Both in the classical tradition and in recent analytic philosophy, much of philosophical theology is concerned with what we might call metaphysical theism, that is, with the notion of God as a metaphysical theory which explains certain facts about the world. This is most visible in the cosmological argument for contingency, where the ability of the theistic hypothesis to explain something that (allegedly) cannot be explained (or explained equally well) without God is given as a reason for belief in God. A lot of our theorizing about God (in this metaphysical mode) then has to do with the question of what is the simplest and most intellectually satisfying variant of the theistic hypothesis for our metaphysical purposes.
If this kind of theorizing succeeds – that is, if the theistic hypothesis really produces a better theory than its competitors – then this is of course extremely important to metaphysicians, but its religious significance is in dispute. Certainly intellectual assent to a metaphysical theory is not what the Christian means by ‘faith’. (Well, maybe, depending on who ‘the Christian’ is, that’s not quite certain. What’s certain is that it’s not what Christians should mean by ‘faith’.) Furthermore, various difficulties and contradictions have been alleged between prominent versions of metaphysical theism and the needs of religion. Thus many recent philosophers and theologians have alleged that some of the classical divine attributes (e.g., atemporality, impassibility, foreknowledge) interfere with God’s ability to enter into the kind of personal relationship essential to Christianity. Others (e.g., Geach) have held that omnipotence implies that God is able to break promises and that this belief tends to undermine religious faith. So metaphysical theism is not the kind of theistic belief relevant to religion, and may even be in some tension with religion.
Religious theism, that is, the sort of doxastic state regarding God that is appropriate for Abrahamic religion, must then be a different sort of thing. Indeed, Plantinga has famously argued (1986, 132-133; 1996, 249) that religious theism is generally not anything like an explanatory theory, and we should not expect it to be justified in anything like the same way. This is precisely correct. Whatever exactly religious theism turns out to be, natural inclination to believe (Plantinga’s sensus divinitatis), religious experience, experience of miracles, and testimony about such things will all be relevant to its formation, but explicit theorizing will not.
Does this make metaphysical theism irrelevant to religious theism? I think not. Plantinga and Alston both argue that theistic belief (and they seem to mean religious theism here) is basic in something like the way beliefs arising directly from sense experience are basic. (Alston, of course, emphasizes the sensory analogue a lot more than Plantinga.) But consider how we actually respond to the deliverances of our senses. We do (and should) distrust our senses if what we seem to sense cannot be made to fit into a coherent picture of the world. Further, although we do not have non-circular justification for our general attitude of trust in the senses, we do have scientific theories about the functioning of our senses and these inform the degree of trust we place in our senses in specific circumstances. Trust in the senses notwithstanding, if I seem to see a pink elephant in my living room, I will conclude that I am hallucinating because I know of no intellectually satisfying background theory on which the presence of a pink elephant in my living room is more likely than my being the victim of a hallucination.
The background theories whereby we determine when and how far to trust our senses become all the more important in cases of contradictory testimony. It is by consideration of the circumstances in which the senses tend to be reliable, and the likelihood of the events in question, that we decide which witnesses to trust, and sometimes even end up giving preference to the testimony of others over our own senses. (For instance, if someone else was in a better position to observe the action.)
Rarely, if ever, are witnesses to events that can be detected in ordinary sense perception as conflicted in their testimony as human beings are about religion. As a result, we stand in dire need of a theory that will tell us both (a) how generally reliable are the processes by which religious theism comes about, and (b) how likely to be true are the particular claims which purport to be justified by this process. Metaphysical theism has an important role to play as such a background theory. An intellectually satisfying theory of the world which allows for the truth of religious theism will render religious theism better justified, just as an intellectually satisfying background theory according to which vision is more reliable in my circumstances than in yours increases my justification for trusting my visual experience over your testimony. If the atheist’s background theory is far more compelling than mine, and can explain the origin of my religious theism as a cognitive malfunction, I may eventually have to concede that I am the victim of such a malfunction, just as there are circumstances in which I may be brought to concede that I am hallucinating.
The undermining of the justification here depends on two factors: conflicting religious testimony, and the relative merits of different explanatory theories. It thus will not tend to undermine the (internalistic) justification of relatively uninformed or unsophisticated religious theists who are either unaware of the conflicting testimony or are not in a position to evaluate the relative merits of the theories in question. However, if the religious theist wants to retain her justification after becoming a sophisticated philosopher, she is going to need metaphysical theism or something very much like it.
(Cross-posted at blog.kennypearce.net.)