Via Andrew Sullivan, Razib Khan over at Gene Expression (image from Wikimedia Commons):
Though I have always been skeptical of God, and an explicit and self-conscious atheist from childhood on, I found religious beliefs peculiar and difficult to comprehend in any intuitive sense. This led me early on to reading the source texts and scriptures, as well as theological commentaries (e.g., Summa Theologica, and I’ve read the whole of the Hebrew Bible and New Testament multiple times, and Genesis dozens). In this way I felt I understood on some deep level why people were religious.But I was wrong. When I read Scott Atran’s In Gods We Trust it opened up a whole landscape of cognitive anthropology which explained with much greater accuracy the paradoxes of religious belief and behavior with which I was confronted. The key insight of cognitive scientists is that for the vast majority of human beings religion is about psychological intuition and social identification, and not theology. A deductive theory of religion derived from axioms of creed fails in large part because there is no evidence that the vast majority of religious believers have internalized the sophisticated aspects of their theologies and scriptures in any deep and substantive sense. To give a concrete example, Sri Lankan Buddhists, Hindus, and Muslims can give explicit explanations to at least a rudimentary level as to the differences of their respective religious beliefs. But when prompted to explain their understanding of the supernatural in a manner which was unscripted, and which was not amenable to a fall back upon indoctrinated verbal formulas, their conceptions of god(s) were fundamentally the same! (see: Theological Incorrectness). The superficiality of theological system building is also evident in the fact that when confronted with radicalism derived from the logic of shared axioms during the Reformation prominent Protestant thinkers fell back upon tradition and revelation to defend the common creeds inherited from the early Roman church.
And that is why one should always been cautious of taking theology, textual analysis, and intellectualism too seriously when it comes to religion. Mathematicians can derive proofs from logical analysis. Those proofs are invariant across individuals and subcultures. They are true in a fundamental sense. Though natural science attempts to validate and refine theories and formal models which are robust, it fails when there is no empirical check upon the model building.Outside of pure math our powers of ratiocination are overwhelmed by subjective decisions along the chain of propositions. Separate theologians and have them derive from first principles, and there will be no similarity in their final inferences about the nature of God and the universe. Elite theological conformity is a function of social conformity, not the power of intellectual rigor. When isolation is imposed upon a community of religious believers for any given period of time they are almost always defined by a rapid shift toward heterodoxy, as they lose contact with the broader elite consensus (see: Dao of Muhammad as an example of how strongly an alien milieu can totally transform a familiar religious group unless that subculture remains in contact with the broader community).
Theology is not a cause of any great robustness on the macro scale. Nor does it explain much of micro scale behavior. Where does that leave us to be “serious” about religion? As Noah Millman stated it requires a deep program of empirical analysis and research of massive multi-disciplinary scope. Almost no one is interested in such a program from what I have seen.