A lot of comments have revolved around whether I am a Post-Modernist when it comes to the definition of religion. This post is to make explicit and clarify my own position so I don't have to waste so much time in the comments. Most readers can therefore ignore this and wait until I go back to posting on genetics or something more interesting! :-) One model of religion goes like so: Axiom (e.g., One must follow all 612 commandments) → Entails → Entails → Specific belief and practice In other words, there is a contingent relationship between the initial set of beliefs, and the elaborated set of religious practices and beliefs which are subsequent to "primitive" and "core" assertions. In other words, the space of states which religious phenomena inhabit is tightly constrained and guided by a simple set of principles and ideas from which the rest follows. To a great extent many religious people might accept this as an accurate description of their faith; defined as it is by a belief in a particular god, a particular creed, and a systematic theology, etc. The average religious person might not be able to master the details of a particular theology, but they accept its validity, and accept the guidance of religious elites who are masters of said theology. These elites then serve as the executors or implementers of a set of logical consequences which emerge plainly from the propositions clearly derivable from first principles. In this conception religious texts and theologies serve as blueprints. Religion as it is simply serves to reflect the nature of that blueprint. Therefore, to understand a particular religion you simply go to the religious texts, and see what those texts say. That will be a reasonable approximate, model if you will, of religion as it is practiced. Texts and theologies serve therefore as the theoretical framework. To test the theory you need go out and observe how religion is practiced. So what happens if practice deviates from what the theory says? One reaction would be to suggest that practice is in error, that it deviates from the expectation of theory simply because of misunderstanding, or, willful neglect of the inferences. For example, most people would agree that most religions teach that adultery is wrong, but many believers continue to engage in adultery because of personal weaknesses despite their acceding to the moral principle that their actions are wrong. I accept the second point; many times people act in willful contradiction of their admitted religious principles because of personal failings. I do not accept the first; that is, that deviations from expectation are error. If I was religious I might accept this as a matter of faith because I accept particular premises about the nature of religion. Specifically, that religion maps non-trivially and transcendentally upon particular truths about the universe. If I was a theist I would also assert religion is a revelation from an entity of unimaginable power and scope. Because the premises of religion are what they are, there must be a true religion, a particular most religious religion which maps perfectly upon the Platonic idea of what a religion should be in the mind of the god who revealed the religion. But personally I don't accept this. I don't think that the initial axioms of religion about god and revelation are anything more than mental constructs; productions of human cognition, not expression of ontological truths.. Because religion is a production of the human mind I believe there is a profound subjectivity to its expression and perception. Additionally, I do not believe there is a Platonic ideal religion which maps onto true religion. All religion is true only insofar as religion is ultimately rooted in neurological material and phenomenological process; gods exist only in the mind's eye. And therefore, I do not totally shrug off the accusation that I am a Post-Modernist when it comes to religion. Since I believe that religion is fundamentally a mental construct, I do not believe that individually it is my place to tell a religious person what their religion is all about. It is what their mind tells them it is. It is what it is. Of course, there is a problem insofar as while I think religion is simply a production of their minds, they believe it is a reflection of some deep truth outside of their minds. We, the religious person and I, disagree on the fundamental nature of religion. They may reject Post-Modernism precisely because they believe that religion is true. I believe that it is false insofar as I am considering the set of assertions which they believe are true, but I believe that religion is true as a mental process. This brings to the disagreement I have with some atheists, a disagreement I would have with myself when I was 18. An atheist believes that the claims of religion are false, but an atheist may believe that there is a true expression of religion which can be back-projected toward its premises. An atheist may reject the premises, but they may hold a model of religion which conceives of it as a set of necessary inferences, a chain of tight propositions back to the original premises. I do not believe that this model reflects reality; that is, it does not reflect the truth of how religion manifests itself in the world around us. I do not believe that religion as it is practiced is tightly constrained by a primitive initial set of beliefs. Instead of an analogy to a logical or mathematical formalism as the theoretical superstructure of religion, I believe that something more akin to law is appropriate. In other words, religion is a matter of interpreting from the premises toward a range of conclusions. The sample space which religion as it is practiced inhabits is very large, and relatively loosely, if at all, constrained by the premises of religion. Rather, the sample space is contingent upon local historical context and its own endogenous evolutionary pathway. Of course, many religious persons will tell you this is not so. But my discussion at this point is not with the religious, but those who reject its truth claims as I do. My contention is that religion is not well characterized as a set of necessary propositions, so deviation from "expectation" is not error, rather, it reflects an interpretative difference along the set of propositions which is a matter of local condition and contingency. You might ask how it is then that religious professionals might agree that "of course A → X", where there are intervening inferences. I believe this is for show and comes about through social consensus. My own study of Chinese Islam suggests that when separated from other religionists a subgroup can quickly deviate outside of the bounds of the consensus, and only reintegration into the world wide information network can correct the "errors" which creep into the inferences. Rather, the true nature of religious logic is better illustrated through its evolution over time, which implies a malleability and loose constraint from premises. After all, the Nicene Creed and the basic corpus of the Bible have been axioms which span nearly 2,000 years, but the normative form of Christianity varied a great deal due to time-sensitive interpretation. Some, but not all, religious people will assert that there isn't any time-sensitive interpretation; that past interpretations were wrong or conditioned by local circumstances, but present interpretations reflect the true spirit of the doctrine. Again, if you accept the presuppositions of a religion as to its transcendent truth value and revelation from god on high this is a reasonable assertion. But if you do not accept the truth value of the religious premises then one must question it, and ask if we are not again seeing local temporal conditions being the important determinants of religious practice. Because of the world wide nature of Christianity or Islam we can see how this dynamic plays out spatially. The African churches of the Anglican communion hold to the dominant view in regards to homosexuality over time of the Christian tradition. The American and Europe branches hold different views. Both groups claim that their perspective in the authentic and true interpretation of the religion, but I think what you're seeing is simply different local conditions. After all the African branches of the Anglican communion don't adhere to all the precepts of Anglicanism as it was in 1600, or Christianity as it was in 300. In fact some "Southern" Christian theologians have argued strongly for an indigenization of Christian practice to accommodate local practices, in part by asserting that Christianity as it was practiced and evolved over the past 2,000 years was in fact Europeanized (e.g., the rejection of polygyny is attributed to Greco-Roman pagan influence, as evidenced by the acceptance of the practice among Jews outside of Europe). At this point I would like to sidestep for a bit into my model of cognition. In short I believe that many cognitive processes are reflexive, or somehow encapsulated from our conscious inspection and awareness. Rationality is like a shimmering surface above the deep roiling waters of our mental processes. The human mind is a collective, and one where there is imperfect communication or unanimity. Mathematics works because its formal system is so precise and clear that there is no possibility of "cognitive creep" fudging the sequence of inferences to suit our own ends. In contrast, verbal logic is subject to interpretation, and so inevitably subjective or exogenous parameters end up shaping its outcome. Wealthy Christians may genuinely believe that their wealth is a gift from god, and that Christ wishes them to be wealthy. From the outside one might note that wealthier Christians seem to come to a particular interpretation in regards to material success, while less wealthy ones come to another, but both might be equally sincere in accepting that their logic was objective. The problem here is that the nature of cognition means that without the straight-jacket of symbolic formalism people easily and unconsciously insert hidden variables into the reasoning process. More concretely, this gets us to something like the Bible. I've been talking as if the premises are clear and distinct even if the propositions entailed are not so much. In fact reading the Bible itself is subject to a great deal of interpretation. "Literal" readings of the Bible are not usually quite so literal, rather, they often "hide" the interpretation by packing it straight into the text without acknowledgment. By this, I mean that Fundamentalists may appeal to a Bible which translates a word or passage in a manner to their liking. Non-Fundamentalists may admit beforehand that there are different readings, or in the process of smoking out the inferences point to the different directions where the text could take you. Fundamentalists may assert that there is no falsity in the Bible, but they eliminate falsification and contradiction simply through expedient reinterpretations of words. Jesus Christ prophesied that he would return before the passing of a generation, but since generation means Jewish people, as long as the Jewish people remain he need not necessarily return (why does generation mean Jewish people when it says generation? Don't ask). Nevertheless, at least there is sense in the Bible. The nature of the Bible is such that it is accessible to a typical person; the stories and ideas extant within are intelligible. What about theology and religious philosophy? To a great extent I don't believe they have sense; that is, I don't think that they mean anything in a direct fashion. I don't think even the theologians themselves understand what they're saying or what it means. That implies to me that the problems with viewing religion as a logical system start out with the axioms. After all this, I think it's pretty clear I don't think as a phenomenon that religion is what religious people think it is. So what is it? I do believe one can make objective generalizations about religion, but I believe to a great extent it is an empirical matter, not one of inferences derived from textual and theological presuppositions. Religion is how it is practiced. Religious people may believe that religion is true, so likely how they are practicing is the closest to true religion in their own mind. But from a non-religious perspective I think it is useful to simply define and characterize it by the distribution of practices and beliefs that people hold, and not by texts or experts. Therefore, one can make generalizations about religions for a particular time and place, but since there are few constraints one can not make universal generalizations. This gets to my point about instrumental utility. A model of religious behavior, a predictive model so to speak, can be constructed, but its priors must be the proximate behaviors and beliefs. An inductive system is within our reach, but I believe any deductive system predicated on religious priors (texts, theologies, etc.) are highly suspect. I do believe that a deductive system which suggests constraints is possible, but I do not believe that it is possible from the world of religious studies, rather, one must look to the social and biological sciences. Since religion is a cognitive phenomenon we must examine the priors which constrain and shape the unfolding of the cognitive process. In my post Richard Dawkins - Islamophobe? I implied that Islam is Creationistic in orientation. I believe this is true, insofar as I believe most Muslims would be what we call Young Earth Creationists. But, this is an empirical matter. There are Muslims who are not Creationists in this fashion. Are they then less "true to Islam"? I don't believe so. Islam is what Muslims believe, if they believe that that is true to Islam that is their opinion and I won't gainsay that. That being said, there is a statistical generalization one can make. On a theoretical level does the nature of Muslim interpretation of the Koran constrain or bias Islam toward Creationism? Possibly. That being said, most Muslims do not read the Koran, most Muslims can not speak Arabic, especially the classical variant within the Koran, and a substantial minority of Muslims are even illiterate. I do not believe that Muslims are by necessary Creationist, rather, that is simply the modal state of Islam here and now. That may change due to interpretation. In other words, an objective model of Muslims can be constructed based on ascertainment of the empirical distribution. This distribution though is in constant flux, and that flux is contingent up a host of parameters, very few of which are ultimately rooted in some sort of religious premise. For an atheist to make an assertion about what the true Islam is is like a geologist to define the most rocky rock. A rock is a rock. Though abbreviated I'll end my own model of explaining religion at this point. But rather I want to shift to some of the atheists who criticize this model. I believe their own rationale for trying to truncate religion into a simply formal system is pretty obvious; you can disprove formalisms. On the other hand, the sprawling complex phenomenon that I describe above is a bigger fish to fry. Like a natural system it requires a great deal of study to re-engineer and model. It takes work, and we're not really there yet because the social sciences have not advanced to the point where we have all the tools necessary to understand the phenomenon we speak of, and which affects our lives on a very deep level. We can't just argue religious people out of religion if the model I'm proposing is correct; we can't just show that it's unreasonable and false because reasoning and falsity isn't really the point of it. My main criticism of The God Delusion is that Richard Dawkins seems to "get" that religion is more than a simple set of beliefs derivable from axioms in the first half of the book...but in the second half he pretends as if it is exactly that to "refute" it. If it was a matter of conjecture and refutation it would be rather tractable, but it isn't. The model of religion that many atheists hold in their mind is simply one thing: wrong. That's just objectively so. But the godless delusion that religion is what an atheist thinks religion is is hard to banish.