bachelor is assumed? It is the concept “bachelor” that is taken by definition to be an “unmarried man”. When we pass a single guy on the street, however, we observe his bare ring finger and therefore find the concept “bachelor” to be a useful description of him. Likewise, “scripture” might (this is a crude rendering, but hopefully useful for present purposes) be defined as “authoritative religious text”, but it is only the concept of “scripture” that is axiomatic in this way. Any given text is then usefully described as scripture insofar as a person or a community takes such a description to be useful in describing it. “All bachelors” doesn’t stand in the realm of contingent facts, but “this bachelor” does. Likewise, “Scripture” should be distinguished from “the Old and New Testaments” or “the Koran” or “the Book of Mormon”, each of which is more or less convincing to a given person as authoritative scripture. Similar arguments can easily enough be laid out for institutional bodies that claim authoritative status.

This is why Daniel turns to consider empirical standards of knowledge rather than talk of axioms.  Such standards seem to fit better in cases where rationalist approaches aren't as helpful.  Of course there are ambiguities to religious truths insofar as they are not always observable or their cause (Allah, the Holy Spirit, etc.) is not always clear (this second problem probably often boiling down to the first problem of unobservability, although Hume had problems with an empiricist basis for knowledge of even the most mundane causes). But I don’t think this means that certain religious claims aren’t contingent facts in the way that "the sky is blue" is (and Daniel doesn't disagree with me here). It just means that they’re not empirically straightforward contingent facts. But does that make the following true? –

“Belief in revealed religion is belief in something that you authorize yourself to consider “true,” not belief in something that any established epistemological standard (either rational or empirical) allows you to claim is true.”
 
When did “rational or empirical standards” become “any established epistemological standard”? And to what extent must we even reject rational or empirical standards simply because religious truths don't comfortably conform to rigidly analytic or sense-observable notions of rational and empirical standards?  I guess I’m just at a loss to understand why this is worth fretting over.  Is the point simply that one can't structure the justification of religious knowledge claims like one might structure a scientific experiment or a geometry textbook?